Working method · 6 min read
A weekly SEO review that ends with one useful action
Use one consistent search comparison, choose one question supported by the evidence, and record a small action with a later recheck. A task marked complete is not proof that search performance improved.
Prepare a narrow review
Choose a complete date window and the same property, search type and filters as the previous review. Open the affected page and keep recent page edits available. If the data is incomplete, record that before selecting a task.
This routine is a suggested working method, not a tested customer outcome. It helps you make a review reproducible. It does not promise a ranking lift, a fixed completion time or a number of enquiries.
Copy this action record
Use a shared document or the existing task workflow available in your account. The record below is ordinary text; it does not schedule a job or send a reminder.
Review date:
Page or page group:
Source and export time:
Previous/current dates and days:
Search type and filters:
Observed change:
What remains uncertain:
One next check or edit:
Owner and completion evidence:
Recheck date and same-segment result:Fictional review: a change without a conclusion
An invented business sees unchanged daily clicks but more daily impressions on its service page. CTR is lower. The owner suspects a query-mix change, but has not checked the queries yet.
The next action is “compare branded and non-branded query groups for this page”, with the date ranges and filters recorded. Completion evidence is the saved comparison and a note about what it shows. A rewrite is not automatically the right next action.
| Record field | Fictional entry |
|---|---|
| Observed change | Daily clicks 10 to 10; impressions 200 to 250; CTR 5% to 4% |
| Uncertainty | More lower-CTR queries may explain the combined rate |
| Next check | Compare the same page’s query groups |
| Completion evidence | Dated export and segment notes |
| Outcome | Not yet known; review again after the next complete window |
Choose work proportional to the evidence
If an intended public page returns an error, investigate delivery before rewriting its description. If a title makes a claim the page cannot support, correct the mismatch. If only combined CTR moved, inspect the segments before treating the snippet as the cause.
A small factual improvement with a visible acceptance check is easier to review than “do SEO”. Keep separate records for implementation completion and observed business outcomes. Enquiries and sales need their own evidence.
Close the loop honestly
- Record what was changed and when it became public.
- Recheck the same segment after a complete window.
- Note seasonality, unrelated site changes and incomplete data.
- Keep an unresolved result as unresolved. Choose the next question from the evidence rather than declaring success.
Sources and editorial method
Written and maintained by Mapseen. Guidance is linked to primary documentation. Examples are fictional; they explain a calculation or decision rather than report customer outcomes. Provider behaviour and report availability can change.
Found a mistake? Open an email draft to Mapseen with the guide URL and the detail to review.