That explanation is transaction-cost economics, the branch of theory concerned with why some exchanges are organised through the market and others inside the firm. Where the resource-based view says which resources matter, transaction-cost economics says why a platform brings them inside through ownership rather than renting them. This note develops that complement, applies it to the build-borrow-buy choice, and tests it against the obvious objection that a long-term contract would be cheaper.

Why firms exist, and why they buy

The question is older than buy-and-build. Coase (1937) asked why firms exist at all, given that markets are supposed to coordinate production efficiently, and answered that using the market is itself costly: there are costs to discovering prices, negotiating, and enforcing contracts, and when those costs are high enough it is cheaper to organise the activity inside a firm under managerial direction. The boundary of the firm sits where the cost of one more internal transaction equals the cost of carrying it out through the market.

Williamson (1979) made that boundary operational. Three features push an exchange out of the market and inside the firm: asset specificity, when the parties must invest in assets specialised to the relationship; uncertainty, when contingencies cannot all be written into a contract; and the hazard of opportunism, when one party can exploit the other once both are committed. The greater these are, the more an arm’s-length contract leaves a firm exposed to hold-up, and the more attractive ownership becomes, because common ownership aligns incentives and lets disputes be settled by fiat rather than by renegotiation.

Build, borrow, or buy

For a growing platform this abstract make-or-buy choice takes a concrete form: build the capability internally, borrow it through a contract or alliance, or buy it through acquisition (Capron & Mitchell, 2012). Their argument is that firms default too readily to one mode, usually the one they know best, and that disciplined growth means matching the mode to the resource. Build when internal resources are close to what is needed; borrow when a capable partner exists and the relationship can be governed by contract; buy when the resource is deeply embedded in another organisation and cannot be cleanly separated from it.

In a platform the choice is recursive, which is what makes it interesting. Capabilities assembled through early add-ons change the calculus for later ones. A capability the platform once had to buy, because it had no foundation to build on, it may later be able to build, because earlier deals supplied the foundation. Conversely, a relationship the platform was happy to borrow through a contract can become specific enough, as volumes grow, that leaving it in the market becomes the riskier option.

Why platforms internalise

Transaction-cost reasoning explains several recurring buy-and-build patterns that a pure synergy story leaves vague. Vertical integration is the clearest. A platform that depends on a distributor or a specialist installer, and that has invested in assets specific to that relationship, faces hold-up risk: the partner can extract value precisely because the platform cannot easily switch. Acquiring the partner removes the hazard by bringing the transaction inside the firm. The same logic explains why platforms internalise scarce capabilities, such as a regional service network or a proprietary product line, rather than contracting for them, once those capabilities become specific to the platform’s strategy and too important to leave exposed to a counterparty’s incentives.

This is also where transaction-cost economics and the resource-based view fit together rather than compete. The resource-based view explains which capabilities are worth controlling, because they are valuable and hard to imitate (Barney, 1991; Dierickx & Cool, 1989); transaction-cost economics explains the governance form, ownership rather than contract, through which the platform secures them. Recent theory casts private equity itself as a specialised intermediary in the market for corporate assets, whose comparative advantage lies in reallocating and governing assets that markets price poorly (Nary & Kaul, 2023), which is a transaction-cost argument about the sponsor as much as about the platform.

A worked illustration: from supplier to subsidiary

Consider a building-products platform that for years has bought a specialised component from an independent regional manufacturer under an ordinary supply contract. Early on this is a textbook borrow: the component is available from several sources, the relationship is non-specific, and a contract governs it cheaply.

As the platform grows, the relationship changes character. The platform redesigns its installed product around this manufacturer’s specification, trains its branch staff on it, and markets it to customers, investments that are specific to this supplier and worth little if the relationship ends. The manufacturer, aware of this, presses for better terms at each renewal. The platform now faces classic hold-up: it is committed, the asset is specific, and the contract cannot anticipate every future contingency. Transaction-cost logic predicts what happens next. The platform acquires the manufacturer, not because the manufacturer is a wonderful standalone business but because internalising the transaction removes a hazard that had become too costly to manage through the market. The borrow became a buy when specificity and uncertainty crossed a threshold.

The objection: wouldn’t a contract be cheaper?

The natural objection is that acquisition is an expensive and clumsy way to solve a contracting problem. Long-term agreements, exclusivity clauses, and well-designed incentives can manage most supplier relationships without the cost and integration burden of ownership, and often they should. Transaction-cost economics agrees: where specificity and uncertainty are low, the market is the right governance form, and a platform that acquires everything it transacts with will overpay and overload its integration capacity.

The objection fails only at the margin transaction-cost economics actually identifies. When assets are highly specific, contingencies cannot be fully specified, and opportunism is a live risk, contracts become incomplete in ways no clause fully closes, and the cost of repeated renegotiation and the exposure to hold-up exceed the cost of ownership. There is also a resource-based limit the contract cannot reach: some of what the platform wants, the tacit know-how and routines embedded in the target, cannot be transferred by contract at all, only by acquiring the organisation that holds them. The discipline cuts both ways, and the failure case proves it. When bidders lack the keystone resources needed to unlock a target’s value, the right move is not to own it; firms that announce deals and then find they cannot create the value divest the target-related resources rather than absorb them (Gibbs et al., 2026). Transaction-cost economics is not an argument for buying; it is an argument for buying only when the market is the costlier option.

Four propositions

Stated plainly, so they can be argued with and tested against cases:

1. Governance, not just resources. The resource-based view says which capabilities to control; transaction-cost economics says when to control them through ownership rather than contract or alliance.

2. Specificity drives internalisation. Platforms acquire, rather than contract, when asset specificity, uncertainty, and the hazard of opportunism make the market exchange too costly to govern.

3. The mode choice is recursive. Capabilities accumulated through early add-ons shift later build-borrow-buy decisions, converting some buys into builds and some borrows into buys.

4. Buy is not the default. Where specificity and uncertainty are low the market is the right form; acquiring everything overloads capacity and destroys value, and the failure case is divestiture.

Why this matters

Read alongside the resource-based view, transaction-cost economics turns buy-and-build from a series of opportunistic purchases into a governed sequence of make-or-buy decisions. It explains why platforms internalise distribution, why a comfortable supplier relationship suddenly becomes an acquisition target, and why the discipline is knowing when not to buy. The companion question, once the platform has decided to acquire, of which target to choose, is the subject of the target-selection note; and the constraint that every internalisation quietly draws down is the absorptive capacity examined later in the Notebook. Ownership solves a governance problem, but it spends the one resource a platform cannot easily replace.

References

Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99–120.

Capron, L., & Mitchell, W. (2012). Build, borrow, or buy: Solving the growth dilemma. Harvard Business Review Press.

Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405.

Dierickx, I., & Cool, K. (1989). Asset stock accumulation and sustainability of competitive advantage. Management Science, 35(12), 1504–1511.

Gibbs, A., Byun, H., & Lim, K. (2026). Build, borrow, buy… or bail: Divestiture following merger and acquisition deal termination. Strategy Science. Advance online publication.

Nary, P., & Kaul, A. (2023). Private equity as an intermediary in the market for corporate assets. Academy of Management Review, 48(4), 719–748.

Williamson, O. E. (1979).Transaction-cost economics: The governance of contractual relations. The Journal of Law and Economics, 22(2), 233–261.

The argument from The First 24 Months Now Decide the Deal — that the operating-model design choices made at entry now produce most of the return certainty — sharpens with the new figure attached. The first 24 months are where the operating composition required to deliver 12% gets built, not where the strategy to deliver 12% gets articulated.

The new math and what it actually demands

The 12% requirement is not a sector-specific number. Bain reports it as the asset-class-wide consequence of the new cost-of-capital environment. For mid-tier PE in building products, however, the requirement lands against a sector where organic-revenue growth has been constrained for years by the same “still-fragile construction demand and an uncertain outlook” the Bain Global M&A Report 2026 described — the demand backdrop in which The Scale Curve Has Run Out — Where Scope Goes Next and The Frequent-Acquirer Premium and What It Demands of the Operating Model already framed the asset class’s strategic position. Entry multiples in the sector have followed the broader market upward without a corresponding lift in underlying demand growth. The gap between what the math requires and what the cycle delivers is therefore wider in building products than in sectors where demand growth has been stronger.

This means the 12% requirement, for a mid-tier PE platform in this sector, is binding in a way it might not be in another sector. There is no demand tailwind to absorb part of the requirement; every basis point of the 12% has to be designed for and delivered through deliberate operating action.

Why no single lever in mid-tier PE building products delivers 12%

The four conventional EBITDA-growth levers all carry sector-specific constraints in building products that limit any single lever’s contribution. Organic-revenue growth is bounded by macro-cyclical construction demand, which compresses the upside available without taking share — and taking share in segments like cement, where the scale curve has run out, is harder than it was a decade ago. Pricing runway is shallow in segments where the largest acquirers have already taken price discipline through prior consolidation. Margin expansion from cost-takeout was the prior decade’s primary lever, but in segments where consolidation is exhausted — cement at 65–70% top-six in the US, the example The Scale Curve Has Run Out — Where Scope Goes Next already cites — the easy cost levers have already been pulled. Capital efficiency improvements through working-capital and asset-turnover discipline can contribute, but the contribution from any single working-capital cycle is bounded.

The mid-tier platforms I watched fail to deliver projected EBITDA growth usually shared a common pattern — the underwriting case had concentrated the growth thesis on one lever, and the operating model had been built to manage that one lever well, leaving the other levers to perform without dedicated cadence or accountability. When the primary lever underdelivered, the platform had no compensating contribution from the others because the operating model had not been designed to produce one.

What composing 12% actually requires operationally inside mid-tier PE

The operating discipline that credibly delivers 12% EBITDA growth in this sector is not a single-lever discipline; it is a multi-lever composition. It requires organic growth from channel and customer expansion running in coordination with inorganic compounding from add-ons that contribute real EBITDA rather than revenue stacking; it requires pricing discipline differentiated by SKU and segment; it requires margin expansion from operating-model improvements that go beyond cost-takeout; and it requires capital-efficiency discipline on asset turnover and working capital. Industrial Patterns Operating Benchmarks in the Building Materials Industry, 2024 edition, measures the gradient across which these levers operate: distribution turns assets 1.95 times for every dollar of capital against manufacturing’s 0.46 to 1.45 times, with refined overhead intensity at 8.8–17.3% versus manufacturing’s 15.1–25.3%. The compositionally feasible 12% looks materially different in a distribution-led platform than in a manufacturing-led one, and the operating model has to be designed for the specific composition the platform’s segment mix actually supports.

Coordinating the composition requires decision-rights architecture that holds multiple lever P&Ls in parallel without defaulting to the dominant lever, and operating cadence structured around lever interactions rather than around any single lever’s progress against target. Integration capacity for inorganic compounding has to be maintained as a continuous operating function rather than an episodic deal-team activity. Each of these elements supports a different lever, and each is undermined when the operating model is built around the assumption that one lever will carry the platform.

In the building products platforms I’ve worked inside and watched closely, the operating model that could credibly produce double-digit EBITDA growth was always more multi-lever than the underwriting case described — pricing, mix, productivity, and inorganic compounding had to be running together, and the operating cadence had to be designed for the composition rather than for any single lever.

Where the academic literature underwrites the operating-side argument

Acharya, Gottschalg, Hahn and Kehoe’s 2013 Review of Financial Studies study of 395 PE transactions establishes the empirical foundation for the operating-side argument: in their sample, “abnormal performance of deals is positive on average after controlling for leverage and sector returns, with higher abnormal performance related to improvement in sales and operating margin during the private phase, relative to quoted peers.” Operational improvements — specifically sales growth and operating margin expansion — explain PE outperformance over their period. The 12% math does not change the source of alpha that the academic literature has identified; it changes the scale of operational improvement required to produce it, which is what changes the operating discipline mid-tier PE has to design platforms around from entry.

The connection to the through-the-cycle operating discipline argument is direct. Through-the-cycle acquisition is one element of the composition required to deliver 12%; the operating capacity to maintain it through down cycles is one component of the broader composition that has to be designed at entry. The pieces fit together because they are different views of the same underlying operating reality: the asset class’s returns now come from operational improvement at a scale most operating models were not built to produce, and the discipline to produce that improvement has to be built into the platform from the start.

What 2026 settles

2026 is the year the operating math demands of mid-tier PE building products platforms shifts decisively from single-lever to composed value creation — and the platforms positioned to deliver the 12% requirement will be the ones whose operating models were designed for composition from entry, not retrofitted onto a scale-era playbook after the math reset.

In buy-and-build environments, the moment is familiar. It’s where leadership stops being theoretical and becomes operational, and most of what determines whether organisations stabilise or fragment at that point has less to do with structure than with how leaders behave inside incomplete systems.

This is where the work of Warren Bennis remains useful, not as leadership philosophy, but as a description of leadership under constraint. Bennis (1989) wrote extensively about leadership emergence in conditions of uncertainty, before formal authority, elegant structures, or aligned teams have settled.

Leadership when authority is incomplete

Bennis’s work didn’t begin with assumptions of perfect conditions. It began with ambiguity: leaders operating before systems had caught up, before roles were fully defined, before trust was settled. That context matters.

In buy-and-build systems, leadership authority is often temporarily incomplete. Titles may be clear, but legitimacy is still forming. Decisions get made before norms are fully reset, and people infer authority from behaviour rather than from org charts.

Bennis understood that in those conditions, credibility comes from consistency, not position. Consistency in how decisions get made, in how trade-offs get explained, and between what gets said and what gets protected. Those behaviours reduce uncertainty long before structure does.

Credibility as a capacity multiplier

Earlier in this section, leadership was framed as a finite resource. Bennis helps explain how that finite resource travels further, or collapses faster, depending on behaviour. Credibility functions as a capacity multiplier.

When leaders are credible, fewer decisions escalate unnecessarily, informal coordination recovers faster, people act without waiting for confirmation, and leadership attention is conserved for genuinely hard problems. When credibility is weak, the opposite holds: even small issues demand senior involvement, decision latency increases, and cadence fragments.

Bennis described this as a failure of trust under pressure rather than a failure of intelligence or intent. The distinction is the central operational point of his work.

