Technical Due Diligence Checklist for Investors

Technical Due Diligence Checklist for Investors

What to ask about a target’s technology before you invest or acquire, area by area, and the red flags that should change the price or stop the deal altogether.

By Jon Silver · 20 November 2026

Technical due diligence can go wrong in two directions. It can shrink to a questionnaire the target fills in about itself, or grow into a deep audit of something that never mattered to the deal. This checklist sits between the two: the areas that affect what a technology business is worth, the questions that get past the sales pitch, and the red flags that should change the price or stop the deal.

It is written for private equity, venture capital and corporate buyers. If you are the one being sold, it is also the list a buyer’s team will work through.

How to Use This Checklist

Not every line matters equally. A five-person SaaS start-up and a 300-person software house carry different risks. Work out which assets carry the value in your investment case, and go deepest there.

Ask for evidence, not assurances. “We have automated tests” is an assurance; a test report from last week’s build is evidence. Where the answer is a document, ask to see it. Where the answer is a system, ask to see that too.

Record strengths too. A list of problems with no context about what works will overstate the risk.

1. Architecture and Design

What to ask

  • Is there a current architecture diagram, and does it match what is actually deployed?
  • Which components are bespoke, which are third-party products, and which are open source?
  • How are the main systems integrated, and what happens when one of them fails?
  • What would have to change to support the growth in the investment case?

Red flags

  • Nobody can draw the architecture without checking with one particular person
  • A monolith with no clear boundaries, where every change risks breaking something unrelated
  • Critical functions built on a platform or product version that is out of support

2. Code Quality and Technical Debt

What to ask

  • Can we have read access to the source code repository, including its history?
  • What proportion of the code is covered by automated tests, and do the tests run on every change?
  • Where does the team itself say the technical debt is, and is there a plan to pay it down?
  • How is code reviewed before it reaches production?

If the value sits in a single application, a code health check answers these questions for that codebase on its own.

Red flags

  • Large modules that nobody wants to touch
  • Tests that exist but do not meaningfully test anything, or a build that is routinely red
  • Hard-coded special cases for individual customers scattered through the code
  • A long-running “rewrite” that is consuming the team while the old system still runs the business

3. Security

What to ask

  • When was the last independent penetration test, what did it find, and what has been fixed?
  • How are secrets such as API keys and passwords stored and rotated?
  • How is access to production systems controlled, and is multi-factor authentication enforced?
  • What security incidents has the business had, and how were they handled?

Red flags

  • Secrets committed to source repositories or embedded in code that runs in the browser
  • Shared administrator accounts, or former staff who still have access
  • No dependency scanning, so known vulnerabilities in third-party libraries go unnoticed
  • Security findings from previous tests still open

4. Infrastructure and Cloud Cost

What to ask

  • Where is everything hosted, and who owns the accounts?
  • What is the monthly cloud bill, how has it trended, and how does it scale with customers or usage?
  • Are backups tested by actually restoring them? When was disaster recovery last rehearsed?
  • Is infrastructure defined as code, or built by hand?

Red flags

  • Cloud accounts held in a founder’s or contractor’s personal name
  • Hosting costs rising faster than revenue, with no clear explanation
  • Backups that have never been restored, or a disaster recovery plan that exists only on paper
  • Production environments that nobody could rebuild from scratch

5. Team and Key-Person Risk

What to ask

  • Who could explain each critical system in depth? Is there more than one person for each?
  • How much of the development is done by contractors or outsourced suppliers, and on what terms?
  • What is staff turnover in the technical team, and why have people left?
  • Is the system documented well enough for a new engineer to become productive?

Red flags

  • One person who holds the knowledge of a business-critical system and has no incentive to stay
  • Key development done by a supplier with no contractual obligation to hand over code or knowledge
  • No documentation beyond the code itself

6. IP and Open-Source Licensing

What to ask

  • Do employment and contractor agreements assign IP in the code to the company?
  • Is there an inventory of open-source components and their licences?
  • Is any code licensed in from third parties, and do those licences survive a change of control?

Red flags

  • Code written by founders or contractors before incorporation, with no assignment
  • Copyleft-licensed components (such as GPL) combined with proprietary code in a way that could oblige the company to release its source
  • No licence inventory at all

7. Data and Privacy

What to ask

  • What personal data is held, where, and on what lawful basis?
  • Where is data stored and processed geographically, including by sub-processors?
  • How are data subject requests and breaches handled? Have there been any?
  • Is production data copied into development or test environments?

Red flags

  • Live customer data in developer laptops or test systems
  • No record of processing activities or data map
  • Data transfers outside the UK or EEA with no clear safeguards

8. AI Usage

AI now appears on both sides of the product: in what it sells and in how it is built.

What to ask

  • Which AI models and services does the product depend on, and on what commercial terms?
  • What customer data is sent to AI providers, and is that covered by customer contracts?
  • How much of the code is written with AI coding tools, and what review and testing does it go through?
  • Could a change in an AI provider’s pricing or terms break the business model?

Red flags

  • “AI-powered” claims that turn out to be a thin wrapper around a third-party API, with no defensible IP
  • AI-generated code shipped without human review, automated testing or security scanning
  • Customer data sent to AI services in breach of customer agreements

AI-built code has its own characteristic weaknesses. We describe them in Vibe Coding Has No Building Control. Where AI is central to the investment case, our AI due diligence looks at models, data, provider dependence and evaluation in more depth.

9. Microsoft 365 and the Internal IT Estate

The product is not the only technology a business runs on, and internal systems are easy to overlook.

What to ask

  • Who administers Microsoft 365 (or Google Workspace), and how many global administrators are there?
  • What business processes run on SharePoint, Power Apps, Power Automate, spreadsheets or Access databases?
  • How is access to sensitive files controlled, and when were permissions last reviewed?
  • Are any internal servers or applications running on software that is out of support?

Where the deal depends on the target’s Microsoft 365 tenant, or on separating it from a parent’s, see our Microsoft 365 due diligence.

Red flags

10. Scalability and Performance

What to ask

  • What is the current load, and what is the largest load the system has handled?
  • Has the system been load tested, and where did it fail?
  • What are the known performance problems, and what do customers complain about?

Red flags

  • Performance problems fixed by adding hardware, with the cause left in place
  • A database design that will not survive the growth in the investment case
  • Bug reports about slowness that keep recurring

11. Roadmap and Technical Debt Plan

What to ask

  • What is on the technology roadmap for the next 12 to 24 months, and what does it cost?
  • How does the roadmap support the commercial plan?
  • What would the team fix first if it had the time?

Red flags

  • A roadmap that depends on a rebuild nobody has costed
  • A commercial plan that assumes features the technology cannot deliver
  • Remediation work that the management team has not budgeted for

Red-Flag Review or Full Due Diligence?

Early in a process you rarely need every line answered in depth. The question is simpler: is there anything here that should stop us, or change the price?

Our Red-flag Review answers that, quickly. If you go ahead, a Full Technical Due Diligence works through the areas above, goes deepest where the deal value sits and ends with graded findings and a board-level presentation, for £30,000 plus VAT in about three weeks.

See how our technical due diligence works, or read how we independently reviewed a £3M low-code project that had gone wrong.

If you are selling a business, our sell-side technology readiness review works through this list for you before a buyer does.

Want engineers to work through this list for you?

A Red-flag Review gives you the biggest technical risks in 5 to 10 working days for £12,000 plus VAT, credited against a Full Technical Due Diligence if you proceed.