A practical walkthrough
Turn evidence into one useful page.
A six-week sprint gives your team a sequence to review. It does not promise a ranking or customer count. Start with a customer question, record what you know, and choose a change you can measure.
1. Choose the business and the question
In Business context, describe the service, audience, market and goal. For online-only SerpCalls, a useful question is “How do I decide what to fix after a ranking change?” Choose Online only; the plan then focuses on website and measurement work.
A home-services team might instead start with a service page that customers use to check coverage and request help. Choose In person: premises or service area only when that describes the business. It does not certify eligibility for a Google Business Profile.
2. Import a comparable search window
In Search Console, select the correct property and a complete 28-day window. Export Queries or Pages with Clicks, Impressions and Position. In Evidence & opportunities, enter those exact dates and the report URL, then paste that single CSV export. Keep query and page exports separate.
For new demand, exclude your brand names and spelling variants in the report. Record the property, search type, country/device filters and extraction date in a measurement note. Missing access is unknown, not zero. Query exports omit some searches, so their row totals need not equal the report totals.
3. Add proof with its limits
Add a Pages evidence note with its source URL, observation date and Sample or Complete coverage. The John Owens example below is a sample of matched searches with recorded positions, not every tracked search. Include losses as well as gains.
Use the audit brief to carry that evidence into the page review. An observed ranking change can suggest a question to investigate; it cannot establish why the change happened.
4. Assign a change and a follow-up
In Week 3, choose an existing page whose intent matches the question. Check its title, service detail, customer proof, mobile contact path and relevant internal links. Record the owner, URL, proposed change and review date in the task notes.
After an approved release, compare an equal window with the same filters. Record qualified enquiries and paid customers separately from clicks. A completed task, a lead and a paid organization are different outcomes.
Real customer evidence
John Owens: recovery with gains and losses
From February 18, 2026 to August 24, 2026, 169 matched organic keyword–city pairs had recorded positions in both completed runs. Grid rows were excluded. This is historical scan evidence, separate from Search Console traffic.
- Average recorded position
- 12.3 → 10.8
- Returned to the top ten
- 40 pairs
- Fell out of the top ten
- 23 pairs
A return means the pair also had a recorded top-ten position before the baseline. Top-three placements fell from 25 to 20. The cohort excludes pairs missing a position in either comparison run; it does not prove that SerpCalls caused the changes or predict another business’s result.
Read the full customer story and methodologyKeep the evidence separate.
The planner prioritizes imported rows with at least 100 impressions and average position 4–20. That is a starting filter, not a forecast. Average position is aggregated search data, not a live city or neighborhood rank. A sitemap URL or a successful page fetch does not establish that Google indexed it; inspect the published URL in Search Console.
Export this preview to keep your work. In a workspace, an owner or admin can save the plan for the team. Review drafts before publishing; task status records your team’s progress and does not publish a page or Business Profile post.
For the underlying measurement guidance, read Google’s Search Console performance guide and indexing guide.