Judgment without the illusion of control

A recurring theme across Bennis’s writing is restraint — not passivity, but the discipline to act without pretending to know more than the system can support.

In acquisition environments, leaders are often tempted to project certainty early. Structure and decisiveness feel stabilising. Premature certainty creates fragility, especially when the organisation is still interpreting what has changed. Bennis emphasised judgment over decisiveness.

Judgment shows up as naming what isn’t yet known, resisting false precision, making provisional decisions visible as provisional, and allowing learning to precede standardisation. The form of leadership doesn’t accelerate execution immediately; it preserves the organisation’s ability to execute later.

Meaning as an organising device

One of Bennis’s most cited ideas is that leaders create meaning. In operational settings, that often gets misread as inspiration or narrative. In practice, meaning is far more functional, it reduces cognitive load.

When people understand why certain priorities are protected, why some initiatives wait, and why authority is exercised unevenly during transition, they make better local decisions without escalating uncertainty. In buy-and-build systems, meaning substitutes for missing structure during periods of transition. It lets organisations function coherently while formal systems are still catching up.

This isn’t symbolic work, it’s load-bearing.

Leadership behaviour inside imperfect cadence

Operating cadence governs when leadership effort gets consumed. Bennis’s work complements that by describing how leaders behave when cadence is uneven.

Effective leaders under strain don’t react to every signal, don’t treat urgency as a proxy for importance, and don’t allow volume to redefine priority. They reinforce rhythm through behaviour: showing up predictably, making decisions at consistent intervals, and resisting the temptation to compress everything into immediacy. Those behaviours stabilise time even before systems do.

Reconciling behaviour and system design

Nothing in Bennis’s work contradicts the argument that leadership capacity is constrained. It reinforces it. The behaviours he described matter precisely because capacity is finite — they let leadership effort go where it matters most, rather than getting dissipated across noise and uncertainty.

Leadership behaviour isn’t a substitute for system design — it’s how leaders operate responsibly inside systems that are still forming. The cases I’ve watched closely confirm a quieter version of Bennis’s claim: when leaders behave consistently in transition, the system holds together long enough for the structure to catch up.

Why this still matters

Buy-and-build strategies magnify moments when leadership must function before structure is complete. In those moments, org charts lag reality, cadence is unstable, and authority is still being negotiated.

Bennis offered a vocabulary for understanding these conditions, not a playbook for them. The vocabulary remains relevant — not because leadership is timeless, but because organisational strain is predictable.

What follows in Integration & Execution is where these behaviours get tested hardest — when plans meet capacity, when systems arrive, and when leadership judgment determines whether execution creates leverage or locks in fragility.

The conditions in 2025 are the conditions in which the premium has historically opened. Bain describes the construction environment as “still-fragile construction demand and an uncertain outlook.” Operating Benchmarks in the Building Materials Industry, 2024 edition, reaches the same picture from underneath, in its current-conditions reading of six federal sources through December 2025: prices off their peaks and segment-divergent, production soft for construction-tied manufacturing, residential and commercial pipelines weakening, and labor demand easing. The Bain framing and the federal-data reading agree. Exactly the inflection where through-the-cycle acquirers have, in prior cycles, captured disproportionate share of the next up cycle. The scale-to-scope rotation traced in The Scale Curve Has Run Out is the strategic shift these acquirers are now positioned to capture. The 690 bps figure is the historical measure of what positioning ahead of the next cycle has been worth. The reading that follows is what that positioning costs operationally, and why mid-tier PE platforms in particular tend to lose the capacity to deliver it.

The 690 bps gap and what it actually measures

The Bain figure is a cumulative measurement of a structural difference, not a one-cycle outperformance. Frequent acquirers do not outperform inactive peers by 690 bps in any single year; they compound a smaller advantage across multiple years and cycles. Operating Benchmarks 2024 reads that cycle reality directly out of the peer-set long history: across 27 years since 1998, fabricated metal products expanded in only 13 of them (48%), nonmetallic minerals in 18 (67%), wood products in 20 (74%), and wholesale trade in 21 (78%). Building products is a segment-divergent cyclical asset, and the cohort math Bain measures runs across roughly a dozen cycle inflections per peer industry over the long window. The compounding is what makes the gap durable, and it is also what makes the gap difficult to recover for a platform that drops out of the frequent-acquirer cohort for a single cycle. Re-entering the cohort requires rebuilding capacity that took multiple cycles to develop, and the capacity rebuild runs slower than the cycle window typically stays open.

The framing matters because the through-the-cycle discipline is usually discussed as a CEO-level choice: invest now while multiples are low, harvest later when they normalize. The framing is correct at the capital-allocation level. It is incomplete at the operating-capacity level, where the actual constraint lives.

Why through-the-cycle is harder for mid-tier PE specifically

Large strategics in building products typically maintain dedicated M&A functions with institutional buffers that survive cost cycles. The M&A team is a department; its staffing is governed by corporate policy and protected by institutional inertia. Down-cycle cuts can erode it but rarely eliminate it. Mid-tier PE platforms in building products can typically have M&A capability embedded inside the platform — integration teams that report to platform leadership, pipeline development that competes with operating-team priorities, post-merger integration playbooks maintained by the same people who run the day-to-day business. The structural vulnerability is asymmetric: the M&A capability that gets cut in a strategic’s down cycle is more easily restored than the M&A capability that gets cut in a mid-tier PE platform, because the strategic’s M&A function is institutional and the platform’s is operational.

The result is that the through-the-cycle dynamic that breaks compounding in buy-and-build breaks more decisively for mid-tier PE platforms than for large strategics. The 690 bps premium is structurally available to both kinds of acquirers. The structural vulnerability to losing it through down-cycle cuts is asymmetrically greater for mid-tier PE — and the gap shows up in the data as outperformance for the platforms whose sponsors recognize the asymmetry and discount their down-cycle cuts accordingly.

What through-the-cycle requires operationally inside mid-tier PE

The operating capacity to keep acquiring through a down cycle is not a single capability; it is a coordinated set. Integration capacity has to be maintained at staffing levels that look overhead-heavy when deal volume is low, and the staffing has to retain the institutional knowledge that PMI execution depends on — which means the people, not just the headcount, have to be preserved. M&A pipelines have to be cultivated as continuous processes, with relationship development running through quiet quarters rather than activated when deal flow picks up. PMI playbooks have to be updated through the down cycle so they don’t decay against current industry conditions. Operating cadence at both the platform and GP levels has to be structured around continuous acquisition rhythm rather than organic-only quarters.

Grant, Nilsson and Nordvall’s 2022 European Management Journal study of successful serial acquirers frames this academically: pre-merger capability has two components — expertise (individual tacit knowledge from repeated execution) and routines (organizational processes and structures) — and the elements that comprise each are “difficult to imitate or acquire” because they take many acquisitions to develop.

The mid-tier building products platforms I watched protect M&A capacity through prior down cycles treated it as an operating function with org-chart status, not as a project team activated when deals appeared — and the distinction was usually visible in the platform’s quiet quarters more than in its active ones. None of these line items survives a CFO’s down-cycle cost-cut review on its own merits. Each survives only because someone at the platform-leadership or GP level has decided the long-cycle capability investment is worth the short-cycle margin cost, and has the institutional standing to enforce that decision when the cost pressure is highest.

Building accidentally up, losing deliberately down

The asymmetry between how mid-tier PE platforms build acquisition capability and how they lose it is the structural source of the 690 bps gap. Capability tends to build accidentally during up cycles: deal flow scales integration teams, pipeline development happens through high-volume reps, PMI playbooks evolve through repeated execution. None of this requires a deliberate investment decision; the up cycle does the work. The capability tends to be lost deliberately during down cycles: integration teams get cut because they do not carry organic revenue, pipeline development gets paused because deal flow has dried up anyway, PMI playbooks get shelved because nobody is using them. Each of these decisions is rational at the moment it is made. The cumulative effect is to dismantle the capability that took the prior up cycle to build, exactly when the next cycle’s premium is starting to form.

In the building products organizations I’ve watched manage through prior down cycles, the capabilities that survived the cuts and the capabilities that didn’t determined the platform’s position when the next cycle opened — and the operating models that kept frequent-acquirer capacity intact were always more deliberate than the ones that didn’t. Long-arc leadership stewardship of operating capability through cycles where the short-term math argues for cuts is the underlying discipline. The 690 bps premium is what that stewardship is worth, measured cumulatively over multiple cycles.

The acquisition decisions nobody is calling acquisition decisions

Through-the-cycle acquisition is not a strategic posture; it is an operating discipline that has to be maintained continuously — and which platforms are positioned to capture the next cycle’s premium is being decided right now, in down-cycle staffing and pipeline and playbook decisions nobody is calling acquisition decisions.

The model is wrong in an important way. In buy-and-build systems, integration load compounds cumulatively rather than additively. Each acquisition alters the baseline from which the next one has to be absorbed, and even when integrations “succeed,” they leave residue: demands on leadership attention, coordination, systems, and learning capacity that don’t fully reset before the next deal arrives. Execution weakens not because teams forget how to integrate, but because the system is carrying more than it appears.

Integration load is a stock

Integration work often gets measured as a flow: hours spent, milestones achieved, synergies captured. The assumption is that once the activity concludes, the load dissipates.

In reality, each integration leaves stocks behind. Each acquisition adds new interfaces between teams, additional decision paths, embedded assumptions in systems and processes, partial standardisation layered on legacy practices, and unresolved cultural and operational differences. None of these disappear when the integration plan is “complete.” They remain in the system, shaping how future work gets done. This is why organisations often feel slower after successful integrations, less fluid rather than weaker.

Why experience doesn’t reset the clock

A common belief in buy-and-build strategies is that experience compounds naturally. The first integration is hard, the second easier, and by the fifth the organisation has “figured it out.”

Experience does accumulate, but not uniformly. What accumulates fastest is coordination complexity. What accumulates more slowly is learning consolidation. Under sustained acquisition pressure, lessons get learned but not fully embedded, new routines coexist with old ones, exceptions multiply faster than they get resolved, and leaders carry more interpretive burden. The organisation becomes more practised at managing integration workstreams while becoming less capable of absorbing additional complexity without friction.

Overlapping integrations change the baseline

Integration load compounds most visibly when integrations overlap, which in serial acquisition environments is the norm rather than the exception. One integration is stabilising while another begins, leadership attention is split across multiple absorption curves, systems are partially aligned in different directions, and teams are asked to adapt before prior changes have settled.

At that point, integration becomes ambient rather than episodic. The organisation operates in a permanent state of adjustment. Execution still happens, performance may still improve, but the margin for error narrows. This is when execution fragility begins to build quietly.

The illusion of stability

One of the most dangerous phases in buy-and-build systems is when performance remains strong while integration load is accumulating. Revenue grows, margins hold, dashboards look clean, and the organisation appears to be coping well. The pattern creates confidence that capacity is scaling alongside ambition.

Integration load doesn’t surface immediately in financial metrics. It shows up first as slower decision cycles, increased escalation, greater reliance on informal fixes, fatigue among high-leverage leaders, and reluctance to revisit earlier decisions. Each of those signals is easy to rationalise individually. Together, they indicate that the system is carrying more than it can easily absorb.

Why this matters for execution

Execution degrades not because people stop trying but because the cost of coordination rises. As integration load compounds, simple decisions require more alignment, exceptions become harder to resolve cleanly, systems become harder to change without disruption, and leaders spend more time reconciling the past than shaping the future. Execution shifts subtly from creating leverage to maintaining coherence. At that stage, execution still looks disciplined and it isn’t yet defensive, but it’s becoming less adaptive.

The path dependency problem

Once integration load accumulates, reversing course becomes difficult. Decisions made under load tend to lock in: systems chosen for speed become permanent, standardisation applied for clarity becomes rigid, workarounds harden into practice, and temporary structures acquire authority. Each of those choices reduces optionality. The organisation doesn’t fail, it becomes path dependent, which is the dynamic When Buy-and-Build Stops Compounding examines from the platform-strategy side. Later execution quality is constrained by decisions made earlier under pressure, often when alternatives were still available.

Why integration load is hard to see

Integration load is rarely named explicitly. It’s distributed across leadership calendars, informal coordination, system complexity, cultural translation work, and unresolved edge cases. Because it’s diffuse, it often gets mistaken for “the cost of growth.” But growth and integration load aren’t the same, growth expands opportunity, integration load consumes capacity. Confusing the two leads organisations to push harder precisely when restraint would preserve long-term execution quality.

Looking forward

Integration load doesn’t announce itself at exit. It reveals itself in how adaptable the organisation appears, how reversible past decisions feel, and how much confidence others place in the system’s ability to perform under new ownership. Those implications will get addressed directly later in this section.

For now, the important point is simpler: each acquisition changes the system that must absorb the next one. Execution quality depends not on the success of any single integration but on how much cumulative load the organisation is carrying when execution is asked to scale. The cases I’ve watched closely confirm a related corollary — by the time the load becomes visible in metrics, the system has usually been carrying it for longer than anyone tracking the metrics realised.

North America’s deal momentum in 2025 made the shift visible. Deal value in building products M&A rose 33% year-to-date in North America through September, even as deal value fell 48% in EMEA and 44% in APAC. The volume came back. What it bought, increasingly, was not more scale.

The HALO trade that traced this rotation at the asset-class level is now showing up inside individual sectors with the same operating-model consequences. Building products is one of those sectors. The reading that follows is what the rotation means for operators on the supply side, and for the mid-tier PE platforms now positioning for the next cycle’s exits.

The named scope deals of 2025 — and what they’re actually buying

Lowe’s acquired Foundation Building Materials for $8.8 billion to extend its reach into professional customers and specialty distribution. Home Depot acquired GMS to continue the professional-channel expansion it began with SRS in 2024. CRH acquired Eco Material Technologies for $2.1 billion, picking up scale in fly ash and pozzolans that no traditional cement scale play could have produced. Holcim’s bid for Xella, in a European parallel, is similarly a scope move, complementary building materials rather than more of what the buyer already had.

