Brand promotion: this article introduces Guangsuan's SEO measurement guidance and Google SEO service. It is a vendor introduction, not an independent review.
Short answer: no single SEO metric decides the next fix. The nine groups answer different questions, so a change in one number is evidence about that number only. Before you rewrite, keep the observation conditions fixed, then split by country, device, page and query group to decide whether to fix the title and description, the page content, or access and indexing.
Which metric answers which question
These nine groups of metrics answer different questions. Use this map to avoid editing the wrong layer. For full definitions, see Guangsuan's 9 SEO metrics guide.
1. Discovery in search
Search Console shows visibility and selections on Google Search. GA4 shows observable activity after people reach your site. Time zones, scope, consent, successful tag execution and canonical attribution can all create differences, so do not force the two tools to agree. For session-based organic acquisition, use the session grouping in Traffic acquisition and keep first-visit grouping separate, because they answer different questions.
Query coverage and position need the same care. GSC average position is an impression-based aggregate for the selected scope, not a fixed live position seen by every searcher. Group queries by product, application and purchasing question, and record language, country, device, branded versus non-branded, and landing page.
2. Selection in results
CTR is selections divided by shows. As a hypothetical teaching illustration only, 20 selections from 1,000 shows is 2%. This shows the arithmetic and is not a pass mark or client result. Compare only within matching intent, position band, device, market and page. More low-position shows can pull the combined rate down while total selections rise, so inspect similar segments before rewriting a title. Write the title to state the page answer accurately and describe genuine content in the description. Do not add an unverified certificate or performance figure to win selections.
3. Arrival and qualified result
A chat tap, completed form, download, basket move and paid order are different outcomes. Test the actual event implementation, then separate tests, duplicates, spam and genuine business enquiries in your own records. GA4's session key event rate uses sittings with a key event divided by sittings. Counting events divided by sittings is a different ratio when one sitting holds several events. Label the calculation you actually use. Search queries as a rule cannot be joined one by one to a single order.
Bounce and engagement also need the GA4 definition for your property. Check measurement first. A one-page visit can still answer the reader question. Relate focus time and later steps to task completion. Needing four pages to find a specification is not necessarily better than finding it on one.
4. Eligibility and trust signals
Maintain a defined set of canonical pages intended for indexing and prioritise important pages within it. Sites contain login pages, duplicate parameter copies and content deliberately kept out, so covering every discoverable URL is not a useful aim. Group Page indexing report entries by their reasons and inspect representative URLs. The inspection tool distinguishes the stored version from the present live check. Requesting indexing does not force acceptance. Fix evidenced blocks, wrong canonical targets, unavailable pages or missing rendered material. “Crawled — currently not indexed” does not by itself diagnose thin copy, and a correctly consolidated duplicate is not by itself a removal task.
For links, total links and total referring sites are different counts. Discovery coverage, refresh timing and de-duplication differ across crawlers, so a missing row in one report does not prove a live link vanished. Reconcile source URL, target URL, anchor, attribute, availability and check time against the delivery list where a link service is used. Going live, getting indexed and moving the target page are separate milestones.
For third-party scores, Moz Domain Authority is a vendor predictive metric, not a public score assigned by Google. Scores from different vendors come from different models and are not interchangeable. Compare peers inside the same tool and time window while also reading actual pages and outcomes.
Separate reporting lag from real drop
If Search Console looks flat while analytics looks almost empty, repair tags, consent handling, streams and filters before judging demand. Note property, dates, search type, page, query, country and device on each export. Leave absent data marked absent rather than filling it with another benchmark.
Your site is not the global average. For September 2026 worldwide, one checkable device-share report shows mobile 60.04% and desktop 39.96%. That is a share of web traffic by device, not a statement about your industry, country or buyer, and it moves by country and year. The only figure that should drive your work is your own Search Console device report.
Apply the device filter and look at your top queries and landing pages separately for desktop and mobile. Decide which question you are asking: “where are we seen” (shows) or “where do people act” (selections). Report both. If the sample is only a few dozen shows, treat the rate as noise. Record the date of any external statistic you cite.
Hypothetical teaching numbers only, not a client result: in one fixed scope, mobile 600 selections, desktop 300 and tablet 100 means a 60% mobile selection share inside that scope. This describes only that scope and does not prove mobile is 60% of orders.
Pre-rewrite checklist and when to ask service
Use matched segments to pick the first action. The Guangsuan method groups this as: check recording, check eligibility, compare close segments, then join visits to checked enquiries.
| What you see | Check first | Sensible next step |
|---|---|---|
| Console steady, analytics almost empty | Tags, consent, stream and filters | Restore reliable recording before judging demand |
| Shows up, combined rate down, selections up | Shift in query and position mix | Compare close segments before rewriting titles |
| Selections up, checked contacts flat | Intent, market, forms and qualification | Review product path and true enquiries |
| Important canonical page not indexed | Access, signals, content and discovery paths | Fix evidenced problem and record follow-up |
- Fix scope: one search type, one complete date range, fixed country and device, brand kept apart from non-brand. Open Pages to find URLs where selections or rate fell, then open queries for that page with shows, selections and position together.
- If shows are high but selections are low on the same query, page, country, device and position band, inspect the actual title and summary shown in results before rewriting the body.
- If the query asks for model, delivery, price or compatibility but the ranking page only explains concepts, the wrong page is ranking. Add product or comparison material first.
- If checks point past one page — template canonical, robots, consent, or whole-site eligibility — bring site address, target country and language, enquiry pages, what changed and when, who can approve code or copy, and how you qualify a contact. Where this fits is described in the Guangsuan Google SEO service.
Record dates, filters, the change made, its owner and what remains unverified.
FAQ
Should flat shows trigger a rewrite? No. Flat shows mean the page is seldom offered for the query. Confirm eligibility, entry links and whether the page type matches what results favour for that task.
Selections up but no contacts — is wording wrong? Not always. That shape points past selection to page type, offer, qualification rules and path to contact. Keep selections and business-confirmed contacts on separate lines.