Due Diligence
In-depth audit carried out before an acquisition, investment or partnership to assess a target's risks and real value.

Due diligence is an in-depth audit and investigation process carried out before an acquisition, an investment or a strategic partnership. Its goal is to assess the target's real value, uncover hidden risks and give the acquirer what it needs to make an informed decision. According to Harvard Business Review, 70% of acquisitions fail to create the expected value, often because of insufficient due diligence.
Technical definition
The dimensions of due diligence
Due diligence is a multidimensional exercise. The financial dimension audits historical accounts, cash, projections and revenue quality (recurring vs. one-off). The legal dimension reviews contracts, intellectual property, pending litigation and regulatory compliance. The HR dimension assesses key people, attrition risk, company culture and employee-related commitments.
Technical due diligence (or tech due diligence) has become a critical dimension for any acquisition involving software. It covers: the quality and maintainability of the source code, the technical architecture and its scalability, accumulated technical debt, security (vulnerabilities, OWASP compliance), infrastructure (cloud, on-premise, hybrid), and dependency on key people.
Adapting to vibe coding
In 2026, technical due diligence has to account for a new reality: code can be generated by AI in a matter of days. Classic criteria (test coverage, naming conventions, documentation) remain relevant but are no longer enough. Assessors now need to audit the development workflow (LLM usage, level of human review), the team's ability to defend its architecture choices, and the risk of "invisible technical debt" specific to AI-generated code.
Current challenges
The tech valuation paradox
The OpenClaw case illustrates a new paradox: a project riddled with vulnerabilities (critical CVEs one after another, malicious extensions by the hundreds in its catalog) sparked a bidding war between Meta and OpenAI to bring in its creator. OpenAI, which ended up hiring him, is buying the talent and the community, not the code. This logic forces M&A professionals to rethink what counts as a "technical asset": intellectual property loses value when AI can reproduce the code, but a creator able to rally a massive community in a few weeks has unique value.
Faster deals
Competitive pressure (several AI giants bidding at the same time) squeezes due diligence timelines. Acquirers have to trade off audit thoroughness against the risk of losing the deal. This tension favors tailored evaluation grids that rank deal-breaker criteria (critical security issues, irreversible dependency) above criteria that can be fixed after the acquisition (technical debt, documentation, tests).
Standards and Specifications
ISO 31000:2018
Risk management — Guidelines, applicable to due diligence processes
AICPA Due Diligence Guidelines
US reference framework for conducting acquisition audits
Frequently Asked Questions
What are the different types of due diligence?
Due diligence covers several dimensions: financial (accounts, cash, projections), legal (contracts, litigation, intellectual property), tax (compliance, reassessment risk), HR (key people, turnover, culture), commercial (market, customers, pipeline) and technical/IT (code, infrastructure, security, technical debt).
Technical due diligence has grown in importance with digitalization, especially for tech acquisitions. It now includes assessing AI workflows and vibe coding practices within target teams.
How long does due diligence take?
A standard due diligence lasts 4 to 12 weeks depending on the target's size and complexity. The phases: scoping (1 week), data room and document collection (1-2 weeks), in-depth analysis (2-6 weeks), final report and negotiation (1-2 weeks).
For high-growth tech acquisitions (startups, open source projects), the timeline is often compressed to 2-4 weeks under competitive pressure. That acceleration increases the risk of missing critical flaws, as the OpenClaw case shows, where deal speed took precedence over the depth of the security audit.
What's changing in tech due diligence in 2026?
The rise of vibe coding is reshaping the evaluation grid. Classic criteria (code quality, test coverage, architecture) remain necessary but are no longer sufficient. You now also need to assess: the AI development workflow (which tools, what level of human review), the team's ability to explain its technical choices, the invisible technical debt of AI-generated code, and the dependency on the original creator.
Talent, traction and community now weigh as much as technical cleanliness in the valuation. A project that is hugely popular on GitHub yet riddled with critical CVEs can be worth more than a clean project with no users.