The trade press still discusses these in scale-era vocabulary because the absolute dollar figures are large. The operating logic underneath them is different. Lowe’s is not buying more retail volume; it is buying access to a customer profile its existing organization is not built to serve. Home Depot is not consolidating distribution; it is building a multi-channel architecture. CRH is not adding more cement; it is buying a capability the cement scale curve cannot deliver. Each transaction is being underwritten on capability or channel adjacency, not on cost-takeout or production-scale arithmetic.Operating Benchmarks 2024 measures the operating gradient these deals cross at the source-data level: distribution turns assets 1.95 times for every dollar of capital against manufacturing's 0.46 to 1.45 times, and runs at 8.8–17.3% refined overhead intensity against manufacturing's 15.1–25.3%. These are not adjacent operating models. A scope deal that crosses the gradient has to hold both.What scope deals demand operationally

The diligence pivot Bain documents, from cost levers to commercial levers, with revenue synergies as the new centre of gravity, is the symptom of a deeper operating-model gap. A scale deal can be underwritten on integration of like with like: combine two cement plants, eliminate overlap, optimize a single product flow. A scope deal cannot. It requires cross-selling architecture across product categories whose sales motions don’t naturally combine; channel management that handles multiple product narratives without flattening them into a single message; customer data hygiene that spans categories most legacy systems were not built to relate; sales organisations carrying differentiated value propositions to overlapping customer bases without diluting any.

It requires, in particular, decision-rights architecture that can hold multiple product-category P&Ls in parallel without defaulting to the single-category operating model the scale era produced. Single-category operating models tend to centralize decisions around the dominant product flow because that’s where margin lives. Multi-category scope businesses fail when the same instinct travels with them — adjacencies starve, capability advantages erode, and the scope premium the deal was underwritten on never materialises operationally.

The diligence-side observation Bain makes about scope being increasingly common because the buy-and-build math has stopped compounding the way it used to in scale-consolidated categories is the same observation read from the M&A side. When the scale curve has run out, doing more scale deals is no longer where compounding lives. Compounding moves to capability, and the operating model has to move with it.

What that asks of the operator who’s been running scale

In the building products organizations I’ve worked inside and watched closely, the operating model that delivered scale-curve advantages, single-category cost takeout, route density, manufacturing optimization, looked materially different from the operating model required to make a scope deal work in practice. The capabilities that produced returns in the scale era were not the wrong capabilities; they were the right capabilities for what the prior cycle’s M&A was actually buying. They are not the right capabilities for what the current cycle’s M&A is buying, and most operators trained on the prior cycle’s playbook find that out somewhere between eighteen and thirty months into a scope integration that the financial case said should already be producing results.

The pivot is not announced. It shows up in hiring choices, in sales-org redesigns, in data architecture investments, in operating cadence decisions that look like routine governance changes at the time they are made and reveal themselves as pivot decisions only in retrospect. The operators who can run scope-led building products platforms over the next cycle will be the ones whose operating models were built, or are being rebuilt, around capability integration rather than cost-takeout integration. The operators who can’t will be the ones whose operating models were optimized for a scale era the asset class has already moved past.

Where mid-tier PE sits inside the rotation

The named transactions above are large-cap strategics doing the scope rotation; mid-tier PE platforms in North American building products are not the actors doing those deals. They are increasingly the supply side. The scope-rotation demand profile that Lowe’s, Home Depot, and CRH are now underwriting creates a real opportunity for mid-tier PE to build specialty platforms specifically positioned for capability acquisition by strategics — rather than for the sponsor-to-sponsor handoff most prior-cycle mid-tier building products platforms were architected around.

The realistic 2027–28 buyer for a well-built mid-tier North American building products platform is increasingly likely to be a Fortune 500 strategic doing scope rotation, rather than a larger sponsor doing a roll-up. The buyer-profile flip from sponsor-handoff readiness to strategic readiness applies to building products in 2026 with unusual precision: the operating-model choices that produce a scope-readable platform, preserved capability definition in each acquired piece, channel and customer-data architecture that maps to a specialty capability thesis, sales organisations that carry differentiated value propositions rather than commoditised cost positions, are the choices being made now, in the next eighteen months, by the mid-tier platforms whose 2027–28 exits will show up in the next cycle’s deal data.

The mid-tier building products platforms I’ve watched get acquired by strategics for capability premiums tended to have one thing in common, they preserved enough product-category definition that the acquirer could see what they were paying for. The platforms that absorbed each acquired piece into a single platform identity got sold, when they got sold, for commodity discounts rather than capability premiums. The platform-and-add-on selection criteria that distinguish acquisitions whose individual case still matters from acquisitions absorbed indistinguishably into the platform are doing more work in this environment than they did in the prior one. In a scope-buyer market, the add-on that retains identifiable definition is the add-on that produces capability-premium pricing at exit; the one that doesn’t is the add-on that flattens the platform’s exit narrative into a roll-up case the strategic buyer pool has already moved past.

What 2026 is settling

2026 is the year the scale curve in North American building products concedes to the scope curve in the segments that matter most, and the operators who can build the cross-category architecture the scope playbook requires will be running the mid-tier platforms that the strategics’ capability demand is now being underwritten against.

In practice, target selection rarely begins with a blank slate. Most acquisition decisions get made from a narrow, time-bound set of available options shaped by market conditions, seller readiness, and timing more than by strategic ideals. Within those constraints, fit becomes less about choosing the perfect company and more about interpreting imperfect options correctly, which is where compatibility starts becoming dangerous in ways that are hard to see at the time.

The seduction of familiarity under constraint

Compatibility is attractive precisely because it feels legible. When options are limited, familiar operating models, overlapping customers, similar leadership profiles, and shared language provide a sense of orientation that reduces cognitive load at the moment decisions have to be made.

From the outside, these deals appear prudent, described as “low-risk,” “obvious,” or “clean.” From inside the platform, they also lower resistance: leaders expect fewer surprises, teams assume integration will be incremental. In constrained environments, familiarity feels not just safe but responsible. That is exactly what makes it risky.

Compatibility reduces visible friction while quietly increasing what’s worth calling structural coupling, the degree to which two organisations become dependent on shared decisions, shared timing, and shared judgment earlier than the system can reliably support. Fit is about how the two interact once joined, not about how similar they look beforehand.

Fit as interaction risk

Because acquisition choice sets are constrained, the critical error is usually less in which target gets selected than in how that target gets framed in the room where the decision is made. In buy-and-build systems, fit is better understood as interaction risk than as compatibility. The questions that matter are different:

• How tightly will decisions become coupled after close?

• Where will autonomy disappear in practice, even if it remains nominally?

• Which assumptions about pace, priorities, and performance will collide?

Highly compatible businesses tend to integrate more deeply and more quickly — not because they must, but because they can. Systems merge before assumptions are tested, informal coordination replaces explicit governance, and decision rights collapse into habit. That accelerates alignment, and it also eliminates insulation. When strain emerges, it propagates faster; when leadership bandwidth tightens, there are fewer buffers; when priorities shift, reversibility is limited.

Fit is therefore a function of how much strain the combined system can absorb at this moment in its development, not of how similar the two organisations appear on a fact sheet.

The quiet failure mode of “easy” deals

Most integration failures I’ve watched closely originated in deals that everyone perceived as easy. Those deals rarely fail dramatically, they erode quietly. Because compatibility masks interaction risk, early warning signs get interpreted as noise rather than signal. Teams assume issues will resolve once systems are fully aligned. Leaders tolerate temporary overload in the belief that scale or synergy will soon simplify the environment.

Deep compatibility often front-loads commitment rather than deferring it. Integration outpaces learning, decisions become entangled before trust is established, and small misalignments cascade because there’s no longer separation to contain them. By the time strain becomes visible, it’s no longer localised, the system is already carrying it everywhere. The problem isn’t execution; it’s interpretation under constraint.

Compatibility and the illusion of knowing

Compatibility also creates an illusion of understanding. When organisations look alike, leaders believe they know what they’re buying. That belief compresses diligence, narrows interpretation, and accelerates commitment, not because the information is better, but because the uncertainty feels lower.

This is particularly dangerous in serial acquisition environments, where prior success with similar targets reinforces confidence. Experience gets mistaken for predictability, and similarity hides uncertainty rather than eliminating it.

Fit gets revealed not by how little changes at close but by how the system behaves months later — when leadership attention is divided, integration work is unfinished, and performance pressure returns. At that point, selection decisions are no longer adjustable.

Reframing fit as a constraint question

A more useful way to think about fit is to ask three questions of the system, not the target:

• Where will interaction demand exceed leadership capacity?

• Which interfaces will require ongoing judgment rather than one-time alignment?

• How much coupling can the platform absorb now, given everything it’s already carrying?

These are questions of constraint rather than compatibility. A target that’s less familiar may fit better if it preserves boundaries, slows coupling, and allows learning to occur before deeper integration is attempted. A target that feels perfectly aligned may fit poorly if it collapses too many decisions into the same time horizon and the same leadership bandwidth.

Fit is a dynamic property of the system under load, not a static property of the target. The same target might fit one platform well and another platform poorly. The same platform might absorb one target cleanly at one moment and the same target poorly six months later.

Why this matters upstream

Because target universes are constrained, the danger isn’t choosing imperfect options, it’s misreading what those options will demand of the system once they arrive. Fit is a design judgment under uncertainty, applied with the platform’s current state in view.

Treating compatibility as a proxy for fit biases selection toward faster integration, deeper coupling, and earlier irreversibility, often without anyone naming that those commitments are being made by default. Understanding the distinction doesn’t simplify target selection, it makes it more honest about what the selection is actually committing to.

The question to put on the table before any target moves into deeper diligence isn’t whether it feels easy. It’s: what kinds of interaction is the system prepared to live with next?

The macro pressure traced in Dry Powder and the Pressure to Move has been landing in different places across the arc this piece closes, in higher entry multiples, in faster sector rotation, in compressed buyer pools, in 76% add-on velocity. The data above is the LP-side print of the same pressure. What it produces on the operator side is the topic this piece is about: a holding pen that nobody calls a holding pen, and one the asset class has not yet priced into the people who have to live inside it.

The LP-friendly answer and what it leaves out

The ILPA LP Sentiment Survey 2025–26, cited in Bain Figure 21, records a striking adaptation: two-thirds of LPs would accept extended holds for better MOIC. The asset class has read this as permission to slow down. The framing is rational at the LP level (preserve unrealized value rather than force exits into thin markets) and at the GP level (extend the hold, accumulate operating value, exit into a better window). It is silent on what extending the hold does to the people who have to run the platform during the extension.

The silence is not careless; it reflects the structural shape of the conversation. LPs negotiate hold-period flexibility with GPs. GPs negotiate extension structures with their own LPs and with continuation-fund providers. Neither conversation routinely involves the operating-side people whose incentive structures, retention architecture, and career arcs were calibrated to a hold the asset class is now extending past. The LP-friendly answer to the distribution drought has, by structural design, no good place for the operator-side cost to be voiced.

The cohort the silence excludes

CEO LTIPs are calibrated to typical-hold assumptions even when individual sponsors are careful not to prescribe exit timelines explicitly. Vesting schedules, performance milestones, and retention-bonus architecture all assume something — usually something close to the asset class’s typical five-to-six-year hold — and the architecture carries forward into extended holds without the calibration being rebuilt. The operator who joined a 2018-vintage platform under those calibrations is now eight years in. The hold may run another two or three years before exit. The first-24-months operating-model choices that defined the platform’s design are now being executed by the same person, on the same incentive structure, eighteen months past the date both were originally calibrated to.

Most of the CEO transitions I’ve followed in PE-backed companies show a similar pattern: tenure was designed for one length of hold, and when the hold extends beyond it, the people who were hired into the original timeline are the ones who carry the cost of the extension. The operating cadence that defined the early hold is harder to maintain at year seven than at year three, not because the CEO has changed, but because the cadence was calibrated to a finite-horizon engagement that has been quietly converted into an indefinite one.

The second-bite math

The asset class’s response to the retention question, where it has one, is the second bite of the apple, rolling LTIPs into the next vintage, offering equity in a continuation fund, structuring a new compensation package for the extended hold. The math is rational at the firm level and was rational at the operator level when the typical second cycle ran four or five more years. It is harder to make rational when the second cycle looks likely to run another six or seven.

The retention conversations I’ve followed in trade-press coverage over the past two years describe a pattern the data hasn’t yet caught: second-bite-of-the-apple offers increasingly arrive at the moment a long-tenure CEO is least able to accept them. The CEO who has just crossed the seven-year mark, who is being asked whether to commit to another six- or seven-year cycle in the same role, is making a different calculation than the CEO who crossed the four-year mark in 2019. Adaptive leadership over a long arc is not the same thing as committing to a second cycle whose terms are functionally identical to the first one — and the asset class has not yet built incentive structures that recognize the difference.

Where this arc lands

The pieces in the arc this closes have traced the macro pressure landing in different places: at the deal in higher entry multiples, at the exit in compressed buyer pools, at the platform in faster sector rotation, at the integration model in 76% add-on velocity. What this piece traces is where the same pressure lands last and most personally — at the operator whose career was structured around the prior cycle’s timelines, whose LTIPs were calibrated to a typical hold the asset class no longer reliably delivers, and who is now being asked, often without the conversation being framed this way, to absorb the cost of the asset class’s extension.

When the LP-friendly answer to the distribution drought is “hold longer for better MOIC,” the asset class is asking a cohort of operator CEOs to absorb the cost of an extension their LTIPs and retention architecture were calibrated for but never explicitly promised — and by end of 2026 a measurable share of those CEOs may be receiving second-bite-of-the-apple offers that potentially come with another six or seven years’ commitment, in a math that no longer adds up the way it did when the first cycle started.

The narrower question

The distribution drought is an LP-vs-GP debate everywhere except inside the platforms running through the extension, where it is something narrower: a stewardship problem for the people who came in to run a business at one timeline and are now being asked to keep running it at another, without the incentive architecture catching up to the change. By the end of 2026, that mismatch will be visible in the second-bite-of-the-apple offers being declined by long-tenure CEOs and in the org charts of the platforms whose retention design hasn’t kept pace with the holds they’re now running.

That phase is the one most accounts of buy-and-build describe. Less often examined is what happens after, when the strategy doesn’t collapse but quietly stops producing incremental advantage. Growth continues but value creation flattens. Integrations still “complete” without learning translating into leverage. Leadership feels busier than sharper, and optionality narrows even as the platform grows larger. Nothing is obviously broken, and something has stopped compounding.

