Content Audit Checklist for Original Research and Evidence

Suppose a fictional software team times five onboarding walkthroughs with colleagues. "Users finish onboarding 40% faster" is too broad if there is no comparable baseline or customer sample. A more accurate heading is "What we learned from five internal onboarding walkthroughs." Describe the task, participants and observations before deciding what to change.

Write the research question, source, sample or scope, collection period, method, comparison basis and limitations. For the fictional walkthrough, note whether everyone used the same account setup and whether the timer included help from a colleague. These conditions affect what the observation means.

If you report a percentage, show the underlying counts and define the calculation. Keep collection methods and observation periods comparable. A change after a product edit is not automatically caused by the edit. When the baseline is missing, report what you observed rather than constructing one from memory.

A small study can still identify a practical issue. If several walkthroughs stalled at an unclear field, show the field and the proposed wording change, using authorized material. Explain why that change is worth testing. Do not turn the observation into a prediction about every user’s completion time.

Confirm permissions, factual authorship, the collection date, source links and the distinction between observed results and proposed actions. Keep private information out of screenshots and downloads. Add an update when the underlying evidence changes. A visible date alone does not make old findings current. End with the decision the research supports and the evidence needed for the next check.