Skip to content
← Resources / Articles
Integration5 min read

The first hundred days decide whether the model was ever real

Synergies are a modelling assumption until somebody with a name and a calendar owns them. Notes from the people who inherit the deal after the champagne.

All M&A Research

Post-merger integration

Updated
Several tower cranes standing over a construction site.
Several tower cranes standing over a construction site.

Key takeaways

  • Synergies are a modelling assumption until a named person with a calendar owns each one. Unowned synergies do not arrive.
  • The integration model has to be chosen and stated out loud. Leaving it ambiguous means every team picks a different one.
  • The people question is the whole question. The staff who make the business work decide within weeks whether to stay.
  • Acquirers consistently underestimate their own capacity to absorb a business while continuing to run themselves.

Deal teams are measured on getting to signing. Integration teams are measured on everything after. The handover between them is the least designed moment in the entire transaction, and it is where a good deal most often becomes an average one.

The pattern integration leads describe is remarkably consistent: they are brought in late, given a synergy number they did not build, handed a business whose management has just spent six months in a data room and is exhausted, and asked to deliver an operating plan by the end of the quarter. What follows is a set of observations from people who do this repeatedly.

Synergies are assumptions until they have an owner

A synergy in a model is a line. A synergy in reality is a named person, with a date, a budget line it comes out of, and a specific decision they are authorised to make. The single most reliable predictor of whether a synergy is delivered is whether that conversion happened in the first month.

The conversion is uncomfortable, because it surfaces which synergies were never real. Procurement savings that assume you can consolidate suppliers you are contractually locked into for three years. Headcount reduction in a function that turns out to be the only thing holding the customer relationship together. Cross-sell that requires a sales team to learn a product with a nine-month sales cycle. Facility consolidation that requires a lease break you do not have.

Doing this early is expensive politically and cheap financially. Doing it in month nine, when the board asks why the run-rate is behind, is the opposite.

Every integration lead has the same request of every deal team: tell us which synergies you actually believe, and which ones were in the model to make the return work. We will deliver the first list. We will spend nine months failing to deliver the second one.

The people question is the whole question

You did not buy a legal entity. In most mid-market transactions you bought a customer base held together by relationships and an operating capability held in the heads of perhaps a dozen people. Almost all destroyed value traces back to those two facts.

Identify the actually-critical people, not the senior ones. They are frequently not on the org chart’s top two rows. The operations manager who knows how the scheduling really works. The one engineer who understands the legacy integration. The account manager three customers will follow.

Talk to them in week one, not month three. Retention packages issued after somebody has taken a recruiter’s call are twice as expensive and half as effective.

Answer the three questions everyone in an acquired company is asking. Do I have a job? Who is my manager? Does what I was working on still matter? Uncertainty on any of these produces attrition, and the ambiguity costs more than any answer would, including an honest “this function is under review and you will know by the fifteenth.”

Expect the founder’s departure to be earlier than agreed. Founders who have sold and stayed describe a fairly predictable arc: a period of goodwill, a first collision with the new owner’s process, and a recalculation. Plan for the capability transfer rather than the retention clause.

Choose an integration model and say it out loud

Most integrations underperform because nobody decided how integrated the business would be. The three defensible models each require different behaviour:

Absorption. The acquired business ceases to exist as a distinct unit. Systems, brand, processes and org all consolidate. Fastest route to cost synergy, highest risk to the acquired business’s distinctiveness, and completely incompatible with an earn-out.

Preservation. The business runs largely as-is, with financial reporting and a handful of group standards. Right for businesses bought for growth, for a market position or for a team, and for anything where the acquired culture is part of the asset. Delivers little cost synergy, so the model had better not have assumed any.

Symbiosis. Selective integration: shared back office, separate go-to-market, or vice versa. The most common choice and the hardest to execute, because every boundary must be explicitly decided rather than assumed.

The failure mode is drift: starting with preservation because it feels kind, then absorbing function by function without ever telling anyone that the model changed.

The first hundred days, sequenced

What integration leads consistently put in each phase:

Before close

Clean rooms where competitively sensitive data can be analysed lawfully. Day-one readiness across payroll, IT access, customer communications and supplier notifications. A named integration lead with a budget. Leadership decisions made. The worst outcome is a leadership vacuum in week one. A communications plan drafted for employees, customers, suppliers and, if relevant, regulators.

Days 1–30

Leadership announced and in place. Every employee has heard directly from a named person about their situation. Top twenty customers contacted by someone senior. Reporting cadence established. The synergy list converted to owners and dates. A stop-doing list. The projects the acquired business will pause so that the integration has capacity, because it does not have spare capacity.

Days 31–70

Operating plan agreed and resourced. Systems roadmap with sequencing that does not put two critical cutovers in the same month. Org design decisions made and communicated, in one wave rather than five. Early wins delivered and visible: something the acquired team asked for and now has.

Days 71–100

Synergy tracking against baseline, with variances explained rather than averaged. Retention checkpoint on the critical-people list. Honest reforecast, including the synergies that will not land. Integration governance either wound down or formally extended, but not left ambiently running for two years.

What the acquirer usually gets wrong about itself

Serial acquirers who integrate well share an unglamorous trait: they know their own capacity. The constraint is almost never the model or the target. It is the number of simultaneous integrations the acquiring organisation’s finance, IT and HR functions can absorb. A number that is usually one, sometimes two, and never the four in the pipeline.

The most valuable discipline in corporate development is therefore not sourcing. It is saying “not now” to a good asset because the last one is not finished.


General commentary on market practice. Not legal, financial or HR advice.