Why early success is a poor signal

Early buy-and-build success often gets read as proof of robustness, the platform can integrate, leadership can stretch, systems can absorb change. The reading is usually wrong. Early success is more often a product of conditions that don’t persist: excess leadership attention available because the second deal hasn’t arrived yet; informal coordination mechanisms still functioning because the team is small enough for them to hold; slack in systems and processes that hasn’t been spent; and the goodwill that comes from novelty and momentum.

Those conditions let organisations compensate for structural gaps through effort. Leaders step in personally, problems get worked around, and learning feels rapid because each acquisition is still meaningfully different from the last. Those behaviours aren’t wrong, they’re often exactly the right thing in the moment. The risk is that they mask whether capability is actually being built or merely borrowed from future capacity. Borrowed capacity has to be repaid, usually right around the third or fourth acquisition.

Experience does not automatically become capability

In theory, repeated acquisition should improve performance: organisations learn, integration routines solidify, decision-making improves. Many frameworks assume a smooth learning curve. In practice, experience often accumulates faster than the organisation’s ability to convert it into durable capability.

The conversion fails because learning in buy-and-build environments isn’t additive by default, it competes with ongoing execution. Integration work consumes attention, leadership absorbs ambiguity that might otherwise have been resolved, and systems get patched rather than redesigned. Over time, experience layers on top of unresolved constraints, and the organisation has seen more without necessarily absorbing more.

This is why later acquisitions can feel harder than earlier ones, even when the organisation is objectively larger and more experienced. The system is carrying more history, more interfaces, and more unexamined assumptions. Learning has occurred. Whether it has compounded is a different question.

The quiet shift from growth to maintenance

A subtle transition often marks the end of compounding. Early in a buy-and-build strategy, leadership attention goes toward building, shaping the platform, establishing norms, defining what “good” looks like. Later, the same attention goes toward maintaining, managing exceptions, resolving friction, preventing drift.

The shift is rarely explicit. It shows up as more time spent coordinating rather than deciding, more effort required to achieve the same outcomes, fewer decisions that feel reversible, and increasing reliance on structure to hold things together. None of those signals indicates failure on its own; together they indicate saturation. At that point, additional acquisitions stop extending capability and start consuming it. Growth continues; the system’s ability to learn from growth plateaus.

Why compounding stops without anyone choosing it

One of the harder things about this dynamic is that no single decision causes it. Compounding stops because early workarounds never get fully retired, integration practices harden before they’re fully understood, leadership bandwidth becomes the limiting factor, and sequencing decisions prioritise momentum over consolidation. Each choice is locally rational. Together, they shift the system from one that builds capability into one that protects coherence.

From inside the platforms I’ve studied carefully, the experience is usually that the organisation is working harder to stay in the same place. From outside, the strategy still looks intact. The gap between those two views is one of the better leading indicators that compounding has quietly stopped.

What this means for buy-and-build strategy

The implication isn’t that buy-and-build inevitably stalls. Compounding is fragile, and it has to be actively protected. It comes from converting experience into capacity faster than complexity consumes it — which means the conversion has to be deliberate, with pauses built in for it to happen.

That conversion requires retiring practices that no longer scale, naming the moments when leadership effort is substituting for system design, and resisting the assumption that past success guarantees future absorption. It also requires the first add-on test — using each new acquisition as a read on whether the platform’s actual capacity has grown since the last one, rather than assuming it has.

Most buy-and-build strategies that stall don’t stall from poor conception, they stall because early success delays the recognition of where the real limits live. By the time the limits become visible, the organisation has often committed to a sequence that can’t easily be unwound.

The next piece in this section examines those limits more directly, starting with the decisions in buy-and-build that cannot be reversed once they are made.

The trajectory that produced the 76% print was already visible. Capital pressure of the kind traced in Dry Powder and the Pressure to Move has favoured add-ons for years, because they deploy faster, face less competition than platform deals, and absorb capital without requiring a new operating thesis. What’s new is not the trend; it is the share threshold. When add-ons account for three out of every four buyouts the asset class makes, the question of what an add-on actually is — to the platform, to the eventual buyer, to the operating model that has to integrate it — stops being a portfolio-construction detail and becomes an architectural one.

The signal at large-cap, the mechanism at mid-market

Two of 2025’s largest sponsor-to-sponsor exits were add-ons being sold separately rather than platforms — Hometown Food’s $601M Chef Boyardee acquisition and NRG’s $560M Texas gas-gen pickup — both documented in PitchBook’s 2025 Annual US PE Middle Market Report. Both transactions were large-cap and should not be read as evidence of MM dynamics. The pattern of add-ons being carved out and sold individually is showing up first in large-cap, where buyer pools are deeper and the standalone underwriting case for a single asset is easier to construct.

The mid-market mechanism is different. MM platforms typically exit whole, with their add-ons going to the buyer as part of the platform sale rather than being separated out. What the 76% print changes for the mid-market is not the exit pattern itself; it is the cumulative effect of an architecture where most of the platform’s growth is now happening through add-ons rather than organically. The MM operator question that follows is not about whether add-ons will be sold separately at exit — most won’t — but about what the platform actually becomes when 76% of its acquired components are add-ons whose individual cases were never required to hold up beyond the integration moment.

What “add-on as the modal acquisition” actually changes

When add-ons were one in three buyouts, a platform could be defined by its initial acquisition and its operating thesis, with add-ons absorbed into that defining identity. When add-ons are three in four, the platform is increasingly defined by the cumulative shape of its add-on architecture — which acquisitions were chosen, how they were sequenced, and what each one contributed beyond capital deployment. The platform’s identity becomes a function of its add-on history rather than the other way around, and most operating models built for the prior ratio do not produce a coherent answer to the question of what the platform now is.

The platform-and-add-on selection criteria the buy-and-build literature has long distinguished are doing more work in this environment than they did in the prior one. When add-on selection drives most of the platform’s eventual shape, applying platform-grade selection criteria to add-on choices is no longer optional — it is the difference between a platform whose add-ons cohere into something valuable and one whose add-ons stack into something that doesn’t.

The compounding question

The 76% print also raises a question that runs underneath the velocity itself: whether the add-on-heavy architecture is producing compounding or merely stacking. Compounding through buy-and-build requires that each acquisition reduce the integration load on the next one — that operating capacity, customer overlap, and capability adjacencies grow through addition. Stacking, by contrast, produces growth in revenue and EBITDA without producing growth in coherent capability, and the difference is usually not visible at the velocity level. A platform doing eight add-ons a year and a platform doing three may both look healthy in deal-flow terms; only one of them is producing a more valuable platform with each acquisition.

The platforms I evaluated as acquisition candidates from a strategic vantage fell into two categories — those whose add-ons still had identifiable cases at the deal table, and those that had absorbed everything into a single identity that was harder to value at exit. The distinction wasn’t velocity-related. The platforms with high add-on counts that read as compounding had usually preserved enough definition in each acquisition that a downstream buyer could see what each piece contributed. The ones that read as stacking had often run integration as the elimination of distinction — converting every add-on into the platform’s existing identity as quickly as possible — and the resulting asset was harder to value precisely because the components had been blended past the point of being individually visible.

The underwriting question, on every acquisition

When the modal acquisition is an add-on and a meaningful share of platforms will eventually be evaluated by buyers who underwrite at the asset level rather than the platform level, every add-on now carries an implicit underwriting requirement that the prior environment’s add-ons did not. Each acquisition has to make sense not only as a contribution to the platform’s current operating thesis but as a piece whose case can still be reconstructed if a downstream buyer asks why it was acquired and what it now contributes. The buy-and-build platforms I’ve watched run integration as absorption — adding companies to a unified identity — produced different results from the platforms that ran integration as portfolio architecture, where each acquired piece kept enough standalone definition to be valued independently.

The buyer-fit question that becomes operationally specific in When the Buyer Profile Flips lands here in a related form. The platforms whose add-ons retain identifiable cases preserve more options at exit — both because some add-ons may be sold separately and because corporate strategics underwrite synergy at the asset level even when they buy whole platforms. The platforms whose add-ons have been absorbed into indistinguishability have narrowed their buyer pool to sponsors willing to underwrite the platform as a single, blended whole, and that pool is meaningfully smaller than it used to be.

What the operating model now has to do

The integration playbook for an environment where 76% of acquisitions are add-ons is not the integration playbook the prior environment produced. Integration capacity becomes binding much faster when the cadence of acquisition is higher and each one carries its own underwriting requirement. The failure modes that produce indistinguishable absorption — speed-prioritized integration that erases standalone definition, technology consolidation that deletes operating-model variation, leadership consolidation that removes the people who carried the add-on’s case — are no longer just integration risks. They are exit-value destruction mechanisms, and they compound at the velocity the asset class is now operating at.

Most operating models built in the prior add-on environment treat integration as the work of making each acquisition disappear into the platform. The work the new environment requires is closer to the opposite — making each acquisition contribute to the platform while preserving enough of its individual case that it remains visible to the people who eventually have to value it.

What the underwriting question actually settles

When 76% of buyouts are add-ons and a meaningful share of those add-ons may eventually be sold separately, the integration choice is no longer a sequencing question — it is an underwriting question being asked of every acquisition the platform makes, and most operating models still answer it as if the add-on were going to disappear into the platform forever.

Bain’s 2026 Global M&A Report supplies the corollary at the broader corporate level. Roughly 60% of major 2025 deals were scope deals — capability adjacency, complex assets, regulated industries — the highest share on record. The HALO concentration in PE matches a wider buy-side rotation toward operationally complex assets, and the alignment between the two data sources is what makes the Q1 PE print structural rather than episodic. Read alongside Dry Powder and the Pressure to Move, what the rotation describes is capital that needed to move finding the assets the public market still values, and those assets being structurally different from the ones the asset class spent the last decade building operating playbooks for.

The other side of the software story

When Capital Concentrates Into the Sector That’s Resetting traced the same rotation from the opposite end. Software at 18% of US PE deal value in 2025 was the asset class concentrated into the sector that was about to reset. HALO at 31.2% in Q1 2026 is the asset class moving into the sector public markets are still rewarding. The two prints are the same shift recorded at different stages — the exit from one preference, the entry into another — and they describe a faster pivot than asset-class rotations usually run.

What’s unusual is the compression. A sector preference that took ten years to build is being unwound across two reporting periods, and the operating models built inside it do not adjust on the same timeline as the capital allocation.

The HALO print and what it actually captures

A 14% long-run average climbing to 31.2% in a single quarter is a specific signal, but it is also a single-quarter print, and the composition matters. HALO is dominated by mega-deals at the largest sponsors, where dedicated infrastructure and industrials funds have been operating for years. The AES take-private is exactly that pattern — strategic capital from sponsors with purpose-built funds, joint with sovereign and pension co-investors, deploying into a regulated asset that has its own operating discipline and its own engineering culture. None of that is new at the firms doing it.

What the print describes more precisely is the share of total US PE capital being absorbed by transactions whose operating logic is materially different from the sector PE has been most concentrated in. The compounding consequence falls on the firms whose playbooks weren’t designed for the asset profile now claiming that share.

Where dedicated funds change the question

The popular framing of the HALO trade — that PE is moving into operationally complex assets it doesn’t know how to run — is the wrong framing. Large sponsors with dedicated infrastructure, industrials, and energy funds have been running capital-heavy regulated assets every day for years, and many do it well. The capability is not the question.

The actual question is where the operator-asset fit lands at firms without that dedicated capacity. The mid-market sponsor with three software platforms, a healthcare services holdco, and one industrial roll-up is being asked, increasingly often, to evaluate HALO-adjacent opportunities — carve-outs from corporates rotating their portfolios, complex services platforms with regulated end-markets, asset-heavy specialty businesses that were previously more clearly outside PE’s preferred profile. The capability gap at those firms is real, and it is not something specialized funds at the largest sponsors solve for.

What the operating playbook actually carries

The operating discipline PE has built inside its preferred sector for the last decade is a specific toolkit. Operating cadence designed for software companies emphasises weekly metrics, monthly cohorts, quarterly cadence reviews, and rapid feedback loops on customer behaviour — disciplines that fit GTM and retention engineering and produce diagnostic signal in those contexts. Capital-heavy regulated assets do not produce signal on the same cadence. Engineering capacity, CAPEX governance, regulatory affairs, and asset-life management run on multi-year timelines, and the operating reviews that work for a SaaS business produce noise rather than diagnosis when applied to a transmission asset or a specialty industrial.

Most of the cross-sector PE operating transitions I’ve followed in the trade press show the same pattern: the metrics that worked in the prior asset class produce misleading signals in the new one, and the corrections come later than they should have. Leadership models built for software-era operating partners emphasise speed of iteration and tolerance for ambiguity in product-market fit. The leadership models that actually fit capital-heavy regulated assets emphasize patience with long capital cycles, stewardship of asset-life economics, and discipline against the kind of operating overlay that produces visible activity but degrades the asset’s underlying logic.

When buy-and-build stops compounding the way it did

Part of what the HALO trade represents is the asset class admitting that compounding through small bolt-on acquisitions is not producing the returns it once did. Scale dealmaking in capital-heavy regulated assets is the structural alternative — fewer, larger transactions, with operating value coming from asset-life management rather than integration synergies. The mid-market sponsor running a six-platform buy-and-build portfolio in services or software is now operating in a context where its peer firms are increasingly making one or two scale deals per year in adjacencies that look operationally foreign.

The org-chart shifts I’ve followed in PE typically precede sector rotations by six to twelve months, becoming visible before the rotation reads as a trend in the data. The hiring patterns at firms with dedicated HALO capacity have been visible for two years. The question is what the org charts at firms without dedicated HALO capacity look like by the end of 2026 — which firms hire ahead of the curve, which hire after the misallocation is visible, and which decide to stay out of HALO-adjacent deals entirely.

What 2026 is sharpening

The HALO trade is a capital-allocation shift, not a capability claim — large sponsors with dedicated infrastructure and industrials funds have been running these assets for years. The operator-side question 2026 is sharpening is what happens at the mid-market firms without dedicated HALO capacity, when a software-era operating playbook gets applied to assets it wasn’t built for — and how visible the mismatch becomes before the next bid arrives.

