The short answer
Select a reporting window that fits the question, and disclose why that period is useful. Keep comparison periods consistent and note material differences. A narrow favorable window can make ordinary variation look like a durable improvement, especially when the underlying counts are small.
A worked example
A fictional guide receives a one-day spike after a newsletter mention. A weekly report shows the spike clearly; a monthly comparison provides different context. Both can be valid observations, but neither should silently become proof of sustained search growth. The report names the newsletter event and preserves the actual period.
A practical checklist
- State the decision the report should inform and choose the window before reviewing the result. Keep the metric and timezone consistent across periods.
- Record campaigns, outages and major content or tracking changes inside each window. Explain whether the periods are genuinely comparable.
- Show absolute counts and the relevant denominator alongside changes. Add enough historical context to distinguish a short event from a repeated pattern.
What to avoid
Avoid changing the window until a positive percentage appears or comparing a partial month with a complete month without labeling it. Do not hide low sample sizes behind polished charts. The chart should communicate the observation, not manufacture confidence.
A useful follow-up
Is the longest reporting window always best?
No. Choose the period for the decision, retain important context and make its limits visible.