Skip to content
Marketing Vallée Marketing Vallée Est. 2017
SEO与数字营销

What 9 SEO Metrics Mean Before You Change a Landing Page

MAbout the authorMarketing Vallée

Brand promotion: this article introduces Guangsuan's SEO measurement guidance and services.

For a team that ships weekly experiments and is judged on pipeline, a mixed SEO report is not yet an order to rewrite. A landing page can look weak because counting rules differ, because the wrong page type is ranking, or because a key address cannot appear at all. This guide walks through Guangsuan (光算) definitions first, then gives a short triage for Marketing Vallée style readers: B2B SaaS and DTC operators who must choose between editing the page and repairing measurement.

What the nine groups count

Fix one comparison frame and keep it for the whole diagnosis: one site view, one search type, one finished date range, one query cluster, one page, one country, one device. Write down language, brand against non-brand, and the landing page on every export. The nine groups answer different questions, so a shift in one total proves only that total.

1. Visits seen in Search and visits observed on site

Search Console reports shows and picks inside Google Search. GA4 reports observable sessions after arrival. A Search Console click is therefore not the same as a GA4 session. Time zones, scope, consent, successful tag execution and canonical attribution can all create differences. If Console is steady while analytics is almost empty, repair tags, consent handling, streams and filters first, then list which landing pages, countries and devices moved.

2. Query cover and mean position

Keep language, country, device, brand split and landing page noted on every pull. Mean position summarizes the selected view, weighted by shows. Counts from outside tools depend on their own lists, markets and collection days. Group queries by product family, use case and buying question. Does the current result page favour a product, tutorial, video or local service? Review whether your page answers that task before changing copy.

3. Pick rate inside close groups

Pick rate is picks divided by shows, times 100. As a hypothetical teaching illustration with invented numbers only, 20 picks from 1,000 shows is 2%. That arithmetic is not a target for every intent. Brand, broad informational and buying queries behave differently, and fresh low-position shows can drag a blended rate down while total picks rise. Compare only matched intent, position band, device and market. State what the page answers and describe only genuine page material.

4. Contacts and ratios

A chat tap, finished form, download, basket move and paid order are separate outcomes. Test the actual event implementation, then distinguish tests, duplicates, spam and genuine business enquiries in your operational records. Name the ratio you use, because sessions with the event divided by sessions differs from events divided by sessions when one visit holds several events. Check mobile forms, contact routes and repeats. Query rows as a rule cannot be tied one by one to a single order.

5. Bounce under GA4 rules

Bounce in GA4 is the share of sessions without engagement, the reverse of engagement rate. It is not the old single-page idea. Default engagement criteria include lasting more than 10 seconds, having a key event, or at least two page or screen views; check the property's actual configuration. One answered page can satisfy the reader, while duplicated view hits from a setup fault can inflate engagement. Verify capture, then compare promise, answer and next step.

6. Focus time and depth against the task

GA4 focus time counts time with the page in front or the app in the foreground. Averages with different bases answer different questions. Needing four pages to find a specification is not necessarily better than finding it on one. Longer time can mean useful study or confusion. Read time with later pages and task completion together, then add the missing spec, unit, condition or next page. For method detail, see Guangsuan's 9 SEO metrics guide.

7. Eligibility from a canonical list

Keep a defined list of canonical addresses meant for indexing and work important ones first. Login areas, parameter copies and pages held out by choice stay outside that list. Group the indexing report by reason and open samples. The inspection tool separates the stored copy from the live test. Requesting indexing does not force acceptance. Fix only the proven block on the exact path and user-agent checked, such as accidental noindex, blocking, wrong canonical target, unavailable page or missing rendered material. A crawled-but-not-indexed note alone does not prove thin copy.

8. Links: delivery log apart from later movement

Link totals and referring-site totals are different dimensions. Cover, refresh timing and de-duplication vary by crawler, so one missing row does not prove a live link vanished. Google's Links report is not a complete current indexing test for every source page. Where outside help is used, match source, target, anchor, attribute, availability and check time against the delivery list. Going live, entering the index and later ranking movement stay three separate stages.

9. Vendor scores inside one scale

Moz Domain Authority is a vendor prediction, not a grade issued by Google. DA, Ahrefs DR and other tools' metrics come from different models and are not interchangeable units of link equity. Compare close peers inside the same tool and window while still reading real links, target pages and search outcomes. Do not tune a vendor score apart from technical repair, page material and checked enquiries.

Page edit or measurement repair first

Use one finished export with identical filters. Open the moved address, then its query list with shows, picks and position. Keep intent, band, device and market fixed.

  • If Console is steady but analytics is near empty, repair capture. Do not judge demand yet.
  • If shows rise, blended pick rate falls and total picks rise, check query and position mix. Compare narrow slices before rewriting titles.
  • If picks rise but business-confirmed contacts stay flat, check intent, market, phone form and qualification. Walk the product path and read real enquiry records.
  • If a key canonical is outside the index, check access, signals, material and discovery path. Repair the evidenced block and log a recheck with date, filters, owner and what stays open.

If the query asks for model, delivery, price or compatibility but the ranking page only explains ideas, the wrong type is ranking. Add product or comparison material and an enquiry entry before wording tweaks. Change a title or snippet-relevant passage only when the same narrow slice shows many shows with few picks and the shown title misstates the page answer. Keep missing data marked as missing.

Checklist before asking for Google SEO help

Guangsuan groups site technical work, content supply and external resources under its Google SEO work, with counts, attributes, upkeep and indexing terms checked against the chosen plan. For whole-site scope, see Guangsuan Google SEO service. Bring facts another person can repeat:

  1. Site address, target country and language on separate lines, plus affected page addresses.
  2. One finished Console export with view, search type, dates, query group, page, country and device noted; keep brand apart from non-brand and phone apart from desktop.
  3. Enquiry definition: which event is primary, how a qualified contact is defined, and how tests, repeats and spam stay separate.
  4. What changed and when, who can approve code or copy, and for indexing doubt the exact URL, inspection date, robots rule path with user-agent and canonical target.
  5. For people-first checks, see Google guidance on creating helpful, reliable, people-first content.

FAQ

Do flat shows require a rewrite? No. Flat shows mean the page is seldom offered for the query. Confirm eligibility, entry links and whether the page type fits what results favour first.

Does a falling blended rate force a title change? Not alone. Check whether new low-position queries widened the mix, then compare narrow segments.

Does longer focus prove success? No. Longer time can mean careful study or confusion. Add the missing condition or next step and check whether the task was completed.

Take the next step

Put this playbook to work on your funnel.

Bring your CAC, channel mix, and conversion data to a 30-minute Growth Diagnostic with a senior strategist. Walk away with a prioritised backlog of experiments.

Book a Growth Diagnostic

Or read the next teardown in The Vallée Lab, or explore how we rebuilt Pipedrive's paid social engine in our case studies.