The macro pressure traced in Dry Powder and the Pressure to Move lands in the software reckoning the way it lands in most sector-specific stress: capital that needed to move went to the sector with the most familiar underwriting case, and the sector then re-priced. The Q1 print is not a verdict on whether AI displaces SaaS — that argument will run for another decade. The more immediate operator question is whether the moat the platform was actually designed around is the moat that survives the reset.

The concentration print and what it actually means

A 14% long-run average climbing to 18% in a single year is not, by itself, a crisis. Sector concentration shifts every cycle — capital follows perceived durability, and software has been the asset class’s preferred underwriting case for most of the last decade. What makes the 18% print structurally different is its simultaneity with the multiple reset. The concentration happened just as public software valuations broke down through their long-run support, and the deals struck inside that 18% were almost all underwritten on multiple assumptions the public market has now invalidated. The vintage 2024–25 software cohort is, on the whole, a cohort that paid for a future the market has stopped paying for.

This is not the situation that produced the 2022 reset, when compression hit a cohort underwritten in 2020–21 at peak vintage. The current reset is hitting deals struck on the assumption that the prior one had fully cleared. That assumption was wrong, and the cohort that absorbed it is the cohort whose operating model now has to make up the difference.

The exit market priced the reset first

PitchBook’s 2025 Annual US PE Middle Market Report records the corollary: middle-market IT exits collapsed roughly 38% year-over-year in 2025, even as overall MM exit activity recovered. The exit market did the pricing work the deal market had not yet done. Sponsors holding software platforms underwritten in 2023–25 found the bid disappearing while the deal market was still pricing new software entries at concentration-era levels.

The lag between the two is not unusual; deal markets typically follow exit markets by two to four quarters in any sector reset. What is unusual is the concentration multiplier. When 18% of US PE deal value sits in a sector whose exit market has just contracted by 38%, the sponsor population carrying that exposure is unusually large — and the operating decisions that now matter are the ones that determine whether each individual platform clears, regardless of whether the sector overall does.

Where the moat actually lives

PitchBook’s analyst note frames the durable software moat with notable precision: switching costs, compliance, retraining requirements, and incumbent integration. The list does not include features, product roadmap, or AI capability. This is closer to The Industrialist’s integration-capacity vocabulary than to most sector commentary, and it is closer to the actual mechanism by which software businesses survive valuation compressions than the features-and-roadmap framing the trade press tends to default to.

The frame from The First 24 Months Now Decide the Deal applies directly. The moat the platform was designed around is one of the operating choices that becomes visible early — typically by month 24. Platforms that built operational lock-in into customer workflows, that hardened compliance and integration into their architecture, that produced retraining costs the customer cannot ignore — those platforms have a moat that survives a multiple reset because the moat is not priced in the multiple. Platforms that built features and rode growth assumptions have a moat that exists only at the multiple the market was paying.

Two software businesses, two reckonings

The reset does not punish “software” in any uniform sense. It punishes thin software — applications whose lock-in is shallow, whose compliance load is light, whose customer can switch without operational consequence — and it largely spares software whose customers cannot actually leave. The platform-selection distinction in the buy-and-build literature applies here in unusually direct form. Buy-and-build software platforms with deep organizational integration look like one kind of asset under reset conditions; thin-application SaaS roll-ups underwritten primarily on ARR growth and multiple expansion look like another.

In the cases I’ve studied, the software platforms that came through prior valuation resets intact were the ones whose customers couldn’t actually leave — not because the alternatives were worse, but because the switching cost was real. The platforms that did not survive prior resets were the ones where the lock-in was narrative rather than operational. The reset itself does the sorting; the operating model determines which side of the sort the platform lands on.

What the operating model is actually building

Most software deals struck in the concentration period were underwritten on a growth-plus-multiple case the reset has now broken. The growth assumption is independently questionable in many cases — AI capability is reshaping what enterprise customers will pay for incremental functionality — but the multiple assumption is the part already invalidated. Sponsors who paid 2024-vintage prices for businesses underwritten on 2024-vintage multiples need their platforms to clear a new market reality, and the operating choices that determine whether they can are being made right now without being framed as software-reckoning responses. Leadership configuration gets framed as operating-model design. Customer workflow integration gets framed as product roadmap. Compliance hardening gets framed as risk management. The choices add up to either a moat the new market will pay for or one it won’t.

Most of the software platforms that survived prior compressions did so because the operating work that built the lock-in had been compounding quietly for eighteen months or more before the compression arrived. The survival was a cumulative result, not a strategic response — eighteen months of operating decisions that, in retrospect, had been building the moat the new market still valued.

What 2026 settles

The software reckoning is not going to be resolved by a single quarter’s print, and the AI-displacement debate will outlast every operator currently in the asset class. What 2026 settles is something narrower: which vintage 2024–25 software platforms had the operating moat the new market still pays for, and which had the narrative moat the old market accepted in its place. The sorting is already underway in the exit market; it will reach the deal market by year-end. 2026 is the year vintage 2024–25 software platforms get marked to a market that no longer believes in multiple expansion — and the platforms that come through it intact will be the ones whose moats were operating moats, not narrative ones.

I’ve watched several leadership teams arrive at this misread, and the misread itself is rarely what causes the damage — the damage usually comes from what happens after, when the team interprets the misread as a signal to push harder rather than pause longer. That interpretation compounds quietly, until something visible breaks.

In the platforms I’ve studied, what looks like discipline-as-conservatism is often the thing keeping the system from consuming the leadership capacity it would need for the next acquisition. The teams that abandoned that discipline rarely failed at the next acquisition itself. They failed two or three deals later, in ways that traced cleanly back to commitments made before the system was ready to carry them.

Why most failures don’t look like failures at the time

Most organisations fail not from pursuing too little opportunity but from pursuing opportunity before the system can absorb it. The damage rarely appears at the point of decision. It shows up later — displaced in time and location, after commitments have hardened and reversibility has quietly disappeared.

Early on, things look fine. Revenue grows, integration milestones are “on track,” leaders stretch and adapt, and the organisation absorbs more than expected. From the outside, momentum looks real and self-reinforcing. Confidence rises, and with it the belief that capacity is expanding naturally alongside ambition.

That phase — the one where everything is working and confidence is rising — is the most dangerous one. What discipline actually does when applied early is preserve optionality: it protects leadership bandwidth, decision quality, and learning capacity before any of those assets visibly deteriorate. Because the protection is invisible in the short term, discipline reads as conservatism. It’s almost impossible for the people inside the system to see clearly until well after the window for using it has closed.

What “not yet” actually costs operators

Saying “not yet” requires explaining why progress should pause even when results look strong. It means trading visible momentum for invisible resilience. And it usually means absorbing frustration from teams who are already carrying heavy loads and feel capable of more — at least for now.

From a deal team’s vantage point, the signal can be even harder to read. The business is performing, the thesis still holds, and the market opportunity hasn’t changed. The instinct to ask why hesitate, why sequence instead of accelerate, why introduce friction when velocity seems available — that instinct is rational. The answer usually has less to do with ambition than with capacity, measured not by how much the organisation can endure but by how well it can decide, coordinate, and learn under increasing load. Those qualities degrade long before financial performance does. By the time the metrics show strain, the system has usually been overloaded for some time.

The conservative effect of undisciplined momentum

This is why undisciplined momentum is far more conservative in effect than it appears. When organisations move before constraints are understood, they lock in assumptions that can’t easily be reversed: capital gets committed against untested mechanisms, leaders become bottlenecks, integration load accumulates faster than it can be resolved, and the cost of stopping rises precisely when stopping would be most valuable. At that point the organisation is no longer choosing speed. The speed is choosing the organisation.

Discipline is easiest to abandon when things are going well. Strong performance creates narrative confidence, and external validation increases — from investors, sellers, and advisors alike. Dissent softens. Teams begin to believe that capability scales automatically with success. In those moments, restraint can feel almost irresponsible, as if leadership is failing to capitalise on momentum that is just sitting there.

But momentum and capacity are different things, and the difference shows up late. Capacity is built quietly through sequencing, stabilisation, and deliberate constraint. It depends on role clarity, decision rights, integration absorption, and the system’s ability to convert experience into learning rather than exhaustion. These are slow variables. They respond poorly to pressure.

What I notice when discipline is absent vs present

When discipline is absent, organisations compensate through heroics. Strong leaders take on more, decisions get centralised, problems get worked around rather than resolved, and learning gives way to coping. From the outside, the system still appears functional. Sometimes it even appears impressive. From the inside, the strain is already visible to anyone honest enough to look at the calendar of the people running the platform.

Disciplined systems behave differently. They create space — letting leaders decide rather than react, making trade-offs explicit rather than implicit, and preserving the ability to slow down without losing credibility because slowing down is framed as stewardship rather than retreat. That preserved ability is the real source of strategic freedom in buy-and-build, and it’s something most platforms try to manufacture too late, after they’ve already spent the discretion they would have needed.

What this means in practice

For operators, discipline protects the organisation from consuming its own leadership capacity in pursuit of short-term progress. For investors, it protects the platform from locking in fragility that can’t be undone later with capital or talent alone.

Discipline doesn’t eliminate risk — it changes when the risk gets taken, and whether the organisation can still survive its own success when the risk surfaces.

That is what restraint, applied early, actually does. The most enabling decisions in complex systems are often the ones that look most conservative in the moment. The clearest sign you got it right is usually that you never had to find out what would have happened if you hadn’t.

One quarter is not a trend. But the mechanism behind the Q1 print is not noise. Corporate balance sheets have rebuilt across two years of constrained M&A, and EY-Parthenon’s most recent CEO survey shows 62% of US CEOs planning M&A activity in the next twelve months — a level of stated intent that maps directly onto the corporate share of exit value PitchBook just recorded. Read alongside the macro pressure traced in Dry Powder and the Pressure to Move, what the Q1 reversal describes is not a temporary swing in deal flow. It is a re-emergence of the corporate strategic as the buyer the sponsor has to design the asset to please — and that re-emergence has consequences on the operating side that show up well before the exit conversation begins.

The 2025 inversion was real, and it set the wrong default

The conditions that produced the 2025 inversion were specific. Sponsor-side dry powder was at record levels, GP-to-GP exits offered a path through a corporate market that had gone quiet, and continuation funds gave platforms one more way to clear without going to a strategic. The inversion was not invented; it reflected actual capital flows and actual exit paths. What the trade press read as a new normal, however, treated those conditions as durable. Most operating playbooks built or refreshed in 2024 and 2025 quietly assumed a sponsor exit. Integration runway was designed for the next GP to extend. Growth-thesis architecture was built to be re-underwritten by another platform investor. Holdco overhead was tolerated because the next sponsor would tolerate it too.

The default shaped the platform. That part matters more than the print itself, because the platform that was being shaped looks slightly different from the platform a corporate strategic underwrites — and the difference compounds the longer it goes unnoticed.

The Q1 2026 reversal has mechanism behind it

The Q1 2026 print could be dismissed as a single-quarter wobble if the underlying drivers were random. They are not. Corporate cash positions have rebuilt across two years of subdued M&A. Strategic premiums have re-opened in sectors where scope dealmaking is back on the table. CEO M&A intent, as the EY-Parthenon survey records it, is at levels that historically precede multi-quarter sequences of strategic acquisitions, not single-quarter spikes. The 26.9% sponsor-to-sponsor figure may move when Q2 prints; the corporate appetite that produced 58.4% on the other side is the part with mechanical durability.

Whether sponsor-to-sponsor recovers to its 2025 share or settles into its longer-run 41.4% is a question about deal flow. Whether the operating model the platform was designed under fits the buyer most likely to actually pay for it is a different question, and the answer is being shaped right now in decisions nobody is labelling as exit decisions.

Two buyer profiles, two underwriting frames

A sponsor-to-sponsor exit and a sale to a corporate strategic look identical at the headline — both are exits at multiple X — but they reward different operating-model choices in the year before they happen. A sponsor underwrites platform fit, integration runway the next GP can extend, and a growth thesis that will survive being re-marketed to LPs. A corporate strategic underwrites synergy: identifiable cost takeout, clean carve-out boundaries, integration-ready data architecture, leadership configurations that map onto a parent company’s existing functions, and a holdco overhead profile that disappears cleanly into the buyer’s existing structure rather than being absorbed at additional cost.

These are not subtle differences. The cases I’ve studied where an asset moved smoothly from a sponsor-handoff plan to a strategic exit usually had the corporate-readiness work already in place by the time the pivot was named. The platforms that read as well-prepared for a strategic exit had been quietly building toward one for at least the prior eighteen months. The platforms that scrambled to reposition once the buyer profile shifted usually showed it.

What the first 24 months were already deciding

The frame from The First 24 Months Now Decide the Deal applies directly. Most of the design choices that determine which buyer can underwrite the asset are made in the early part of the hold — the period when integration architecture is set, when operating-model design is still pliable, and when overhead structure is being built rather than dismantled. By the time exit conversations start, those choices are already mostly fixed. The platforms that absorbed the 2025 inversion as the new normal and built sponsor-handoff defaults into their early-hold design are now eighteen months into a configuration that fits one buyer and not the other.

This is the part of the buyer-profile flip that operators control, and it is also the part the macro data does not show. The Q1 PitchBook print describes what cleared in the quarter; it does not describe how many platforms were ready for the buyer profile that actually showed up.

The choices nobody is calling exit decisions

The decisions that determine which buyer can underwrite the asset are usually framed as something else at the time they are made. Carve-out boundaries are framed as integration choices. Data architecture is framed as systems work. Leadership configurations are framed as operating-model design. Holdco overhead is framed as governance. None of them is described, internally or externally, as an exit decision — and yet each one constrains the buyer profile the platform can plausibly attract eighteen to thirty months later.

In many of the platforms I’ve looked at carefully, the buyer-profile question is hardly ever named directly during the first hold years. It is decided through a sequence of operating choices that look like normal hold-period decisions and only resolves into a buyer-profile commitment in retrospect. A platform whose data sits across three GPs’ template stacks and four bolt-on acquisitions’ inherited systems is not unsellable to a corporate strategic; it is just substantially harder to underwrite as one. A platform whose holdco runs at PE-typical scale because the sponsor was building toward a GP handoff that would tolerate it is not unattractive to a strategic; it is just discounted at the table because the strategic has to absorb the cost of taking it apart. Each of these features makes the deal a different deal — and the difference compounds at exit, where the corporate buyer’s underwriting frame either lines up with the operating reality or doesn’t.

