Ask how a buy-and-build platform chose its add-ons and you will usually get a tidy story: a thesis was set, targets were screened against it, and the best ones were bought. The tidy story is rarely true. In most platforms the thesis is not fixed at the outset and then executed; it is built, tested, and revised as the platform learns what actually works and what actually comes up for sale.
There are two ways buy-and-build is run, and the distinction matters. In the first, a predefined roll-up thesis drives everything: a sponsor commits to consolidating a fragmented sector and then selects a platform and add-ons to execute that plan. In the second, the sponsor acquires a promising platform first and constructs the add-on thesis afterwards, refining it as the hold period unfolds. The platforms worth studying are mostly the second kind, and it is in that mode that the gap between plan and reality is widest. This note uses process theory to describe how the thesis really forms.
Deliberate and emergent strategy
The cleanest language for this comes from Mintzberg and Waters (1985), who distinguished deliberate strategy, the part that is intended and realised as planned, from emergent strategy, the part that forms through a stream of decisions in the absence of, or despite, intention. Real strategies, they argued, are neither purely one nor the other but lie on a continuum, and the most useful mode for an uncertain environment is what they called an umbrella strategy: leadership sets broad boundaries and lets the specifics emerge within them. A platform-first buy-and-build is an umbrella strategy almost by definition. The sponsor sets a direction, consolidate around this platform in this sector, and the realised sequence of add-ons emerges from the opportunities that actually arrive.
Why this needs process theory
Describing how a thesis forms over time requires tools built for sequences, not snapshots. Process research treats organisational phenomena as sequences of events whose order and timing carry explanatory weight, rather than as variables measured at a point (Langley, 1999; Van de Ven & Poole, 1995). It also insists on moving from narrative to mechanism: a chronology of what happened is only the surface, and explanation requires the generative structure beneath it (Pentland, 1999). This is exactly what the target-selection literature lacks. The pre-deal phase has been modelled as a discrete, static event rather than an unfolding process (Welch et al., 2020), which is precisely the framing a process account of buy-and-build has to replace.
The acquisition-programme perspective supplies the link to performance. The right unit of analysis is the programme, not the single deal, and both the rate and the variability of acquisitions across the sequence bear on how it performs (Laamanen & Keil, 2008). Selection criteria, on this view, are not fixed inputs; they are outputs of a process that updates with each deal.
A worked illustration: the thesis that bent
Take a platform that begins with a clear deliberate thesis: consolidate independent installation contractors in a regional building-products market, betting that scale in labour and scheduling will lift margins. The first two add-ons are chosen to fit that plan. They also teach the platform something its plan did not contain: the real margin pressure, and the real source of customer loyalty, sits upstream in distribution, not in installation labour.
A purely deliberate platform ignores this and keeps buying installers. A purely opportunistic one chases whatever is for sale. The platform that does well does neither: it keeps the umbrella, consolidation in this region, but bends the thesis toward owning distribution, and its next add-ons are distributors rather than installers. The realised strategy is part deliberate, the regional consolidation held throughout, and part emergent, the shift to distribution learned from the first deals. Neither the original plan nor pure opportunism would have produced it.
The objection: isn’t this just undisciplined drift?
The sharp objection is that emergent strategy is a flattering name for a lack of discipline, a way to dignify chasing deals and rationalise it after the fact. The risk is real, and it is the same hazard the target-selection note raised: a flexible thesis can excuse buying whatever appears. But the deliberate-emergent distinction is not a licence to drift; it is a description of where the discipline has to sit. In an umbrella strategy the boundaries are firm even though the specifics are open, and the test of a well-run emergent process is whether the thesis updates through structured learning, post-deal reviews that change the criteria, rather than through opportunism that merely reacts. Mintzberg and Waters (1985) themselves separate emergent strategy, which is coherent even though unplanned, from what they call unconnected strategy, which is not. The discipline of a platform-first roll-up is not a fixed plan; it is a governed process for revising the plan, and a platform that cannot tell the difference between learning and drifting will not survive either.
Four propositions
Stated plainly, so they can be argued with and tested against cases:
Thesis as output, not input. In platform-first buy-and-build, the add-on thesis is built and revised across the hold period; selection criteria are outputs of a process, not fixed inputs.
Realised strategy is mixed. The realised sequence of add-ons is part deliberate and part emergent, and treating it as either pure plan or pure opportunism misdescribes it.
Process carries the explanation. The order and timing of early deals shape which later deals are seen and chosen, so selection must be studied as a sequence, not a discrete event.
Discipline is in the updating. Emergent strategy is disciplined when the thesis updates through structured learning within firm boundaries, and undisciplined when it merely reacts to what is for sale.
Why this matters
Seeing the add-on thesis as a deliberate-emergent process explains something the tidy story cannot: why two sponsors who buy the same platform end up with different companies. The difference is not the plan they started with but the process by which they revised it. This is the connective tissue of the Notebook. The resource-based view says advantage accumulates across the sequence; the target-selection note says each choice embeds the next; and process theory says the criteria themselves evolve. For the dissertation this is the central claim, that target selection in buy-and-build is a path-dependent, recursive process rather than a series of independent fit judgments, and the rest of the empirical work exists to trace how that process actually runs.
References
Langley, A. (1999). Strategies for theorizing from process data. Academy of Management Review, 24(4), 691–710.
Laamanen, T., & Keil, T. (2008). Performance of serial acquirers: Toward an acquisition program perspective. Strategic Management Journal, 29(6), 663–672.
Mintzberg, H., & Waters, J. A. (1985). Of strategies, deliberate and emergent. Strategic Management Journal, 6(3), 257–272.
Pentland, B. T. (1999). Building process theory with narrative: From description to explanation. Academy of Management Review, 24(4), 711–724.
Van de Ven, A. H., & Poole, M. S. (1995). Explaining development and change in organizations. Academy of Management Review, 20(3), 510–540.
Welch, X., Pavićević, S., Keil, T., & Laamanen, T. (2020).The pre-deal phase of mergers and acquisitions: A review and research agenda. Journal of Management, 46(6), 843–878.