Vendor technical due diligence (VDD) is an independent review of a business's technology, commissioned by the seller before a sale and shared with bidders, with the eventual buyer usually able to rely on it under a reliance letter. Done well, it starts with a private readiness review 12 to 24 months before the sale, so the seller can fix the cheap, high-impact problems first: missing IP assignments, open-source licence conflicts, open penetration test findings, admin access held by one person, key-person dependency, cloud waste, missing documentation and AI governance. Large re-platforms are usually better costed and disclosed than started.
I've been through technical due diligence from both sides: as a founder and CEO being diligenced in a sale and a merger, and as an investor commissioning it. Sellers consistently underestimate how much of what a buyer finds was cheap to fix a year earlier. They also overestimate how much a buyer will pay for work that is half-finished on launch day.
A problem you have found, costed and disclosed is a line in the report. The same problem found by the buyer is a negotiation.
What is vendor technical due diligence?
A sell-side tech due diligence is the same work a buyer's technical due diligence adviser would do, commissioned instead by the seller. The adviser reviews architecture, code, engineering practice, security, IP and open source, the team, running costs, the product roadmap and, increasingly, AI. The output is a report written for bidders. In a competitive process it sits in the data room alongside the financial, tax and legal vendor reports.
It helps to separate two pieces of work that often get called the same thing.
| Readiness review | Vendor due diligence report | |
|---|---|---|
| When | 12 to 24 months before launch | Two to four months before launch |
| Audience | The seller only | Bidders, then the buyer |
| Purpose | Find problems while there's time to fix them | Give buyers an independent, reliable view to price from |
| Tone | Blunt, internal, includes things that may never be written down | Factual, balanced, every finding evidenced |
| Reliance | None | Reliance letter to the eventual buyer and often its lenders |
The readiness review is where most of the value sits. It's the reason a VDD report can be clean.
When is vendor technical due diligence worth it?
It earns its cost when technology is central to value and the process is competitive. A software or tech-enabled services business going to an auction with several bidders benefits most: one independent report replaces five sets of buy-side questions, shortens the timetable, and stops each bidder inventing its own estimate of the technical debt.
It's worth less in a bilateral process with a buyer who will run deep diligence anyway, in a small deal where the technology is peripheral to value, or where the seller isn't willing to act on what it finds. A VDD report that reveals problems the seller then ignores is worse than none.
My rule of thumb: if a buyer's investment case depends on the product, the platform or an AI story, do it. If the technology is a standard set of business systems in a services company, a readiness review is enough, and the buyer's own IT diligence will cover the rest.
Who reads a VDD report and how much do buyers rely on it?
Bidders read it first, in the first round, to decide what to bid and what to ask. The winning bidder then usually gets a reliance letter from the adviser, which gives it a contractual right to rely on the report, subject to an agreed liability cap and terms. Lenders financing the deal often take reliance too.
Reliance doesn't mean the buyer stops asking questions. Most buyers commission targeted confirmatory work on the areas their plan depends on. Warranty and indemnity insurers expect it: Browne Jacobson's guidance on W&I insurance and sell-side due diligence notes insurers look for the buyer to "kick the tyres" of the vendor reports, and that addressing the reports to the buyer doesn't remove that need. Insurers are also looking harder at technology risk. In July 2026 the International Underwriting Association's Transactional Liability Committee published a report on cyber risk in W&I underwriting, warning that W&I insurers may be over-reliant on cyber policies that often end at completion, and that risk analysis frequently skips penetration testing and technical validation.
So the realistic goal for a VDD report is that buyers trust it enough to price from it and limit their own work to confirmation. That only happens if the report is independent and the findings are specific. A report that reads like marketing gets ignored, and then the seller has paid for nothing.
