SEO Metrics First: Which Number Decides the Next Fix
Brand promotion: this article introduces Guangsuan's SEO guides and services.
Your paid dashboard looks healthy, yet add-to-cart and enquiry counts do not follow. Before you rewrite a product page, stop and separate the readings. Search visibility, arrival on site, task completion, and confirmed business results are recorded in different systems under different rules. This guide shows a DTC and B2B paid-media team how to read those layers in order, then pick one page to repair first.
Lock one comparable scope before any comparison
Pull every export from the same property, same search type, and one finished date range. Leave out the incomplete last day. Write down language, country, device, whether the query names your brand, and which page it lands on. Group queries by product family, use case, and purchase question. Save filters, export date, and whether data are complete so another person can repeat the check.
Search Console counts shows and picks inside Google Search. GA4 counts what could be observed after arrival. Time zone, coverage, consent, tag firing, and canonical credit can split the two totals, so do not force them to match. For session-based acquisition, keep session grouping apart from first-visit grouping because each answers a separate question.
Discovery: were you shown, and for what queries?
Shows count when a normal result entered a results page the searcher actually opened. Picks count when that result was chosen. Average position is a show-weighted summary for the view you selected. It is not one live rank every searcher saw. An outside keyword tally rests on its own list, market, and collection dates, so two tools can differ without either being broken.
Find the pages, countries, and devices behind the shift. Confirm capture before you judge season, stock, ranking, or copy. A smaller organic share can also mean paid spend grew. Open a falling or promising query, note whether the live results lean to product, guide, video, or local format, then ask whether your page serves that same task. Do not copy markup seen elsewhere only because it appears in source.
Selection: did the snippet earn the pick?
Pick rate is picks divided by shows, times 100. As a hypothetical teaching illustration only, with invented numbers, 20 picks from 1,000 shows equals 2%. That line shows arithmetic. It is not a target, not a client outcome, and not a pass mark for every intent.
Compare only inside matching intent, position band, device, market, and page. Added low-position shows can drag a blended rate down while total picks climb. Where the same query, page, country, device, and band shows many times but is seldom picked, read the title link and description actually shown before you touch the body. State product type and task plainly and describe real page matter. Do not add an unverified certificate, speed figure, or benefit claim to win picks.
Arrival and enquiry: did visits become accepted contacts?
A chat tap, finished form, download, basket action, and paid order are separate outcomes. Test the live event, then separate tests, repeats, spam, and real enquiries in your own log. Name the formula in use. One ratio uses visits with the wanted event divided by visits. Events divided by visits is a separate ratio when one visit holds several events. As a rule, queries cannot be tied one by one to a single order.
Read landing-page visits, visits with the wanted event, and business-confirmed contacts together for the same page and range. Check the phone form, contact route, and de-duplication. Keep absent data marked absent rather than filling it with another benchmark.
Task fit: did the visit solve the job?
Use the GA4 meaning fixed in your property. Bounce is the opposite of engagement under that definition. It is not the old one-page idea. The default engaged case can cover staying past a short time threshold, completing the chosen event, or viewing at least two pages or screens. Confirm how your property is set, because defaults can be changed. A one-page stay can satisfy the need. Repeated view hits from a setup fault can inflate engagement, so verify capture first.
GA4 focus time captures time with the page in front of the reader or the app in use. Four pages to locate a spec is not better than one. Added length can mean study or confusion. Link time, later pages, and completion together. Supply the missing spec with units, applicable condition, worked check, or next step rather than adding headings or images by quota.
Eligibility and support: could the page be found and trusted?
Keep a short list of canonical URLs you want indexed and work on the important ones first. Login areas, parameter copies, and matter held out by choice remain in crawl space, so coverage of every discovered address is not a useful aim. Group the indexing report by reason and open samples. Asking for indexing starts a process. It does not force entry. Repair proven blocks such as unintended noindex, blocking, wrong canonical target, unavailable page, or missing rendered matter, then log date, filter, edit, owner, and what still needs recheck.
Total links and referring sites are separate counts. Coverage, refresh timing, and de-duplication vary by crawler, so an absent row in one report does not show a live link vanished. A links report is also not a live check that each source page is currently indexed. Where link help is used, match source address, target address, anchor, attribute, availability, and check time against the delivery list. Going live, entering the index, and later movement are separate stages. Third-party authority scores come from separate vendor models and are not interchangeable. Compare peers inside the same tool and time window while reading real pages and outcomes.
Whose device and country are you fixing?
For September 2026 worldwide, StatCounter reports mobile 60.04% and desktop 39.96% as a global traffic share. That dated figure describes worldwide traffic by device. It does not describe your industry, country, or buyer, and it moves by country and year.
Work only from your own Search Console device report. Apply the device filter and look at top queries and pages separately for desktop and mobile. Decide which question you ask: where you were seen by shows, or where people acted by picks. Report both. A row with very few shows can swing on one pick, so do not sort by lowest rate alone. As a hypothetical teaching illustration only for one fixed scope, mobile 300 shows with 9 picks is 3%, desktop 300 shows with 21 picks in the same scope does not by itself show desktop copy is stronger. Check whether mobile queries seek fast specs while desktop queries seek comparison, and whether the mobile title truncates or the spec table is hard to tap.
Fix order you can run this week
- Console holds data, analytics near empty: check tags, consent, streams, and filters. Restore reliable capture before judging demand.
- Shows up, blended rate down, picks up: check whether new low-position queries widened the mix. Compare close segments before editing titles.
- Picks up, accepted contacts flat: check intent, market, offer detail, form ease, and qualification rules. Walk the product path and reply route.
- Priority canonical outside index: check access, signals, matter, and discovery path. Fix the proven block and log recheck.
Method detail follows Guangsuan's 9 SEO metrics guide. For device-split method, see Guangsuan's device traffic comparison. If checks point past one page, such as a template canonical or a robots rule for an exact path, bring site address, target country and language, affected URLs, one completed range with filters, what changed and when, who can approve code or copy, and how you qualify a contact. That is the point to ask about technical repair, content supply, and outside 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 page type matches what results favour for that task, then decide.
Picks are up but accepted contacts are flat. Is the title wrong?
Not necessarily. That shape points past picking 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.
Ready to turn this into pipeline?
Book a 30-minute paid strategy audit with a senior strategist. We'll map your current spend, find the leaks, and model the 90-day lift.
Get My Free Strategy Audit