The most common failure in due diligence is not a missed finding. It is a thorough, well-evidenced answer to the wrong question. That failure is set in motion at the very start, in scoping, which is also the phase most people rush because it does not feel like real work yet. It is the phase that determines whether everything after it is useful.
Scoping is where you decide what the assessment is actually for. Three questions have to be answered before any evidence is gathered.
1. What decision does this assessment serve?
An investor about to lead a Series B has different questions than an enterprise choosing between two vendors, and both differ from an acquirer integrating a target post-close. The assessment has to be built backward from the decision. If you do not know exactly what the client will do with the report, you cannot know what belongs in it. "Assess the AI" is not a scope. "Tell us whether this company's core model is defensible enough to justify the valuation" is.
2. What is actually in scope?
AI companies are rarely a single system. There is the headline model, the data pipeline feeding it, the infrastructure around it, and often several smaller models doing unglamorous work. Scoping decides which of these the engagement covers and, just as importantly, which it does not. An assessment that tries to cover everything at equal depth covers nothing well. Naming what is out of scope is part of doing the job honestly.
3. What evidence will we require, and what standards apply?
This is where you set expectations before they become a fight. What artefacts will the target need to provide, what evidence threshold will you hold findings to, and which standards frame the analysis. Deciding up front that you will reference, say, the EU AI Act for regulatory classification and ISO 25010 for software quality means the analysis later has a backbone rather than being assembled ad hoc.
How investment and procurement scoping differ
The two engagement types pull scoping in different directions. Investment diligence weights the questions that affect value and risk over the hold period: defensibility, team capability, scalability, governance maturity. Procurement diligence weights fit and integration: will this system work in our environment, on our data, within our obligations. The same target assessed for an investment and for a purchase produces two legitimately different scopes, because the decisions are different.
The cost of skipping it
We have seen engagements where the analysis was excellent and the scope was wrong, and the client got a beautiful report that did not help them make their actual decision. Scoping is cheap. Redoing an assessment because it answered the wrong question is not. Spend the time at the start defining the decision, and the rest of the work has somewhere to land.
If you are weighing an AI investment, acquisition, vendor selection, or training programme, our team is happy to start with a conversation about scope and approach.
The views and findings in this article are shared for general information only. They are high-level perspectives, not legal, financial, regulatory, or other professional advice, and should not be relied upon for any specific decision or circumstance. For guidance tailored to your situation, please consult a qualified adviser.