The eighteen-month window

For platforms targeting a 2027–28 exit, the window in which operating-model choices remain open is shorter than the gap to exit suggests. Most of the design decisions that determine buyer-profile fit will be effectively locked in by mid-2027. That leaves roughly eighteen months in which the pivot from sponsor-handoff readiness to strategic readiness — if it is going to happen — has to be made operationally, not narratively. The pivot is not announced. It does not appear in board materials. It shows up in the choices made about carve-out boundaries, data architecture, leadership configuration, and overhead structure during the months when those choices are still treated as operating choices rather than exit choices.

What makes this difficult is that the buyer-profile question rarely arrives as a clear signal. The sponsor-to-sponsor share will fluctuate quarter to quarter. Corporate appetite will move with cash positions and CEO confidence. The platform’s actual buyer in 2027 will be visible only in retrospect. The operating decisions, however, have to be made now — and they are being made now, whether or not the buyer-profile question is being asked alongside them.

What’s actually being decided right now

Sponsor-to-sponsor will recover. Corporate appetite will fluctuate. The buyer profile of any individual platform’s eventual exit will not be visible until the exit happens. None of those uncertainties changes the operational reality that runs underneath them. If the realistic 2027–28 buyer for your platform is a Fortune 500 strategic rather than another sponsor, the operating model that gets you there is not the one most platforms were designed to produce — and which buyer you’re building for is being decided right now in choices nobody is calling exit decisions.

The Resource-Based View (RBV) of the firm offers the most rigorous available answer. It conceptualises acquisitions not as market-entry moves or scale plays but as the deliberate accumulation of heterogeneous, imperfectly mobile resource bundles under uncertainty. That framing fits buy-and-build unusually well, because in a platform assembled through a sequence of deals, advantage depends far less on any single transaction than on the construction of a resource base over time.

This note positions RBV as the primary analytical lens for value creation in buy-and-build. It also takes the lens seriously enough to test it: to ask where its assumptions hold, where the evidence is thinner than its advocates suggest, and where its sharpest critics are right. Dynamic Capabilities enters later as a disciplined extension rather than a competing framework, and the note’s central claims are stated as explicit propositions at the end. A lens worth using is a lens worth holding to account.

Origins of the Resource-Based View

The intellectual roots of RBV trace to Edith Penrose (1959), who argued that a firm is best understood as a collection of productive resources whose deployment both constrains and enables growth. Growth, in her account, is bounded not by the market but by the firm’s own capacity to absorb and direct new resources, the limit later writers called the “Penrose effect.” That idea is the seed of everything this note argues: a platform can only acquire as fast as it can absorb.

Wernerfelt (1984) formalised the insight, proposing that analysing a firm through its resource positions, rather than its product-market positions, yields different and often sharper strategic conclusions. Four of his propositions bear directly on buy-and-build:

• Firms differ systematically in the resources they control.

• Those differences persist because factor markets are imperfect.

• Acquisitions are purchases of resource bundles, not just businesses.

• Strategy is the balance between exploiting existing resources and developing new ones.

Barney (1986, 1991) then specified the conditions under which resources generate sustained advantage, and later folded in the organisational dimension: the “O” of the VRIO test, meaning the firm’s capacity to actually exploit what it holds (Barney, 1995).

Core assumptions

RBV rests on two foundational assumptions (Barney, 1991). The first is resource heterogeneity: firms in the same industry control different bundles of assets, capabilities, and knowledge. The second is resource immobility: some of those resources are costly to trade, imitate, or redeploy across firms. From these follows the central proposition: sustained competitive advantage arises when a firm controls resources that are valuable, rare, imperfectly imitable, and non-substitutable (the VRIN criteria) and has the organisational capacity to deploy them.

Resources are defined broadly: tangible assets, intangible assets such as reputation and relationships, organisational processes, and embedded knowledge and routines. That breadth matters for buy-and-build, because much of what creates value in a platform sits in non-codified, tacit, or relational assets that are not easily visible in a data room. A target’s management depth, its installed-base relationships, its tacit operating know-how: these rarely appear cleanly on a quality-of-earnings schedule, yet they are often what determines whether an add-on compounds or disappoints.

Acquisitions as resource transactions, and the problem of efficient factor markets

RBV’s most underappreciated contribution is its treatment of acquisitions. Barney (1986) argued that resources are bought and sold in strategic factor markets, the markets for the inputs needed to execute a strategy, and that these markets price resources according to buyers’ expectations of the value they will create. The same target may be worth materially different amounts to different buyers; synergy is not intrinsic to the target but buyer-specific; and overpayment is common when expectations converge or optimism dominates.

This framing aligns closely with observed buy-and-build dynamics. Platforms repeatedly acquire similar firms; value depends on how well a target’s resources complement the existing base; and sequencing matters because earlier acquisitions shape the platform’s capacity to absorb the next one. Read through RBV, a successful buy-and-build assembles complementary resource positions that competitors cannot easily replicate; it is not simply buying cheap assets.

But Barney’s own argument contains the sharpest objection to the whole enterprise. If a strategic factor market is efficient, the price of a resource already impounds its expected value, and the acquirer earns only a normal return: the advantage is competed away at the auction. This is not an abstract worry for buy-and-build. Most targets are sold through organised processes rather than surfacing at random (Boone & Mulherin, 2007), and the more intermediated and competitive the process, the more fully the price should capture the value any disciplined buyer expects to extract. So how can a platform earn excess returns at all?

RBV offers three answers, and buy-and-build relies on all of them. First, value is buyer-specific because the platform can do something with the target that other bidders cannot: a target’s worth depends on the acquirer’s capacity to deploy its own complementary resources against it (Kaul & Wu, 2016). Second, the platform reduces the competition it faces, through proprietary origination and through private targets, which trade at a discount that compensates for greater information asymmetry (Capron & Shen, 2007). Third, and most importantly, the resources that matter most are not for sale in the factor market at all.

Resource accumulation and time

Dierickx and Cool (1989) supplied the refinement that makes RBV usable for buy-and-build. They challenged the assumption that strategically valuable assets can simply be purchased, arguing that the resources that matter most, such as reputation, culture, routines, and integration know-how, are stocks accumulated over time through path-dependent flows. As they put it:

“Strategic asset stocks are accumulated by choosing appropriate time paths of flows over a period of time.”

Three of their mechanisms travel directly into the platform setting. Time-compression diseconomies mean a capability built over five years cannot be bought in one; asset-mass efficiencies mean those who already hold a stock accumulate further stock more cheaply; and interconnectedness means stocks reinforce one another, so an integration capability and a reputation for being a good acquirer grow together. The implication is blunt: not all resources are tradable, acquisition alone does not confer capability, and the order and pacing of deals matter as much as selection itself.

This is why integration capability is, in a buy-and-build platform, a strategic asset in its own right. It cannot be acquired wholesale; it is built through repeated execution, and it is learned deliberately, since firms that codify and accumulate integration experience develop a capability that measurably improves later acquisitions (Zollo & Singh, 2004). The same logic governs what a platform can even recognise as a good target: its capacity to absorb new resources depends on the related knowledge it already holds (Cohen & Levinthal, 1990). Because these accumulated stocks are exactly the resources a competitor cannot buy at auction, they are also the ones the efficient-factor-market objection cannot erode.

A worked illustration: two platforms, one thesis

The mechanism is easier to see in a concrete case. The following is stylised, a composite rather than a specific company, but every move in it is drawn from the dynamics above.

Consider two private-equity platforms pursuing the same thesis: consolidate a fragmented regional market of building-products distributors and installers. Both target the same fundamentals, namely recurring renovation demand, sub-scale independents with no succession plan, and purchasing fragmentation that promises procurement synergy. On paper, their pipelines are nearly identical, and they often bid for the same assets at similar multiples.

Platform A treats its first two acquisitions as capability-building exercises rather than scale plays. It uses them to construct a repeatable template: a shared procurement function, a common branch-operating model, and a standardised onboarding sequence for acquired management. It deliberately holds cadence below what its capital could support until that template works. By the time it accelerates, each new add-on is worth more to A than to any other bidder, because A can drop it onto an operating system that already exists. The same target, valued by a financial buyer with no template, is worth less, because the synergy lives in A’s resource base, not in the target.

Platform B front-loads cadence to put capital to work and show early momentum. Each acquisition is integrated ad hoc, by whichever executive is free. Integration debt compounds: systems never converge, acquired managers leave, and the procurement synergy that justified the premiums is only partly captured. B is not buying worse companies or paying materially more; it is failing to accumulate the capability that converts a target into value. Three years in, the two platforms hold similar assets bought at similar prices and have diverged sharply in performance.

RBV explains that divergence in a way operational accounts cannot. The difference is not asset quality, price discipline, or market timing, since those were comparable. It is the stock of integration capability A accumulated and B did not, and the buyer-specific value that capability created. The advantage was built, not bought.

What the evidence actually says

The illustration is consistent with the empirical record, though that record is more qualified than RBV’s advocates sometimes imply. Capron (1999), studying horizontal acquisitions, found that the redeployment of resources between acquirer and target, in both directions, improves long-term performance, which is the redeployment-and-recombination logic at the centre of this note. Whether acquirers should prefer targets whose resources are similar to their own or complementary to them is contingent on strategic aim (Yu et al., 2016), and a target’s value depends on the acquirer’s capacity to deploy resources against it rather than on standalone quality (Kaul & Wu, 2016).

Honesty requires a caveat the earlier version of this note glossed. The broader claim that resource or strategic “fit” reliably predicts acquisition performance is not settled; the relatedness literature is genuinely mixed, with effects that vary by measure, period, and context. The defensible version is narrower, and is stated as a proposition below: fit matters when it is operationalised as the acquirer’s demonstrated capacity to deploy complementary resources, not as abstract industry similarity.

The private-equity literature adds the setting-specific evidence. Operational engineering, the industry and operating expertise a sponsor applies to its portfolio, is the capability that distinguishes leading firms (Kaplan & Strömberg, 2009), and inorganic growth through add-ons has become central to the PE business model (Hammer et al., 2017). Sponsors can earn above-average returns despite paying premiums for add-ons, through a combination of top-line growth and multiple expansion (Hammer et al., 2022), and the strategy creates value under identifiable industry and platform conditions (Bansraj & Smit, 2017). Recent theory casts private equity as a specialised intermediary in the market for corporate assets, whose selection logic differs systematically from that of strategic acquirers (Nary & Kaul, 2023), which is precisely why a resource-deployment account, rather than a generic synergy account, is the right lens.

The limits of the lens, taken seriously

RBV has real critics, and the strongest of them deserve a hearing rather than a footnote. The most damaging charge is near-tautology: if valuable resources are those that produce advantage, and advantage is the evidence that resources were valuable, the theory risks explaining outcomes by relabelling them (Priem & Butler, 2001). Reviews of the empirical literature reinforce the worry from a different angle, since intangible resources are hard to operationalise, the expected duration of advantage is left vague, and overall empirical support is modest and uneven (Armstrong & Shimizu, 2007; Newbert, 2007). And the efficient-factor-market objection, raised above, never fully goes away.

These critiques bite hardest against predictive uses of RBV, attempts to forecast which firm will win from a checklist of resources. They bite far less against explanatory use, where the goal is to understand the mechanism by which advantage is constructed across a sequence of deals. The defence is not to wave the critiques away but to accept the discipline they imply: specify the resource and the deployment mechanism in advance, rather than inferring them from the outcome. That is the point of stating propositions before observing cases, and it is the standard the rest of this Notebook tries to hold to.

RBV and Dynamic Capabilities: a disciplined extension

Dynamic Capabilities theory (Teece et al., 1997) extends RBV by emphasising a firm’s ability to sense opportunities, seize them, and reconfigure its resource base as the environment changes. Applied carefully it adds genuine explanatory power; applied loosely it becomes an all-purpose hand-wave for “the platform is good at adapting.” Even its proponents worry about this: Eisenhardt and Martin (2000) recast dynamic capabilities as specific, identifiable processes precisely to rescue the concept from tautology, while Helfat and Peteraf (2003) offered the RBV-compatible notion of capability lifecycles to describe how capabilities are founded, developed, and mature over time.

In a buy-and-build context the framework is best held narrowly: as a second-order capability that conditions the effectiveness of resource accumulation, and as the explanation for why some platforms integrate repeatedly while others stall after the first add-on. RBV explains which resources matter and why they may generate advantage; Dynamic Capabilities explains whether the firm can keep redeploying them as complexity rises. Both lenses are needed; neither alone is sufficient; and the second should not be stretched past its evidence to cover gaps in the first.

Four propositions

Stated plainly, so they can be argued with and tested against cases:

1. Buyer-specific value. In buy-and-build, a target’s value is buyer-specific and rises with the platform’s capacity to deploy complementary resources against it, not with the target’s standalone quality.

2. Accumulation, not purchase. The resources most decisive for platform advantage, such as integration capability, reputation, and operating routines, are accumulated through path-dependent execution and cannot be acquired wholesale; they are therefore not competed away in factor markets.

3. Sequencing as a constraint. The order and pacing of acquisitions shape future feasibility; early deals expand or narrow the set of targets a platform can later absorb.

4. Heterogeneity from capability. Performance differences across platforms pursuing nominally identical theses are explained primarily by differences in accumulated capability, not by differences in asset quality, price paid, or market timing.

Why this lens matters

Positioning RBV as the primary lens clarifies several things operational framings tend to obscure: acquisitions are resource bets, not growth events; value creation is buyer-specific and path-dependent; sequencing shapes future optionality; and integration capacity is itself a strategic asset (a claim Integration Capacity Is the Binding Constraintdevelops from the platform side, and Why We Acquire: Motives Before Targets takes forward into how targets are actually chosen).

