Two companies sign, the lawyers shake hands, the press release goes out, and the hard part begins. An acquisition is a promise about the future: that two organisations will be worth more joined than apart. Post-merger integration (PMI) is the work of actually keeping that promise, and it is where most deals quietly come undone.

The quick version

  • Most of the value lives after close. A deal model is a forecast; integration is the only thing that turns it into cash. Studies put the M&A failure rate at 70–90%, and the failures cluster in execution, not in the price paid.
  • Decide what kind of integration this is. Full absorption, light-touch preservation, or something in between, pick deliberately, because over-integrating a business you bought for its difference destroys the thing you paid for.
  • Speed in the first 100 days matters. Stand up a dedicated integration office, name the few sources of value that justify the deal, and protect the base business so it keeps running while you join the two.
  • Culture is the silent failure mode. Roughly a quarter of executives name culture clash as the top reason integration fails, and it shows up as your best people leaving, not as a survey result.

The idea in depth

Value isn't bought, it's made, after signing

The most useful reframe in all of M&A is also the least glamorous: the deal does not create value, the integration does. In their foundational study Managing Acquisitions (1991), INSEAD's Philippe Haspeslagh and David Jemison drew a hard line between two things deal teams love to blur. Strategic fit sets the potential value of an acquisition; organisational fit determines how hard that value will be to actually capture. But the value that gets created flows from neither, it flows from how well you manage the integration process itself. The synergy in the model is, until you do the work, a hypothesis.

So the move is: treat the closing date as the start line, not the finish. Before you sign, write down the handful of specific things that have to go right after close for the numbers to be real, the contract you'll cross-sell, the duplicate platform you'll retire, the team you must keep, and assign each one an owner. If you can't name them, you've bought a forecast, not a business.

The limitation: Haspeslagh and Jemison's framework is a lens, not a formula. It tells you to think about fit and process; it doesn't tell you the right answer for your deal. Two integration leaders using the same model will reach different plans, and should, because the right amount of integration depends on why you bought the company.

Pick your integration approach on purpose

Not every acquisition should be integrated the same way, and the single most expensive mistake is to absorb a company you bought precisely because it was different. Haspeslagh and Jemison sort integration by two questions: how much strategic interdependence the two firms need (to share capabilities), and how much organisational autonomy the acquired business needs to keep what made it valuable. Cross those and you get four postures: absorption (high need to combine, low need to stay separate, fold it in fully), preservation (keep it at arm's length and let it run), symbiosis (gradually intertwine while protecting the core), and holding (no real operational integration at all).

So the move is: name your posture before Day One, and let it govern every downstream call, systems, brand, reporting lines, which leaders stay. If the thesis is "buy their distinctive culture and talent," preservation protects the asset; ramming them onto your HR system and your ways of working on day 30 is how you watch the founders resign by day 90.

quadrantChart
    title Choosing an integration approach
    x-axis "Low need to combine" --> "High need to combine"
    y-axis "Low autonomy needed" --> "High autonomy needed"
    quadrant-1 "Symbiosis: intertwine, protect the core"
    quadrant-2 "Preservation: keep it separate, let it run"
    quadrant-3 "Holding: no real integration"
    quadrant-4 "Absorption: fold it in fully"
The integration depends on why you bought the company, not on a default playbook. After Haspeslagh & Jemison (1991). Leaders Loop

Why deals fail, and where

The failure statistics are sobering and stubborn. Writing in Harvard Business Review in 2016, strategy scholar Roger Martin observed that "M&A is a mug's game, in which some 70% to 90% of acquisitions are abysmal failures." His diagnosis is sharp: acquirers obsess over what they'll get from the target, when the deals that pay off are usually the ones where the acquirer has something distinctive to give it, a capability, a channel, a balance sheet, that makes the combined business genuinely better. A deal justified only by "we'll extract value from them" tends to extract it from your own shareholders instead.

And when integration fails, culture is disproportionately to blame. McKinsey's work on the human side of mergers finds that around a quarter of executives point to a lack of cultural cohesion as the primary reason integration efforts fall short, even though some 95% of executives say cultural fit is critical going in. The gap between knowing it matters and doing something about it is the whole problem. McKinsey's practical advice is to stop talking about culture as "values" and start treating it as management practices, "the way we do things around here": how decisions get made, how performance is judged, who gets promoted.

