Discovery comes before the decision to build
Discovery is the stage in which a team learns about users, constraints and the problem before committing to build anything. The GOV.UK Service Manual is direct about it: you should not start building your service in discovery (GOV.UK).
What the team finds out
- Users' goals, how they reach them today and the wider journey, including offline steps.
- Accessibility needs, digital skills and internet access.
- Constraints such as legislation, contracts, legacy technology and existing processes.
- Opportunities, for example reusing data that another team already holds.
- What the problem costs today and how success will be measured.
How long it takes
There is no fixed length. The same guidance describes around 4 to 8 weeks as typical, with the purpose of the discovery setting the length, and notes that problems nobody has researched before may need longer. For a small internal tool, a much shorter, focused analysis may be enough; what matters is that the goal is written down first.
Alternatives are a valid finding
Discovery should also consider options other than new software: better content, a changed process, or an existing product. The guidance states that stopping at the end of discovery is not a failure if the research shows it is the best option. A discovery that only confirms a decision already taken is not doing its job.
How this maps to our services
Our requirements analysis service covers the discovery questions for a single product: problem, users, constraints and a prioritised backlog. The room booking demonstration shows what such artefacts look like for an invented example.
Sources
Sources checked on 17 November 2026.