Begin with a complete period
Choose dates that avoid an unfinished reporting period. Note the selected property and keep the window comparable with your previous review. Maple Orbit’s comparison tool uses the immediately preceding period of the same length.
Compare my selected property’s last complete 28-day period with the preceding 28 days. State both date ranges. Report clicks, impressions, CTR and position, and distinguish observations from possible explanations.
Follow one change into the detail
If total clicks moved, request page rows for both periods. Ask the assistant to match URLs and disclose that it is comparing bounded lists. A URL missing from one result cannot safely be assigned zero clicks.
Daily rows can help locate when a change began. Then use your own change log to ask whether a deployment, an unavailable product range, or a campaign deserves a closer look. Maple Orbit does not maintain an annotation system or identify causes automatically.
Finish with one action and a check date
Record what you observed, what you believe may explain it, and what evidence would change your mind. Review the page before editing it. If you do make a change, keep its date and original version.
For a small website, a handful of clicks can create a large percentage swing. Give the next review enough data to be meaningful; a seven-day routine does not make seven days a reliable test. Do not declare that an edit worked from one extra order or click.
Keep a separate record of sales or enquiries if that is the business outcome. Search traffic is useful evidence, but it is not revenue.
Further reading
Official Google documentation. Product instructions distinguish the browser beta from the invited connector pilot. Educational guidance does not imply an automated feature.