Used carefully, as an explanatory lens disciplined by propositions rather than a predictive checklist, RBV makes buy-and-build legible as a systematic process of building advantage under constraint. That legibility is what the rest of the Thesis Notebook builds on: Real Options, where the lens strains and reversibility is overstated; Absorptive Capacity, where learning becomes the binding mechanism; and the pre-deal research, where the resource-bundle framing meets selection in practice. My own view, stated once and plainly, is that the resource-based account is the most honest description we have of why two platforms running the same playbook end up so far apart, and that its looseness is a fair price for getting the mechanism right.

References

Armstrong, C. E., & Shimizu, K. (2007). A review of approaches to empirical research on the resource-based view of the firm. Journal of Management, 33(6), 959–986.

Bansraj, D. S., & Smit, H. T. J. (2017). Optimal conditions for buy-and-build acquisitions [Preliminary version]. Erasmus School of Economics.

Barney, J. B. (1986). Strategic factor markets: Expectations, luck, and business strategy. Management Science, 32(10), 1231–1241.

Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99–120.

Barney, J. B. (1995). Looking inside for competitive advantage. Academy of Management Executive, 9(4), 49–61.

Boone, A. L., & Mulherin, J. H. (2007). How are firms sold?. The Journal of Finance, 62(2), 847–875.

Capron, L. (1999). The long-term performance of horizontal acquisitions. Strategic Management Journal, 20(11), 987–1018.

Capron, L., & Shen, J. C. (2007). Acquisitions of private vs. public firms: Private information, target selection, and acquirer returns. Strategic Management Journal, 28(9), 891–911.

Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128–152.

Dierickx, I., & Cool, K. (1989). Asset stock accumulation and sustainability of competitive advantage. Management Science, 35(12), 1504–1511.

Eisenhardt, K. M., & Martin, J. A. (2000). Dynamic capabilities: What are they?. Strategic Management Journal, 21(10–11), 1105–1121.

Hammer, B., Knauer, A., Pflücke, M., & Schwetzler, B. (2017). Inorganic growth strategies and the evolution of the private equity business model. Journal of Corporate Finance, 45, 31–63.

Hammer, B., Marcotty-Dehm, N., Schweizer, D., & Schwetzler, B. (2022). Pricing and value creation in private equity-backed buy-and-build strategies. Journal of Corporate Finance, 77, 102285.

Helfat, C. E., & Peteraf, M. A. (2003). The dynamic resource-based view: Capability lifecycles. Strategic Management Journal, 24(10), 997–1010.

Kaplan, S. N., & Strömberg, P. (2009). Leveraged buyouts and private equity. Journal of Economic Perspectives, 23(1), 121–146.

Kaul, A., & Wu, B. (2016). A capabilities-based perspective on target selection in acquisitions. Strategic Management Journal, 37(7), 1220–1239.

Nary, P., & Kaul, A. (2023). Private equity as an intermediary in the market for corporate assets. Academy of Management Review, 48(4), 719–748.

Newbert, S. L. (2007). Empirical research on the resource-based view of the firm: An assessment and suggestions for future research. Strategic Management Journal, 28(2), 121–146.

Penrose, E. T. (1959). The theory of the growth of the firm. Oxford University Press.

Priem, R. L., & Butler, J. E. (2001). Is the resource-based “view” a useful perspective for strategic management research?. Academy of Management Review, 26(1), 22–40.

Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509–533.

Wernerfelt, B. (1984). A resource-based view of the firm. Strategic Management Journal, 5(2), 171–180.

Yu, Y., Umashankar, N., & Rao, V. R. (2016). Choosing the right target: Relative preferences for resource similarity and complementarity in acquisition choice. Strategic Management Journal, 37(8), 1808–1825.

Zollo, M., & Singh, H. (2004).Deliberate learning in corporate acquisitions: Post-acquisition strategies and integration capability in U.S. bank mergers. Strategic Management Journal, 25(13), 1233–1256.

The macro pressure traced in Dry Powder and the Pressure to Move — capital aging in a system that cannot recycle it — lands at the deal in this form. Higher entry multiples reduce the margin for error. Longer holds extend the time over which that error must be carried. What used to be absorbed across the life of the deal is now exposed earlier, and the slack a slow first year used to enjoy has narrowed in ways that don’t always announce themselves at the time.

The release valve no longer functions

For much of the prior cycle, multiple expansion functioned as a release valve. It did not eliminate execution risk, but it softened the consequences. A platform that underperformed early could still exit into a stronger pricing environment. A delayed integration could still benefit from valuation lift on the way out. Together those mechanisms created a structural asymmetry — early underperformance was often survivable, and late recovery remained possible — and most operating playbooks of the prior decade quietly assumed that asymmetry would hold.

That asymmetry has weakened. When entry multiples are already elevated and the cost of capital has moved against further expansion, the back end of the deal cannot be counted on to do the work the middle once did. What remains is operational performance — and the timing of when it becomes visible. The shift sounds incremental from the outside; from inside the platform, it changes where the burden of proof sits and when it has to be discharged.

Underwriting has moved to the floor

This shift is now explicit in how deals are described. In Grant Thornton’s sponsor commentary inside PitchBook’s 2025 Annual US PE Middle Market Report, sponsors describe a change of emphasis that maps directly onto the math: underwriting is anchored less on upside cases and more on downside resilience, and increasingly on whether a credible floor can be established early. The same commentary is direct on what this requires of the early hold: “Buyers are highly focused on getting the first two years right.”

This is not a change in ambition; it is a change in what must be proven, and when. Growth theses still matter. They simply no longer carry the deal alone. The floor — what the asset reliably produces under conditions that are not generous — has become the new underwriting frontier, and it has to be visible early enough to recalibrate the rest of the hold around it.

What the first 24 months now carry

The early period after close has always mattered. What has changed is how much of the outcome now depends on it. In the first 24 months, integration assumptions are tested under real conditions, leadership capacity is either sufficient or it isn’t, and sequencing choices begin to compound rather than reset. None of these dynamics is new. What is new is that they are no longer deferrable. The system reveals them earlier because the structure no longer absorbs delay, and the platforms that miss the reveal don’t get a quieter second year to make up the ground.

Cadence tightens — not as a choice, but as a consequence. Proof points are required sooner, not because investors demand them but because the underlying math demands them. What used to emerge gradually now separates quickly, and the separation is rarely loud enough to read as failure when it begins.

The compressed feedback loop

In prior environments, feedback was delayed. Performance issues could take years to surface fully. Integration strain could remain partially hidden inside growth numbers. Narratives could outrun operating reality for extended periods, and a platform’s external story often led its internal one by a year or more. That delay created room for interpretation, and most of the operating playbooks built in that period implicitly relied on it.

Today the feedback loop is shorter. Not because organizations have become more disciplined — but because the system is less forgiving. A weak first year is harder to offset. A mis-sequenced integration compounds earlier. A leadership gap becomes visible under load, not over time. The distinction between a working deal and a strained one forms sooner, and the early signs of strain — the ones that used to be debatable for a year — now resolve into evidence within months.

Early hold versus extended hold

Holding periods have lengthened, but not all extensions mean the same thing. Some reflect timing — waiting for markets to reopen, or for conditions to stabilize before exit. Others reflect structure — a deal that has not established a stable operating base early enough to support exit at all. The two look identical at the aggregate level, and they are routinely conflated in fund-level reporting.

Inside the first 24 months, the difference becomes clear. Whether integration has stabilized or continues to consume capacity, whether performance is compounding or being managed, whether optionality is expanding or narrowing — these distinctions resolve in the early window, not the late one. By the time a deal reaches year five or six, the path is usually set, and what looks like an extended hold for upside is often an extended hold for repair.

Playbooks from a different market

Many operating models in use today were shaped in a different environment. The 2014–2020 vintages operated with lower entry multiples, more reliable expansion, and shorter expected holds, and in that context a slow first year was not ideal but was often recoverable. That experience persists in how platforms are run. Integration pacing, performance expectations, and sequencing decisions still reflect a system that allowed time to compensate for early strain — and the operators who internalized those rhythms most successfully are sometimes the ones least equipped to read the new ones. The environment has changed faster than the playbooks built inside it.

Where the difference shows up

The separation between platforms is not always visible at exit. It shows up earlier, and most of the time it shows up in the same handful of places. The platforms I’ve watched closely tend to declare themselves by month 24 — whether integration has reduced complexity or merely redistributed it, whether leadership is deciding under load or reacting to it, whether the base business is stable or quietly carrying strain that hasn’t yet been named. From the outside, both paths can still look functional well past that point. From the inside, they are already diverging.

The shift in where risk lives

The change is not that deals have become riskier. It is that risk has moved. The risks that mattered most in the prior cycle — exit timing, market conditions, valuation expansion — have been displaced by risks that resolve earlier and inside the platform: early integration, sequencing under constraint, leadership capacity under load. This does not make outcomes more predictable. It makes them visible sooner, and that visibility is the change operators need to design around.

Most operating models I’ve seen carry an implicit assumption that the first year is a setup period — that the real work begins once integration stabilizes and leadership has settled in. That assumption was reasonable in the market the playbooks were built in. It is not reasonable now. The first 24 months are no longer the runway; they are most of the deal.

The question for 2026

The first 24 months do not determine the outcome by themselves. But they now determine whether the outcome remains open. What used to be resolved over the life of the deal is now decided earlier — not by design, but by the structure of the system itself. Multiple expansion was the back-end mechanism that gave operating teams time to learn. With that mechanism gone, learning has to compound forward from close, and the platforms that come out of this period strongest will be the ones whose first 24 months produced visible operating evidence rather than visible operating activity. 2026 is the year that distinction begins to surface in fund-level performance, and the data PitchBook and Bain are publishing this cycle will look obvious in retrospect — once the cohort that absorbed the shift early is visible, and the cohort that didn’t is no longer presenting at LP conferences.

Cadence is the organisation’s temporal architecture

Operating cadence often gets reduced to meetings: weekly check-ins, monthly reviews, quarterly updates. That picture undersells it. Cadence is the organisation’s temporal architecture. It shapes how quickly issues surface, how long ambiguity persists, how often leaders are interrupted, and how pressure accumulates across the system. Cadence determines whether leadership attention is focused or fragmented — long before anyone feels “busy.”

The empirical finding behind this isn’t new. Mintzberg’s classic study of managerial work (Mintzberg 1973) documented that senior executives’ activity is fragmented, varied, and brief — an average task duration of about nine minutes for the executives he observed. The fragmentation was mostly produced by the cadence of demand on their attention, not by their own choices about how to work. That is the dynamic this piece is about, applied to the specific structure of buy-and-build platforms.

It is also why two platforms with similar strategies and similar talent can feel radically different to lead. The platforms I’ve watched closely run on visibly different rhythms, and the difference is rarely captured in any document.

Why cadence breaks after acquisitions

Integrations disrupt cadence before they disrupt performance. New reporting cycles get introduced, legacy rhythms collide, exceptions multiply, and informal coordination gives way to formal review. What once flowed predictably now arrives unevenly. Leaders feel this as a constant low-level interruption — issues surfacing out of sequence, decisions demanded without context, escalations arriving earlier or later than expected. None of it looks dramatic. The texture of leadership work changes anyway, and leaders end up reacting sideways more than leading forward. (This is part of what the first 30–90 days is actually about — cadence breakage is one of the things stabilisation is doing.)

Cadence governs attention, not just information

Most operating models assume better information improves decisions. Cadence determines whether leaders can use the information at all. When cadence is stable, signals arrive with enough frequency to be meaningful, leaders can distinguish noise from trend, and decisions are spaced far enough apart for learning to happen. When cadence degrades, everything feels urgent, weak signals get treated as strong ones, and leaders are asked to decide before patterns have had time to emerge.

That’s how leadership quality erodes without anyone making worse decisions individually. The system is asking for judgment faster than judgment can mature.

Overlapping cadences create hidden load

Buy-and-build platforms often carry several cadences simultaneously: the legacy operating rhythms of the acquired businesses, new platform-level reporting cycles, deal-driven integration milestones, and investor or board review timelines. Each cadence makes sense on its own. Together, they overlap, and the overlap is costly — not because of the meetings it creates, but because of the context-switching it imposes on the people running the platform. Leaders end up operating across time horizons simultaneously: immediate issues, short-term integration tasks, long-term platform development. This is how leadership effort increases even when headcount doesn’t.

How cadence consumes leadership capacity

Leadership capacity is finite. Cadence determines how fast that finite capacity gets consumed. Fast, irregular cadence burns capacity quickly; slow, coherent cadence allows recovery and learning. That doesn’t make slow organisations better — it makes intentional cadence a precondition for leadership effectiveness, where accidental cadence quietly depletes it.

Most leadership strain attributed to “too much going on” is a cadence problem in disguise.

Why adding structure often makes it worse

When cadence strain becomes visible, organisations usually respond by adding structure: more meetings, more reviews, more dashboards. These interventions increase visibility, but they also increase cadence density. Issues surface more frequently, decisions get requested sooner, and leaders feel more informed and more overloaded at the same time.

That’s why some platforms feel heavier after “professionalisation” rather than lighter. The system has become more active without becoming more selective. Cadence rose, judgment didn’t.

Cadence as a design choice

Treating cadence as a leadership system changes the question that has to be answered. The question stops being “do we have the right meetings?” and starts being “have we designed a rhythm that lets leadership capacity compound rather than fragment?” That requires accepting trade-offs most operating models avoid: not every issue should surface immediately, not every decision should be synchronised, and not every integration milestone deserves leadership attention.

Cadence forces prioritisation — of issues, but more importantly of time.

Why this matters in buy-and-build

Buy-and-build strategies introduce complexity in discrete jumps. Cadence determines whether those jumps get absorbed or amplified. When cadence respects leadership capacity, learning consolidates between deals, decision quality stabilises, and leadership effort compounds into organisational capability. When cadence ignores capacity, pressure accumulates invisibly, leaders become reactive, and the platform feels brittle long before results decline.

Cadence is a leadership decision with strategic consequences, not an execution detail.

Closing the leadership and operating loop

Three ideas now sit together across this section:

• Leadership capacity is finite.

• Decision rights determine how that capacity is used.

• Operating cadence determines how fast it is consumed.

Together, these three ideas explain why buy-and-build platforms strain in familiar ways even with strong leaders, aligned teams, and well-executed deals. Leadership in buy-and-build is the system that governs attention, authority, and time.

