What an acquisition premortem is
A post-mortem asks why something died. A premortem asks the same question before the event, while there is still time to change the outcome. The psychologist Gary Klein set out the method in Harvard Business Review in 2007: tell the team the project has failed, then ask each person to write down why.
The method rests on research into what is called prospective hindsight. In a 1989 study by Mitchell, Russo and Pennington, people who imagined an outcome had already happened came up with around 30% more reasons for it than people asked what might happen. "What could go wrong?" invites reassurance. "It went wrong, what happened?" invites an answer.
Applied to a deal, the exercise is simple to describe. It is six months after completion. The synergy has not arrived, two of the people you were buying have resigned, and a customer the model depended on has given notice. The premortem is the written explanation of how you got there, produced while the deal can still be repriced, conditioned or stopped.
Why due diligence does not cover it
Legal, financial and commercial due diligence all point at the target. They establish that the business is real, that the numbers hold, and that the contracts say what the seller claims. None of them asks whether the buyer can absorb what it is about to own.
That gap is a common place for integrations to go wrong. A target can be sound while the acquirer has no spare leadership capacity, an operating model that does not match, and no decision on which systems will survive. Those are facts about the buyer, and standard due diligence is not engaged to check them.
| Due diligence | Acquisition premortem | |
|---|---|---|
| Subject | The target | The combination, and the acquirer in particular |
| Question | Is this business what it claims to be? | Assuming the deal failed, why did it? |
| Evidence | Documents, accounts, contracts | Operating model, leadership capacity, who owns which relationships |
| Output | Findings and warranties | A failure narrative with named owners and first actions |
| Affects | Price and legal protection | Price, conditions and whether the integration plan is achievable |
What a useful premortem tests
A premortem that produces forty risks has failed. The discipline is to test a small number of areas where acquisitions repeatedly break, and to keep each finding separate so one serious problem is not averaged away by five healthy scores. These are the six areas Tollgate scores.
Operating model compatibility
Whether the two businesses are organised to deliver in the same way. A firm that sells fixed projects and a firm that sells ongoing managed services can share a sector and still be structurally incompatible. The failure story usually reads: we bought a different kind of business and ran it like ours.
Change capacity
Whether the acquirer's leadership has real time to run an integration on top of the day job. Capacity is almost always assumed in the deal model and almost never measured. Bandwidth and capability are separate questions and need separate answers.
Workforce and retention
Which individuals the deal value depends on, what their contracts say, and which incentives change on completion. If the answer to "who holds this client?" is one person, that is a finding.
Revenue and relationships
How concentrated the revenue is, who personally owns the key accounts, and whether delivery capacity covers the work already sold. Committed work with nobody free to deliver it is a problem the buyer inherits on day one.
Customer continuity
What changes for the customer during transition, and who is accountable for noticing when it goes wrong. Customers rarely leave because of the deal announcement. They leave three months later, when their usual contact has gone and nobody replaced them.
Technology decision
Integrate, replace or run separately. The premortem does not need the full technical answer, but it does need the decision made before signature. Left open, it gets made by default after completion, at higher cost.
How to run one yourself
You can run a basic acquisition premortem internally in half a day. It will not be independent, but it is far better than not doing it.
- State the deal thesis in one sentence, with the number attached. If the room cannot agree the sentence, you have your first finding.
- Set the date. It is six months after completion and the thesis has been missed. This is stated as fact, not as a possibility.
- Each person writes their reasons alone. Ten minutes, no discussion. Writing first stops the most senior voice setting the answer for everyone else.
- Sort the reasons into target and acquirer. Most teams are surprised by how many land on their own side of the line.
- For each cause, record four things. The evidence for it today, a named owner inside your organisation, a first 90-day action, and whether it should change price or conditions.
- Write the retrospective paragraph. One paragraph, in the past tense, explaining why the deal failed. File it with the board paper.
The paragraph is the point of the exercise. If the deal does disappoint, the explanation has already been written, dated, and seen by the people who approved it.
Where an internal premortem falls short
The people in the room are the people whose capacity is being assumed. Nobody scores their own bandwidth honestly two weeks before a deal they have spent six months pursuing. Momentum does the rest: by this stage the question has quietly changed from "should we?" to "how do we get it over the line?"
An independent premortem removes that pressure. It tests the acquirer with the same scepticism the lawyers and accountants apply to the target, and it reports in the same terms those workstreams use: risk to the deal, risk to price.
What it should lead to
A premortem is not a go or no-go verdict. It gives the board a basis for one of three decisions, taken with the evidence in front of them.
- Proceed as planned. The failure causes are known, owned and manageable inside the existing integration plan.
- Proceed with conditions. Specific findings change the price, the retention terms, the timetable or who leads the integration.
- Reclassify the deal. The gap between the two operating models is large enough that this is a change programme with an acquisition attached, and should be resourced and governed as one.
Tollgate is built on CUSCOE, the operating model CXBridge uses to define what clean function boundaries and accountability look like. That is what gives the premortem a fixed standard to test both businesses against.
Common questions
Is an acquisition premortem the same as a risk register?
No. A risk register lists what might go wrong and tends to grow until nobody reads it. A premortem starts from the assumption that the deal has already failed, which forces the team to commit to the few causes they believe are most likely. It ends in a short written narrative with owners attached, not a scored list.
When should an acquisition premortem be run?
After heads of terms and before signature. Earlier than that, the deal thesis and integration assumptions are not firm enough to test. Later than that, the findings can no longer change the price, the conditions or the integration plan.
Who should be in the room?
The people who will carry the integration, not only the people who sourced the deal. That means the operational leaders whose teams absorb the work, plus finance and whoever owns the customer relationships on both sides. Each person writes their reasons alone before any discussion.
Does a premortem replace due diligence or integration planning?
It replaces neither. Due diligence tests whether the target is what it claims to be. Integration planning sets out what happens after completion. The premortem sits between them and tests whether the acquirer can carry out that plan while still running its existing business.