Due diligence isn't one activity. Early on, a buyer is exploring: is there anything here that should stop us, and what is this business really? Later, once a price is agreed in principle, the buyer is confirming: is what we were told true, and is anything missing? The two phases ask different questions, use different documents and fail in different ways. Treating them as one is how teams spend confirmatory budgets on exploratory questions, or close with exploratory gaps.
The two phases side by side
| Exploratory | Confirmatory | |
|---|---|---|
| When | Before or around the letter of intent | After the LOI, before signing |
| Question | Should we pursue this, and at roughly what value? | Is what we were told true, and complete? |
| Inputs | CIM, public filings, management presentation, a partial data room | The full data room, contracts, interviews, site visits |
| Output | A go / no-go, and a list of concerns | Findings that feed price, the purchase agreement, disclosure schedules and insurance |
| Main risk | Missing a deal-breaker until money is committed | Confirming the story instead of testing it |
These terms are practice, not law. Firms draw the line in different places, and on a fast process the phases overlap. The distinction is still useful because it tells you what question each hour of review is meant to answer.
What exploratory diligence should find
The exploratory phase is about deal-breakers and value drivers. Its job is not completeness. Its job is to find, cheaply, anything that would make the rest of the process pointless:
- sanctions or anti-bribery exposure;
- litigation or regulatory action that is real and live, not hypothetical;
- a going-concern doubt, a covenant breach or debt that can't survive the transaction;
- key contracts or licences that terminate on a change of control;
- a merger-control problem that would block or delay closing.
The constraint is that the data room is incomplete. A good exploratory review records what it couldn't see, so the confirmatory phase knows what to ask for.
What confirmatory diligence should do
Confirmatory work tests specific statements against evidence. Revenue in the model against invoices. Headcount in the presentation against payroll. The customer list against the contracts. Its outputs go into legal documents: price adjustments, specific indemnities, representations and warranties, and the disclosure schedules against them.
The step before both: first-pass screening
There is a step many teams now run before either phase, which is mechanical rather than analytical: screen every document in hand for risk signals before anyone reads it in full.
Screening doesn't replace exploratory diligence. It decides where exploratory diligence starts. Instead of an associate reading documents in the order the seller uploaded them, the team starts with the documents that scored highest in a deal-blocking category and the ones with suspicious silence.
| Stage | Who | Output |
|---|---|---|
| Screen | Software, checked by a reviewer | Ranked findings with quotes, and named gaps |
| Explore | Deal team and advisers | Go / no-go, concerns list, value view |
| Confirm | Advisers and specialists | Evidence for price, contract and insurance |
This is the job Anweshna is built for: first-pass risk screening of deal documents across 15 M&A risk categories, with every finding checked against its source text. It tells a team where to dig first. It isn't independent verification, and it doesn't perform either diligence phase on its own.
Screening again during confirmatory work
Data rooms fill late, and the documents that arrive late are disproportionately the ones a seller was slow to share. Re-screening each new batch as it lands catches the late surprise while there is still time to act on it. The same logic applies after the exploratory phase is done: new documents deserve the same first pass the early ones got.
How to tell which phase you are really in
- If you are still deciding whether to do the deal, you are exploring, whatever the engagement letter says.
- If a finding would change the price or the contract but not the decision, you are confirming.
- If a finding would change the decision and you are already confirming, stop and treat it as an exploratory question again. That is what the phase line is for.
Which documents belong to which phase
| Document | Exploratory use | Confirmatory use |
|---|---|---|
| CIM or teaser | The seller's story, read for claims to test later | Rarely used: superseded by evidence |
| Public filings | Litigation, going-concern language, auditor changes, risk factors | Cross-checked against management accounts |
| Top customer and supplier contracts | Change-of-control and termination terms | Full terms, pricing, renewal history |
| Financial statements | Trends, one-offs, debt and covenants | Quality of earnings, working capital, net debt |
| Regulatory correspondence | Live investigations or enforcement | Remediation status and cost |
| Employment and benefit documents | Key-person and change-of-control payments | Full liability and integration cost |
The same document often appears in both phases. What changes is the question asked of it: exploratory reading asks "could this stop us?", confirmatory reading asks "is this exactly what we were told?"
Signs the exploratory phase isn't finished
- The team can't name the three issues most likely to stop the deal.
- Whole categories, such as anti-bribery, data privacy or merger control, have no documents in the data room and no follow-up request.
- Findings are paraphrased from the CIM rather than quoted from contracts or filings.
- Nobody has checked which key contracts change on a change of control.
Any one of these is a reason to finish exploring before paying for confirmation.
Bottom line
Exploratory diligence looks for reasons to stop. Confirmatory diligence tests what you were told. A first-pass screen comes before both and decides where the expensive hours go first. Keep the three distinct, and each one gets cheaper. For timings, see How Long Does Due Diligence Take?
Sources
This page describes common practice; "exploratory" and "confirmatory" are working terms, not legal definitions, and it makes no external statistical claim. We cite only sources we have retrieved and read; see our methodology. Nothing here is legal advice.