Preparing for tech due diligence: the fix list, in order
The list below is ordered by how quickly and cheaply each item can be fixed relative to how much it worries a buyer. Start at the top. The first four are almost always worth doing whatever the timetable. The full version, 97 items across 15 workstreams with timings and owners, is our exit readiness checklist.
| # | Issue | Typical time to fix | Cost to fix | What it does to a deal if left |
|---|---|---|---|---|
| 1 | Admin access held by one person | Days | Negligible | Key-person and security finding, often a completion condition |
| 2 | Missing IP assignments | Weeks to months | Legal fees, sometimes a payment to an ex-contractor | Condition of completion, specific indemnity |
| 3 | Open pen test findings | Weeks | Engineering time plus a retest | Price chip or specific indemnity, harder W&I cover |
| 4 | Open-source licence conflicts | Weeks to months | Engineering time to replace or isolate components | Specific indemnity, remediation before completion |
| 5 | Missing documentation | One to three months | Engineering time | Slower diligence, more key-person concern |
| 6 | Cloud cost waste | One to three months | Engineering time, often pays for itself | Margin questions, lower EBITDA |
| 7 | AI governance gaps | One to three months | Mostly management time | AI claims discounted, governance questions in the SPA |
| 8 | Key-person dependency | Six to 18 months | Hiring, retention packages | Retention demands, earn-out structure, price |
1. Single admin access
One founder or long-serving contractor holding the only credentials to the production cloud account, the domain registrar, the app store listings or the code repository. It's the cheapest item on the list and one of the most alarming to a buyer. Move every critical account to company-owned identities, put at least two named people on each with multi-factor authentication, and keep a register. It takes days.
2. IP assignments
Every employee and contractor who has written code should have signed something that assigns the IP to the company. The gaps are usually early contractors and the agency that built version one. Missing assignments become a condition of completion, and a former contractor who learns the business is being sold has every reason to drive a hard bargain. Find them a year out, while the conversation is still cheap.
3. Penetration test findings
Buyers ask for the last two pen test reports and evidence of what happened to the findings. Critical or high findings left open for a year tell a buyer more about the engineering culture than the certificate on the website. Commission a fresh test early enough to fix what it finds and retest before the VDD fieldwork.
4. Open-source licence conflicts
This is more common than most CTOs think. Black Duck's 2026 Open Source Security and Risk Analysis, based on 947 commercial codebases from audits that included 197 M&A transactions, found licence conflicts in 68% of them, up from 56% the year before. The same report found 92% of codebases contained components four or more years out of date, and only 54% of organisations evaluating AI-generated code for IP and licensing risks. The typical problem is copyleft code (GPL, AGPL) inside a product the business distributes or hosts, or AI-generated code that has carried in licensed material without its licence.
Run a software composition analysis scan a year out, not two months out. Replacing or isolating a component is straightforward with time and fraught under a deadline.
5. Documentation
A current architecture diagram drawn by the team, runbooks for deployment and incident response, and enough written down that someone other than the author could take over each critical system. This speeds up diligence and reduces the key-person finding. Write it for the next engineer you hire. A buyer will find that version more useful too.
6. Cloud cost
Buyers compare hosting cost growth with revenue growth. Flexera's 2026 State of the Cloud report found respondents estimated 29% of their cloud spend was wasted, up from 27%, with AI workloads blamed for the first rise in five years. Most businesses that haven't looked will find unused environments, oversized instances and storage nobody needs. The saving goes straight to EBITDA, which at exit is multiplied.
7. AI governance
If AI features in the equity story, the buyer will test it. Grant Thornton's 2026 AI Impact Survey found only 9% of private equity leaders very confident they could pass an AI governance audit within 90 days. At portfolio company level, a short policy, a register of AI tools and the data going into them, and clear vendor terms on whether customer data trains third-party models answer most of what a buyer will ask. Building the evidence for the AI claims themselves is a bigger job, and it's covered in will buyers pay for your AI?
8. Key-person risk
The hardest item on the list, because it can't be fixed with a document. If one engineer wrote most of the core product and nobody else understands it, a buyer will demand retention arrangements, push value into an earn-out, or reduce the price. It takes a year or more to fix properly: a second senior engineer, paired work on the critical systems, documentation that someone else has actually used. Start it first even though it finishes last.
What to leave alone before a sale
Some work makes the business look worse during a process than it would if left untouched.
