Kennisbank

The deal thesis: why it exists and why it goes missing

Every acquisition starts with a piece of reasoning. Why these two companies together are worth more than apart, which synergy explains that difference, and within what timeframe that should become visible. That reasoning is called the deal thesis. It appears in the investment memo, has been presented to an investment committee or board of directors, and has received a signature somewhere.

The problem is not that the deal thesis is wrong. The problem is that after closing it often no longer appears explicitly anywhere. The memo disappears into an archive, the deal team moves on to the next transaction, and the people who have to carry out the integration have rarely seen the full document — sometimes deliberately, to avoid sharing negotiation-sensitive information too widely. What remains is a vague sense that there should be 'synergy', without anyone on the work floor being able to say which one, how much, and from whom.

From thesis to synergy baseline

The first step is not execution, but translation. A deal thesis like 'economies of scale in procurement' or 'cross-sell between customer bases' is not a workable assignment. It needs to be converted into a list of synergies, each with a name, an amount, an assumption and an owner. That is what a synergy baseline does: the thesis is pulled apart into components that can be tracked, tested and, if necessary, dropped separately.

That translation immediately raises the question of how you substantiate a synergy that initially existed on paper only as an assumption. A synergy that only exists because it sounded good during the negotiation rarely survives the first months after closing. A synergy with a calculation, a data source and a delivery timeline does — not because the figure guarantees it will land, but because it can be tested.

Why the thesis evaporates

There are a few recurring ways in which a deal thesis gets lost along the way.

The first is ownership. A synergy without a name is a synergy that no one thinks about in the morning. The deal team knew who had drawn up the thesis, but no one recorded who carries it after closing. That is exactly the question that needs to be answered before a synergy gets a chance: who owns a synergy once the deal team has moved on to the next transaction.

The second way is assumption erosion. A synergy amount rests on a market price, a volume, a number of FTEs or a supplier contract at the moment of the deal. Six months later that market price has changed, that volume has dropped, or that contract has been terminated. No one reported it, because no one remembered that the amount depended on it. That is why which assumption underlies your synergy amount is a question you need to ask yourself not once, but repeatedly.

The third way is silent irrelevance. The world changes, the organisation changes, and a synergy that made sense on day one no longer does after a year. No one dares to say that out loud, because the figure has already appeared in a report to the board of directors. That way a dead synergy keeps running in an overview for months, until someone asks how you test whether a synergy is still achievable and the answer turns out to be uncomfortable.

What tools do and do not do here

A synergy baseline, a Day 1 plan and a Day 100 plan do not solve this by themselves. They structure the question so that it keeps being asked. A list with owners, assumptions and test moments does not prevent a synergy from evaporating — it makes visible when that happens, so that someone can choose to drop that synergy instead of letting it quietly sink away.

That is a deliberate limitation. The tool cannot judge whether an integration is worthwhile; it can only organise the information with which someone makes that judgement. And that judgement needs to come back regularly, not only at the start. An integration office that tracks every synergy with a simple status — on track, under pressure, stopped — prevents a failing synergy from running unnoticed for months. That is the idea behind how you track synergies with traffic lights: not to embellish progress, but to surface setbacks early.

And sometimes the conclusion is that a synergy is not going to land. That is not a failure of the process, it is what the process is for. The question what you do with a synergy that does not land should be on the agenda by default, alongside the question of whether components should actually be merged at all. Integration is not automatically a good thing; sometimes a business unit retains more value if it keeps operating separately, and the deal thesis needs to keep that possibility open rather than ruling it out.

This page describes a way of working, not a track record. There is no completed engagement being referred to, and the tool that supports the synergy baseline, the Day 1 plan and the Day 100 plan is under construction. Anyone who wants to work with this can sign up for the waiting list.

The next question: who does the work

Translating a deal thesis into a synergy baseline, revisiting assumptions, assigning ownership and maintaining traffic lights is itself also work — repeated, structured, and partly predictable in form. That makes it a suitable candidate for the question that plays a role beyond integration alone: which part of this kind of task can be taken over by AI, and which part cannot. FTE TO AI's werkscan calculates that per task, not as a gut-feel estimate but as a breakdown of what a task involves and where automation does or does not get a grip.