A clearer way to start and review.

Practical guidance for choosing a question, understanding the evidence and deciding what happens next.

Start with a question you need to answer.

A useful brief describes the decision the findings will support. “Where might visitors hesitate before making an enquiry?” gives a website review a clearer purpose than “Check everything.”

Choose a use case. Compare the intended inputs, scope and output in the use-case library.
Gather the context. Identify the website, priority pages, business goal and information you are authorised to use.
Confirm the boundaries. Agree which work is included, what will be delivered and any cost before starting.
Allow time for review. Plan who will check the evidence and apply the business context before using the findings.

Find a use case

Separate what was observed from what is recommended.

A finding should let you inspect the source and decide whether the recommendation makes sense for the business. Read the observation first, then its evidence, limitations and proposed next step.

Illustrative finding · example.com/contact

Observation: The enquiry form asks for information that could be collected later.
Evidence to inspect: The captured page, the form fields and the context in which a visitor sees them.
Recommended experiment: Test a shorter first step with a clearer explanation of what happens next.
Your judgement: Confirm which fields the business needs. A shorter form is a hypothesis to test, not a promised conversion improvement.

Keep well-supported findings, correct incomplete interpretations and exclude findings that do not hold up. Specialist conclusions still need appropriately qualified judgement.

Explore the interactive report example

Coverage gaps are part of the result.

An agreed review covers a bounded set of pages, checks and actions. A discovered URL is not proof that its page was captured successfully, and a screenshot alone does not prove that a form or cart interaction worked.

Inaccessible pages: blocked, unavailable or out-of-scope pages should be identified as gaps.
Performance: available real-user measurements and lab diagnostics answer different questions. They should be labelled separately.
Conditional checks: applicability depends on the website and agreed service; a framework is not a guarantee that every check applies.
Missing evidence: an unverified area should remain unverified rather than being presented as a pass.

Read the coverage and limitations alongside the findings when deciding whether more investigation is needed.

Review before sharing the conclusion.

Use your business knowledge to prioritise the findings and add practical recommendations. For client delivery, the responsible reviewer should approve the report and check its branding, scope and limitations.

PDF and Word outputs depend on the agreed service. Review the final document as well as the findings, and keep the evidence behind important recommendations available.

Fabric prepares the groundwork. People remain responsible for the advice, decisions and actions that follow.

Read about agency delivery