Skip to content
MoeLink — Field Notes

What 9 SEO Metrics Mean Before You Change a Product Page

For a performance marketing team: what the 9 SEO metric groups each mean, how to fix country and device scope, and how to choose between snippet, page content, or indexing checks before editing a product page.

MMoeLink SEO与数字营销

Brand promotion: this article introduces Guangsuan's SEO measurement guidance. It is a vendor introduction, not an independent review.

A performance team that lives in paid dashboards can misread organic numbers before a product-page edit. Paid clicks reconcile to spend in one system. Organic readings come from separate systems with separate definitions. This guide explains the nine metric groups in plain order, then shows how to lock scope by country and device before you decide what to change. Method definitions follow Guangsuan's 9 SEO metrics guide. Where repair spans templates or many URLs, whole-site scope is described in the Guangsuan Google SEO service.

Direct answer for a team about to edit a PDP

The nine metric groups answer different questions, so a move in one group does not prove the whole page improved. Discovery tells whether the page was shown and selected. Selection tells whether the snippet earned the click among similar results. Arrival and qualification tell whether visits produced contacts your team accepts. Eligibility and trust tell whether the page could be indexed and whether supporting signals were recorded correctly. Lock search type, a complete date range, country, device, and brand separation first, then read those four layers in order.

Fix the reading conditions first

Use one property, one search type, and one finished date range for every file. An unfinished recent day cannot be compared with a finished historical day. Keep language, country, device, brand status and landing page recorded for each export. Group queries by product, use case, and buying question rather than sorting one large mixed list by clicks.

GA4 shows observable activity after arrival on the site. Search Console shows shows and selections on Google Search. Zones, coverage, consent, tag firing, and which canonical received credit can separate the two totals, so do not require them to match. For session-based acquisition, keep session grouping apart from first-visit grouping because they answer different questions.

Average position is a show-weighted aggregate for the view you selected. It is not one live rank every searcher saw. A third-party keyword count depends on its database, market, and collection dates, so two tools can report different counts without either being wrong.

What the nine groups mean on a product page

1. Discovery: shows, selections, queries, position

Shows count when a normal web result enters the results page a user opened. A later page that was never opened does not earn a show only because the query occurred. Selections count when that result is chosen. Locate the landing pages, countries and devices driving the change. Verify recording before judging seasonality, availability, ranking, or copy. A smaller organic share can also mean paid spend grew.

For queries, keep conditions fixed. Look at representative queries that fell or that offer a useful opening. Ask whether the current results favour a product page, tutorial, video, or local page, then check whether your page serves that same task. Do not copy markup seen elsewhere only because it appears in source.

2. Selection: rate compared only among likes

Selection rate is selections divided by shows, multiplied by 100%. As a hypothetical teaching illustration only, with invented numbers, 20 selections from 1,000 shows is 2%. This line shows the arithmetic. It is not a target, not a client outcome, and not a pass mark for every intent or layout.

Compare only inside matching intent, position band, device, market, and page. More low-position shows can pull a combined rate down while total selections rise. If the same query, page, country, device, and position band has many shows but few selections, read the title and summary actually shown before rewriting the body. State the product type and task honestly and describe true page content. Do not insert an unproven certificate, speed figure, or benefit claim to win selections.

3. Arrival and qualification: sessions, events, accepted contacts

A chat tap, finished form, download, basket move, and paid order are different results. Test the event on the live page. Then separate tests, repeats, spam, and real contacts in your own log. Queries as a rule cannot be tied one by one to a single order.

Label the calculation you actually use. One ratio uses sittings with a key event divided by sittings. Counting events divided by sittings is a different ratio when one sitting holds several events. Check the mobile form, contact route, and de-duplication. Review landing-page sittings, sittings with the wanted event, and business-confirmed contacts together.

4. Engagement: bounce, focus time, and depth tied to the task

Use the GA4 definition set in your property. Bounce is the opposite of engagement under that definition. It is not simply the old single-page-session definition. Default engagement criteria include lasting more than 10 seconds, having a key event, or at least two page or screen views; check the property actual configuration.

A one-page visit may answer the reader question. Duplicated page views can make engagement look artificially strong, so check recording first. GA4 focus time measures time with the page in focus or the app in the foreground. Needing four pages to find a specification is not necessarily better than finding it on one. Longer time can mean useful study or confusion. Interpret time, later pages, and task completion together. Add the missing specification, condition, example, or next step rather than adding headings or images by quota.

5. Eligibility and trust: indexing, links, third-party scores

Keep a defined set of canonical pages you plan to have indexed, and work through important pages first. Sites hold login pages, parameter copies, and material 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. Requesting indexing does not force acceptance. A crawled-but-not-indexed state alone does not prove thin copy, and a correctly consolidated duplicate is not by itself a removal job. Fix proven blocks such as accidental noindex, blocking, wrong canonical targets, unavailable pages, or missing rendered material.

Total links and 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. Google's Links report is not a complete current indexing test for every source page. Where a link service is used, reconcile source URL, target URL, anchor, attribute, availability, and check time against the delivery list. Going live, getting indexed, and later movement are separate milestones.

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.

Use country and device to pick the first repair

Your own Search Console device report should drive the work. For September 2026 worldwide, StatCounter reports mobile 60.04% and desktop 39.96% as a global traffic share. That figure describes worldwide traffic by device. It does not describe your industry, country, or buyer, and it moves by country and year.

Worked check your team can repeat: pick one product-page group, one country, and one device. Open Pages to find a URL where selections fell. Open its query list with shows, selections, and position together. Mark new, existing, and irrelevant queries. A row with very few shows can swing on one click, so do not sort by lowest rate alone.

As a hypothetical teaching illustration only for one fixed scope, mobile 300 shows with 9 selections is 9 divided by 300, or 3%. Desktop 300 shows with 21 selections in the same scope does not by itself prove desktop copy is better. Check whether mobile queries ask for fast specs while desktop queries ask for comparison, and whether the mobile title is cut off or the spec table is hard to tap. For September 2026 worldwide, StatCounter reports mobile 60.04% and desktop 39.96% as a global traffic share, but that global share never replaces the split in your own report.

Matched-segment actions

  • Console steady, analytics almost empty: check tags, consent, streams, and filters. Restore reliable recording before judging demand.
  • Shows up, combined rate down, selections up: check whether new low-position queries widened the mix. Compare close segments before touching titles.
  • Selections up, accepted contacts flat: check intent, market, forms, and qualification rules. Review the product path, shipping, and contact route.
  • Key product page not indexed: check access, signals, content, and discovery. Fix the proven block and log follow-up.

Record dates, filters, the change made, its owner, and what remains unverified. Leave missing data marked as missing rather than filling it with another benchmark.

What to prepare before asking Guangsuan

Bring facts another person could use to repeat your reading: site address, target country and language, affected URLs, one complete date range with filters, what changed and when, who can approve code or copy, and how you qualify a contact. Note tags, consent handling, and report filters if analytics looks almost empty while Console shows data. If checks point past one page, such as template canonical, a robots rule for an exact path and user-agent, consent breaking capture, or whole-section eligibility, that is the point to ask about technical repair, content supply, and external resources together.

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, then decide.

Selections are up but accepted contacts are flat. Is the title wrong?

Not necessarily. That shape points past selection to page type, offer detail, form ease, and qualification rules. Keep visits and business-confirmed contacts on separate lines while you check the mobile form and reply path.

See every redirect as a revenue event.

Book a live walkthrough of the MoeLink attribution dashboard. We'll route a real link through your stack in the call.

Get a live demo