Editorial draft available to read. Factual and language review is pending. Examples are illustrative.
Define the exact claim
Specify measure, period and observed population. Distinguish implemented, observed and demonstrated causal effect. Technical delivery can be verified while commercial impact remains uncertain. Different evidence levels require different wording.
Consider alternative explanations
Check concurrent changes in campaigns, pricing, demand, staffing and measurement. Review comparable periods and missing data. If effects cannot be isolated, state the limitation. An observation remains useful when enough context is supplied.
Correct case-study wording
Teaching example: enquiries rise after redesign while advertising spend also changes. Describe both and the period without attributing everything to design. Verify source, permission and calculation. Another person should reproduce the figure and recognise remaining hypotheses.
Key takeaways
- Label the evidence level and avoid turning correlation into a promise.