Use verified Search Console data to decide where SEO work is most likely to matter.
Evidence firstUse measurable signals to decide what deserves attention, keep changes controlled and verify the result after deployment.
Begin with the correct property and period
Confirm you are reading the right domain property and a period long enough to reduce daily noise. For many sites, 28 days is useful for operational review, while longer comparisons help with seasonality.
Look beyond total clicks
Clicks alone hide useful detail. Review impressions, CTR and average position by query and page. A page with substantial impressions and a position around the first two result pages may offer a clearer optimisation opportunity than a page with no search visibility at all.
Find mismatches between query and page
Check whether the page Google is showing is genuinely the best answer for the query. If several pages appear for the same intent, investigate cannibalisation. If the right page appears but CTR is weak, review the title, description and relevance before creating new content.
Prioritise near-ranking pages carefully
Positions are averages, not fixed ranks. Treat them as directional evidence. Focus on queries with meaningful impressions, relevant landing pages and a realistic path to improvement through stronger content, internal links or technical cleanup.
Record the evidence with the action
When you create an SEO task, keep the query, page, impressions, CTR and position that justified it. That makes later review much easier and prevents the work programme from becoming opinion-led.
Practical checklist
Practical checklist
- Correct Search Console property selected
- Date range recorded
- Queries reviewed with landing pages
- Near-ranking opportunities identified
- Cannibalisation checked
- Evidence attached to each action