Integration & Execution, the next section, examines what happens when that system meets reality.

Read together, those three numbers describe a system trying to recycle capital in two directions at once: pushing out new deployment from aging dry powder and pulling in liquidity from holdings that won’t move. None of the individual numbers is unusual in isolation. The combination is structural, and it lands on sequencing more directly than most fund-level commentary acknowledges.

The external clock

Every fund contains an implicit clock. Limited partners expect deployment within defined windows. Value creation has to be demonstrated within defined hold periods. Capital that sits idle erodes credibility; capital that sits unrealised constrains the next raise. The external cadence is rational — it disciplines capital allocation and prevents indefinite deferral of judgment.

Buy-and-build platforms run on more than LP timelines. They run on integration capacity, leadership bandwidth, and absorption limits that don’t automatically expand because dry powder did. The tension this piece is about is what happens when those clocks diverge — particularly now, when the LP-side clock is running faster than it has in fifteen years and the operator-side clock has not changed at all.

Sequencing under capital pressure

In a buy-and-build environment, sequencing is the system’s coordinating mechanism. It governs when complexity is introduced, how learning compounds, and whether integration effort declines over time. Sequencing converts strategy into trajectory, and capital pressure alters sequencing incentives in directions that aren’t obvious until afterward.

When dry powder ages, the cost of waiting rises. Deals that might have been deferred become urgent, platforms that are “nearly ready” get treated as ready, and targets that stretch distance slightly more than ideal become tolerable. Add-on selection criteria get applied more loosely than platform-selection criteria warrant — a conflation pattern the data picks up at the platform level only after the fact.

None of these shifts looks dramatic in isolation. But sequencing is sensitive to marginal changes, and introducing complexity earlier than capacity supports doesn’t fail immediately — it accumulates. Leadership attention fragments slightly, integration timelines extend slightly, exception handling rises slightly. From the outside, deal flow continues. From the inside, cadence shifts.

Guardrails and investor pressure

Every disciplined buy-and-build system needs guardrails. Guardrails aren’t rigid prohibitions; they’re boundaries that prevent capital availability from dictating pace. When investor pressure rises — and it has, with vintage 2022–23 dry powder now visibly aging into the back half of its deployment window — those guardrails get tested.

The question isn’t whether firms feel pressure to deploy. They do; Bain’s data is unambiguous on that. The question is whether the system can say no, or not yet. Saying no is costly: under-deployment risk, fundraising friction, reputational drag in competitive markets. Saying yes prematurely is also costly — it pulls forward complexity without expanding absorption capacity, and the cost surfaces months later in the next acquisition’s integration.

The interaction between guardrails and investor pressure is structural, not behavioural. It determines whether sequencing remains coordinated or becomes capital-driven. Most of the platforms I’ve watched make this choice quietly, without anyone naming it as a choice.

Capacity does not scale with dry powder

A common implicit assumption in strong fundraising cycles is that scale expands capability. Larger funds, larger platforms, larger teams. But leadership and operating capacity don’t scale linearly with capital under management. Hiring increases headcount; it doesn’t automatically increase shared judgment, integration fluency, or decision coherence. Integration discipline gets learned through repetition under manageable load — and when load increases too quickly, learning slows rather than compounds.

Capacity is built deliberately, and dry powder doesn’t build it. If anything, abundant capital can mask capacity constraints temporarily. Platforms keep acquiring, operating partners add initiatives, boards increase meeting cadence. The activity feels like momentum until absorption limits surface — usually one or two acquisitions later, and usually after the constraint has already become harder to undo.

The illusion of optionality

High-dry-powder environments are often described as offering optionality. More capital means more potential moves. Structurally, optionality only exists if the system retains flexibility — and when sequencing accelerates under pressure, optionality can narrow rather than expand.

Each acquisition introduces commitments that are hard to walk back: cultural, operational, financial. Integration choices harden architecture. Technology stacks consolidate. Leadership roles crystallise. Optionality increases when each move improves the next; it decreases when each move consumes more capacity than it builds. Capital availability doesn’t determine which path the system takes — sequencing discipline does.

Where this leaves operators in 2026

Two-thirds of LPs in the ILPA Sentiment Survey 2025–26 said they would accept extended holds for better MOIC, which gives some platforms permission to slow down. But that permission is conditional — it depends on the platform actually compounding MOIC during the extended hold. The platforms I’ve watched at the edge of their capacity usually arrive at the point of needing more time only after they’ve already used the time they had. For a platform already at the edge of its absorption capacity, extending the hold is the right call only if the additional time is being spent building capability, not buying more time to absorb the last acquisition.

Capital pressure isn’t uniformly distorting. External clocks can counteract inertia, prevent over-analysis, and force decision clarity. In platforms with under-utilised capacity, deployment pressure may align with readiness. In platforms that have already compounded learning and reduced integration load over time, increased pace may be sustainable. The question — and the place this piece comes from — is which kind of platform a given operator is actually running, and whether that read is honest.

Accumulation, not event

One quiet feature of capital-driven sequencing is that failure rarely shows up as a single event. There’s no obvious breaking point. Integration stretches slightly, leadership load stays elevated slightly longer, cadence between acquisitions shortens slightly, and operating routines revert to exceptions slightly more often. Each increment is manageable. The system appears intact. But trajectory shifts gradually from compounding to accumulation, and the distinction becomes visible only in hindsight — when optionality has narrowed and sequencing flexibility has diminished.

The question for buy-and-build in 2026

Dry powder will fluctuate. Investor timelines will exist. Capital will periodically outpace opportunity, and opportunity will periodically outpace capital. The question for buy-and-build platforms is whether the system’s coordinating logic remains intact under pressure — not how to eliminate the pressure.

Three operator-level questions map cleanly onto the Bain data:

• Does sequencing still govern when complexity gets introduced, or has the deployment clock taken that role over?

• Does integration discipline still reduce the load on the next acquisition, or are the last two integrations still bleeding capacity into the current one?

• Does leadership capacity expand faster than commitments, or has the platform crossed the line where every additional yes consumes more capacity than the previous one built?

If those three mechanisms hold, the capital pressure becomes useful — it sharpens decisions that might have drifted otherwise. If they degrade, the same pressure becomes distortion. The external clock can’t be removed. Whether it dictates trajectory is something each platform decides for itself, usually one acquisition at a time.

Dry powder is a visible metric. Capacity is not. The tension between them is structural and recurring, and 2026 is the year that tension is sharpest in modern PE history. The platforms that come out of this period strongest will be the ones that took the question seriously while the data still left them room to.

Platform Selection and Add-On Selection Are Different Decisions covers the conceptual distinction: platform selection is an industry-and-resource decision, add-on selection is an interaction-with-platform decision, and the criteria that should dominate each are nearly opposite. This piece is about how that distinction shows up — or fails to — in the actual room where the calls get made.

What changes between the platform call and the first add-on call

The platform call is, in practice, made by deal-team thinking. Industry analysis dominates, resource fit gets attention, and diligence centres on the target as a stand-alone asset — because that’s effectively what it is. There’s no platform yet for it to interact with. Management quality, customer concentration, contract risk, market position: standard PE diligence, applied well.

The first add-on call is the first time that framework has to start including the platform’s current state as a primary input. In most rooms I’ve watched, that input doesn’t enter the analysis cleanly. The deal team frames the add-on the way they’d frame any small acquisition in this space. The integration team frames it through their current load. The two views often don’t reconcile, and the reconciliation happens implicitly — usually by deferring to the deal team, because the deal team has the rhythm and the deck.

The result is a yes that gets made on platform-style criteria when the decision actually being made is an add-on decision. That asymmetry compounds quietly across the next several deals.

The signals operators read differently

There’s a small set of signals that should weigh more heavily on add-on decisions than on platform decisions, and most of them are rarely on the standard diligence checklist:

• Where the platform’s leadership bandwidth currently is. Not an org chart question — a Thursday-afternoon question.

• What the integration team is actually carrying right now, and from which prior deal.

• Whether the platform’s operating cadence has stabilised after the last absorption, or is still in transition.

• How much “cushion” exists in the calendar of the people who’ll have to actually run the new integration.

• Whether the add-on creates capacity for the next add-on or consumes it.

None of these show up well in a deal memo. They show up in conversations with operating partners, in calendar audits, and in honest answers to “how’s the last one going?” The operators I know who are good at add-on selection ask different questions than the deal team, and they ask them earlier in the process, not at the end.

The temporal mismatch

Platform decisions and add-on decisions also operate on different time horizons, and the framework rarely makes that explicit.

A platform decision is effectively a 7- to 12-year decision. The thesis has to hold against an exit window most of the way out. Industry structure, competitive moat, the trajectory of multiples — these matter because they have to compound over the full hold period.

An add-on decision is more like a 2- to 3-year decision. The relevant horizon is the period across which the platform has to absorb the acquisition and stabilise enough to either consider another one or run more cleanly into exit. The interaction effects show up over months, not years. Applying platform-decision discount rates to add-on decisions makes everything look better than it should. The same valuation that’s reasonable for a stand-alone industry bet becomes aggressive once the platform’s actual integration capacity is priced in.

The first add-on test

The first add-on a platform completes is usually the one that reveals whether the platform decision was right.

Not because the add-on itself is the test — the add-on is just a normal acquisition. The test is what the platform’s response to integrating the first add-on shows about the bet that was made on the platform itself.

If the integration goes cleanly, the platform team gets confident, integration capacity feels abundant, and the next deal usually arrives on the calendar early. If it goes poorly, the platform team learns something the deal team didn’t: the resource bundle wasn’t quite what the thesis assumed, the operating model has more friction than the diligence captured, and the sequencing has to slow down.

I’ve watched both outcomes. The platforms that got the first add-on wrong rarely admit it as a platform-decision error — they treat it as an integration execution problem and try to fix it through more process. That’s almost always wrong. The error usually sits at the interaction between the platform’s actual capacity and the add-on selection criteria, and the fix is to recalibrate the criteria, not to add governance.

The question to put on the table

Before any add-on closes — particularly the first one — the question worth asking out loud is narrower than the deal memo usually frames it: given the state this specific platform is actually in, with these specific people carrying this specific load, is this the add-on we should be doing right now, or is it the add-on the deal team would be doing if they had a clean platform to drop it into?

If the answer is “yes, this one, now,” proceed. If the answer is “well, the platform is still finishing the last thing, but the deal team has been working on this one for six months,” the framework is doing the work the platform itself should be doing — and the framework is wrong for the job.

That recalibration is one of the cleaner ways to tell whether the deal team and the operating team are reading the same platform.

The error sits one layer earlier than the framework. Platform selection and add-on selection are different decisions, governed by different logics, and the criteria that should dominate each are nearly opposite to one another in places that matter.

What platform selection is actually deciding

Platform selection is, at root, an industry decision. It is a bet on the structural attractiveness of a sector and on the long-term economics of operating inside it. This can’t be emphasised enough. The question being answered is whether this is an industry where a focused, well-capitalised platform can compound advantage over a 7- to 12-year horizon, against the realistic competitive set.

That question has a long pedigree in strategy. Porter’s Five Forces (Porter 1980) is the canonical framework for thinking through it: competitive intensity, threat of new entrants, buyer power, supplier power, and substitution dynamics. The resource-based view (Wernerfelt 1984, Barney 1991) adds the resource lens — what does the platform need to control to make the industry economics actually work for it? Picking a platform means picking an industry-position-resource bundle that has to hold up under conditions you can’t yet describe.

Other inputs at the platform stage: the availability of attractive add-ons in the sector (a platform with no plausible add-on path is a single-asset bet, not a buy-and-build); the structural moat or scale advantage the platform can build over time; and the exit landscape 7–12 years out — who buys these in size, at what multiples, under what conditions.

Platform decisions are slow, expensive to reverse, and rare. A typical PE platform thesis gets built over months and lived with for the better part of a decade.

What add-on selection is actually deciding

Add-on selection is a fundamentally different question. It is a bet on the interaction between this specific add-on and the specific platform that already exists, given the integration capacity that platform currently has and the sequencing of what’s already been absorbed.

The question being answered isn’t “is this a good business?” — that question is mostly settled by the time an add-on reaches diligence. The question is whether, given everything the platform is already carrying, it can absorb this one well, and whether absorbing it improves the platform’s position for the next add-on.

The criteria that matter at this layer are nearly the inverse of the platform layer: operational overlap with the platform (how much capacity will integration consume?), integration burden relative to current load (what’s already in flight?), leadership bandwidth in the platform team (who has to actually run the absorption?), sequencing (what does this one make easier or harder for the next?), and cultural and process compatibility — not just industry fit. Add-on decisions are governed more by integration capacity than by industry structure, and they are faster, more frequent, and partially reversible (add-ons can be divested cleanly more often than platforms can).

Where the framework misfires

The conflation usually goes one direction. Deal teams that selected a platform on industry-and-resource grounds keep using industry-and-resource framing for the add-ons. The add-on diligence centres on customer concentration, contract risk, financial performance, and management quality. Those things matter — they are necessary conditions, but they are not sufficient ones.

What the framework misses, when it’s the wrong framework, is the question that actually drives outcomes at the add-on layer: how much of the platform’s absorptive capacity will this acquisition consume, and how much will it leave in reserve for the one after this? The original acquisition motive tells you why this kind of add-on is on the list at all; it doesn’t tell you whether this one is the right next one for the platform.

That question doesn’t appear in a Porter-style analysis because Porter is about the industry, not about the platform. It appears only when add-on diligence is built around the buying platform’s current state, not just the target’s stand-alone characteristics.

Why this matters

Both decisions are real, and both deserve rigorous diligence — but they need different rigour. Platform selection needs to ask whether the industry and the resource bundle can hold up over a long horizon. Add-on selection needs to ask whether the platform that already exists can absorb this one well now.

Conflating them produces a recognisable failure pattern: the deal team approves an add-on on platform-criteria grounds; the integration team is left with a target that fits the industry but not the platform’s current state; and twelve to eighteen months later the post-mortem identifies “integration challenges” — when the real problem was the criteria the add-on was approved against in the first place.

How Platform Calls and Add-On Calls Get Made Differently takes up the operator side of this — how the call gets made differently in the room, and what the first add-on usually reveals about whether the platform decision was right.