The practical step: make culture an operational workstream with an owner and a deadline, not an offsite. Map the half-dozen everyday practices that differ most between the two firms, how a spending decision gets approved, how a disagreement gets escalated, and decide explicitly which company's way wins for each. Ambiguity here is what your best performers quit over.

"M&A is a mug's game, in which some 70% to 90% of acquisitions are abysmal failures.", Roger Martin, Harvard Business Review, 2016

The limitation: that 70–90% figure is widely cited and worth taking seriously, but "failure" is defined differently across studies, some mean shareholder return, some mean stated synergies, some mean accounting performance. Read it as a strong directional warning that integration is hard, not as a precise law of nature.

A worked example

The following figures are illustrative, to show the mechanics, not a real company.

Imagine a mid-sized software firm, NorthPay, acquiring a smaller, beloved scheduling app, Rota, for £40m. The deal model promised £6m of annual value: £2m from cutting Rota's duplicate cloud and back-office costs, and £4m from selling Rota into NorthPay's much larger customer base. On paper, a clean win.

The integration lead does three things differently from the failed version of this story. First, she picks a posture: symbiosis, not absorption. Rota was bought for its product culture and its loyal users, so the brand, the product team and the deployment cadence stay intact for a year, only finance, cloud hosting and legal merge on Day One. The £2m cost synergy comes from the "safe" half; the £4m revenue synergy is protected by leaving the bit customers love untouched.

Second, she names the value drivers and guards the base business. A small integration office tracks exactly two things, the platform migration and the cross-sell pipeline, while the rest of both companies are told, in writing, to keep hitting this quarter's numbers. Third, she runs culture as a real workstream: within the first month, the two firms agree whose expense-approval and release-sign-off practices win, so a Rota engineer knows on day 20 how to ship. The cross-sell lands slower than the model assumed, revenue synergies usually do, but the people stay, and the £4m arrives in year two instead of evaporating with a wave of resignations.

flowchart LR
    A(["Pre-close: write the
integration thesis"]) --> B(["Day One: legal close,
business keeps running"])
    B --> C(["First 100 days:
capture quick wins,
settle culture practices"])
    C --> D(["Beyond: revenue
synergies, steady
operating cadence"])
Integration is a sequence, not a switch, the thesis precedes the close, and the value trails it. Leaders Loop

Frequently asked questions

What's the difference between due diligence and integration?

Diligence happens before the deal and asks "should we buy this, and at what price?" Integration happens after and asks "now that we own it, how do we make the combined business actually work?" The trap is treating them as separate teams who never talk. The strongest acquirers carry the same short list of value drivers, Bain calls it the deal thesis, the five-to-ten most important sources of value, from diligence straight into the integration plan, so the people capturing the value are the people who priced it. See Due diligence.

How long does post-merger integration take?

The intense phase is roughly the first 100 days, where the goal is business continuity on Day One and a stable operating rhythm by the end. But full value capture runs longer: cost synergies tend to land within 12–18 months, while revenue synergies are slower and less certain, they often take two to three years and frequently undershoot the model. Plan for a sprint and a marathon; teams that disband the integration office after 100 days usually leave the harder synergies on the table.

Who should run the integration?

A dedicated integration leader with genuine authority and CEO air cover, not the deal team that negotiated it (their job is done) and not a committee. The common structure is an Integration Management Office (IMO): a small, senior, cross-company group that coordinates workstreams, routes the flood of decisions, and reports to the top. The point of the IMO is to let the other 95% of the company keep running the base business while a focused few do the joining.

Should we rebrand and merge systems immediately?

Only if your integration posture is absorption. Speed is an asset for the obvious, low-regret moves, payroll, email, a single source of financial truth. It's a liability for the things you bought the company for. If a distinctive brand or product culture is part of the value, forcing it onto your systems and identity early is how you destroy the asset while feeling productive. Match the pace to the posture.

What's the single biggest mistake?

Letting the base business drift while everyone stares at the integration. The deal is a fraction of the company's value; the existing business is the rest. If your salespeople spend the first quarter in town halls about the merger instead of selling, you can hit every integration milestone and still miss the number, and a merger that misses the number is, to the market, a failed one.

Related in the Toolkit

Where to go next