AI is making judgement the most valuable asset in corporate and M&A advisory
The emergence of AI is compressing much of the analysis that once was needed for corporate and M&A advisory, raising uncomfortable questions around fees. Philip Goldhahn, Director at EM Advisory, argues that as AI makes analysis increasingly accessible, experience, context and commercial judgement will become even more important.
During a recent due diligence, a buyer asked one of our clients a detailed question about a core metric. These are key moments in sale processes where speed and precision are important, and a slow or unconvincing answer invites the buyer to assume the worst. Until recently, answering it would have meant several days of work. The client would extract the data, one of our analysts would clean and analyse it, and we would prepare a response that could withstand further scrutiny.
This time, the client connected Claude to its Google data warehouse and generated the analysis within minutes.
What remained was the part that had always required judgement: deciding what the analysis meant in the context of the question and the transaction. The preparation took hours instead of days. The decision at the end of it was unchanged, as was the cost of getting it wrong.
From analysis to judgement
Advisers have always been paid for judgement, but that judgement has traditionally been bundled with a considerable amount of analysis and document production. Before an adviser could interpret a set of information, somebody had to assemble it. Before deciding which buyers might be interested, an analyst had to research them. Much of that work remains necessary, but AI is compressing the time it takes and allowing teams to go further.
Buyer research is one example. A process involving dozens of potential acquirers once imposed practical limits on how deeply each one could be analysed. Advisers ranked buyers on previous engagement and a handful of visible transactions. With AI, advisers can now examine previous acquisitions, management commentary and stated priorities in far greater detail, and test the rationale for an approach against what is known about each company. The same names still appear on the list. We can now say with more confidence which of them to call first.
Buyers are using the same AI tools. A seller can analyse a potential acquirer more deeply, but an acquirer can also analyse the seller more thoroughly and much earlier in the process. Buyers can test a company’s narrative, interrogate financial performance and identify inconsistencies faster than before. The practical effect is that the weak explanation which may have previously survived a first meeting now rarely does. Preparation that used to wait until the data room opened has moved to the front of the process.
The uncomfortable version of that symmetry is commercial. If a client can generate in an afternoon the analysis that once occupied a team for a week, the client will reasonably ask what the fee is for. Advisers who cannot answer that with something other than the work they produce will find it answered for them, in the fee negotiation.
The adviser’s intangibles
As both sides become better at analysing information, access to it stops being an advantage. What cannot be prompted is what advisers accumulate privately: years of conversations with investors, acquirers, lenders and lawyers and the experience of doing hundreds of different transactions.
I recently spoke with the CEO of a private company who set out their acquisition strategy, the capabilities the company wanted to buy, its growth priorities and how it viewed potential targets. Almost none of that was public. Conversations like that build an understanding of how acquirers think, not just what they say they want. That shapes how we advise clients on narrative and timing.
Context alone is only part of the value of an adviser. The judgement lies in deciding how much weight to give it. A stated priority, whether public or private, may be genuine, tentative or unlikely to survive a board discussion.
That judgement extends beyond information. M&A transactions are shaped by personal incentives, trust and emotion as much as financial analysis. A founder who keeps returning to what happens to their team is not asking about employment clauses; they are deciding whether they can live with the buyer. Experienced advisers recognise the pauses, repeated questions and changes in tone that reveal what may really be driving the other side, and place those observations within a broader commercial context before recommending a course of action.
AI can analyse large amounts of data, identify patterns and help test assumptions. At some point, however, the client needs to make a choice with incomplete information. The role of the adviser is to make a recommendation and be accountable for it, when they ask: Should we accept the offer or keep pushing? Is this buyer losing interest, or negotiating? Which differences in positions should be addressed with deal structure, and which differences are not solvable?
André Meyer, who led Lazard’s US operations for several decades, remarked once that “the merger business is ten per cent financial analysis and ninety per cent psychoanalysis”. He was describing a market without spreadsheets, let alone large language models, and the ratio was slightly exaggerated. But the observation is just as true now as it was then: transactions depend on understanding what people want and how they are likely to behave.
The obvious objection is that judgement can soon be done by AI. Much of what advisers call judgement is pattern recognition across dozens of transactions, and pattern recognition is precisely what AI systems are getting better at. But what sits behind it is narrower than pattern recognition: experience and trust. A model can produce a view, but it cannot answer a call of a founder and give them confidence that the decision is right.
Conclusion
AI will continue to transform corporate advisory. It will reduce the time required to produce research and analysis while continuing to raise expectations across the profession. The advisers who create the greatest value will use AI to spend less time gathering information and more time testing assumptions, positioning businesses and helping clients make better decisions. Information is becoming increasingly accessible. Experience, context and commercial judgement remain much harder to replicate.
Great advice was never just about documents and analysis, it’s about trust, meeting the specific client needs and solving for a solution that weights the different needs for that specific client.
About the author: Philip Goldhahn is a corporate advisor specialising in M&A and strategic transactions. He has helped close over 40 transactions across technology, business services and consumer sectors.