Don't start a re-platform you can't finish before launch. A half-migrated system is two systems, both partly documented, and a buyer will price the risk of the second half going wrong. Write a costed plan for it instead and put it in the report. A buyer can model a known £400k project. A buyer can't model a migration that's 60% done with the lead engineer on notice.
Don't chase a certification you don't need. ISO 27001 started six months before a process is unlikely to finish and reads as a scramble. Cyber Essentials Plus is achievable quickly and, alongside a clean pen test, answers most mid-market buyers.
Don't build features to dress up the roadmap. Buyers check roadmap delivery against what shipped, as covered in the technology due diligence checklist, and a rushed feature with no usage counts against you.
And don't hide things. A problem you have found, costed and disclosed is a line in the report. The same problem found by the buyer is a negotiation.
A vendor technical due diligence timeline
Counting back from the planned launch of the sale process:
- 24 to 18 months out: readiness review. Score every area a buyer will look at, cost each finding, and agree the fix list with the board.
- 18 to 6 months out: fix. Work down the list in order. Start key-person work immediately. Run the open-source scan and the pen test early enough to remediate and retest.
- Six to four months out: appoint the VDD adviser and agree scope and reliance terms. Make sure the scope matches the questions the likely buyers will ask, including AI.
- Four to two months out: VDD fieldwork, typically two to four weeks for a mid-market business, followed by a draft, management comments and a final report.
- Launch: the report goes into the data room. Keep the evidence behind it ready for confirmatory questions.
- Before signing: a short bring-down update if the process runs long, so the report still reflects the business.
The AI part of the story follows the same clock, with one difference: measured results need time to accrue, so AI use cases you want credited need to be live 12 to 18 months before launch, which leaves six to twelve months of measured results in the data room.
How much does vendor technical due diligence cost?
Our exit readiness baseline is the readiness review: £15k to £25k over two to three weeks, producing the costed fix list and a timeline back from the planned sale. A vendor technology due diligence report written for bidders, with reliance for the eventual buyer, costs £25k to £40k depending on the size and complexity of the technology estate. Funds often commission the baseline at the start of exit planning and the report a few months before launch.
If you're 12 to 24 months from a sale and want to know what a buyer's technical adviser will find, our exit readiness work starts with a short call to scope the baseline.
Questions
What is the difference between vendor due diligence and a pre-sale readiness review?
A readiness review is private to the seller. It is run 12 to 24 months before a sale to find problems while there is time to fix them, and nobody else sees it. A vendor due diligence report is written for bidders, usually two to four months before launch, and the eventual buyer can rely on it under a reliance letter. Running the first makes the second shorter, cheaper and cleaner.
Does a buyer still do their own technical due diligence if there is a VDD report?
Usually, yes, though less of it. Most buyers use the vendor report to understand the business quickly and price it, then commission targeted confirmatory work on the areas that matter most to their plan, often security, scalability or AI. Warranty and indemnity insurers also expect the buyer to test the vendor reports rather than accept them unread, so a VDD shortens buy-side work rather than replacing it.
Should the seller choose the same firm for vendor and buy-side work?
The seller chooses the vendor adviser. What matters is that bidders see the firm as independent and competent, because a report they distrust adds nothing. Pick an adviser buyers in your sector recognise, agree that the report will be addressed to the eventual buyer under a reliance letter, and keep management's commentary separate from the adviser's findings.
Can vendor technical due diligence cover AI?
It should, if AI features in the equity story or the business could be exposed to AI disruption. The AI section covers an inventory of AI systems and vendors, usage and unit cost data, data rights, governance, and an honest view on exposure. Buyers increasingly expect it, and an independent view on the AI claims carries more weight than management's own description.
What happens if the VDD report finds something serious?
Then it has done its job early. The seller can fix it, cost it, or decide how to present it before bidders see the report. Serious findings are usually written up with management's remediation plan and a cost. That gives buyers a known, priced issue rather than a discovery late in exclusivity, which is when price chips and specific indemnities tend to appear.