GA4 and Search Console: Tracking the SEO KPIs That Actually Matter

For years, SEO reporting meant a ranking-tracker screenshot and a Sessions number pulled from whatever dashboard happened to be open, both dressed up with green up-arrows. Rankings climbed, traffic climbed, and everyone assumed the strategy was working. But position 3 for a keyword nobody converts on is worthless, and a 40 percent jump in sessions from a viral listicle that bounces in nine seconds tells you nothing about revenue. Vanity metrics, raw traffic, keyword rankings viewed in isolation, impressions without context, are seductive because they are easy to screenshot and instantly gratifying. They rarely say anything about whether SEO moved the business forward.

Google Analytics 4 and Search Console, used correctly and linked together, tell a different story: not did traffic go up, but did the right visitors land on the right pages, engage with the content, and complete an action tied to revenue. Since Universal Analytics fully sunset in mid-2024, GA4’s events-based model and updated key events terminology have changed how SEO KPIs should be framed, and most teams are still reporting on the old bounce-rate, sessions-only mental model. This guide walks through setting both platforms up correctly, the reports that actually matter, how to build a defensible KPI dashboard, using GA4 Explorations for custom cuts of the data, catching content decay before it costs rankings, attribution nuance for organic, and the measurement mistakes that quietly waste SEO budget every month.

Setting Up GA4 and Search Console Properly for SEO Tracking

Nothing in this article works if the underlying data is wrong, and the setup mistakes here are more common than most audits admit. Start with the GA4 property itself. Confirm the data stream is tied to the actual production domain, not a staging environment left connected from an old redesign, and that internal traffic, the office IP range, the agency’s IPs, the client’s own team, is excluded through a data filter rather than just noted as known noise. An agency running its own SEO checks all day can otherwise inflate engaged sessions by five to ten percent without anyone noticing.

Next, link GA4 to Search Console. This is done inside GA4 under Admin, Property Settings, Search Console Links. It requires Owner access on the Search Console property and Editor or Administrator access in GA4, and it pulls query, click, impression, and position data into GA4’s Search Console reports and into Acquisition. Linking does not retroactively backfill history, data only flows from the link date forward, so this should be one of the very first things configured on a new site rather than something fixed six months into a campaign when someone finally asks for organic query data inside GA4.

On the Search Console side, decide between a Domain property, verified through DNS, which aggregates http, https, www, and all subdomains into a single view, and a URL-prefix property, which is locked to exactly one protocol and host. For SEO KPI tracking, a Domain property is almost always the right choice, because it prevents traffic and query data from being artificially split across three or four separate properties for what is really one website. If a site already has fragmented URL-prefix properties left over from a past migration, add a Domain property alongside them rather than trying to merge the history. Finally, confirm key events are configured in GA4 before you need them for a report, not after a stakeholder asks for organic conversion numbers and nothing has been tracking.

Core Search Console Reports Every SEO Should Master

Search Console is the only place that shows what happened in the search results themselves before a click ever occurs, which makes it the right starting point for any why-isn’t-organic-growing question that GA4 alone cannot answer.

The Performance report is the backbone: clicks, impressions, average CTR, and average position, filterable by query, page, country, device, and search appearance. The metric most teams misread is average position. It is an average rank across every impression for a query, including page-four rankings that quietly drag the number down, so a page can show an average position of 8 while actually ranking number 2 for its primary term and number 40 for a dozen tangential ones. Segment by query before drawing any conclusion from position alone. CTR is the more actionable number here: a page ranking well but with CTR meaningfully below the norm for that position usually points to a weak title tag, missing schema markup, or a competitor with a more compelling snippet.

The Index Coverage report, now labeled the Pages report, shows which URLs are indexed, excluded, or carry errors, broken down by reasons such as Crawled – currently not indexed or Duplicate without user-selected canonical. This report is a technical-health leading indicator; a rising not-indexed count across a category of pages usually shows up as a traffic drop inside GA4 two to six weeks later.

The Core Web Vitals report groups URLs by status, Good, Needs Improvement, or Poor, across LCP, INP, and CLS, using real Chrome User Experience Report field data on a rolling roughly 28-day window. INP replaced FID as the responsiveness metric back in 2024, so any legacy dashboard still referencing FID is stale and should be updated. The Links report, external and internal, rounds this out. It will not replace a dedicated backlink tool, but a sudden drop in referring domains, or internal link concentration accidentally starving a priority page of internal links, is worth a monthly check.

Core GA4 Reports That Matter for SEO

Once Search Console shows what happened in the search result, GA4 shows what happened after the click. Start in Reports, Acquisition, Traffic acquisition, filtered to the Organic Search channel under session default channel grouping, to see sessions, engaged sessions, engagement rate, and average engagement time by channel. Engaged sessions, a session lasting ten seconds or more, containing a key event, or containing two or more page views, replaced the old bounce-rate framing, and engagement rate is a genuinely more useful content-quality signal than bounce rate ever was, because it rewards real reading and interaction instead of penalizing a legitimate single-page visit, a common and healthy pattern for a blog post that fully answers its query on one page.

From there, filter key events, formerly called conversions and now configured under Admin, Events, mark as key event, by Organic Search, to see which organic sessions actually completed something the business counts: a demo request, an add to cart, a newsletter signup. Cross-reference this against the Monetization, Ecommerce purchases report on any site that sells directly, so organic gets tied to actual revenue rather than just raw event counts.

Landing page performance sits under Engagement, Landing page, or can be built as an Exploration, showing which URLs act as the entry point for sessions. Cross-referencing this with the linked Search Console report, page by page clicks and impressions next to engaged sessions and key events for the same URL, is usually the fastest way to find pages that rank and get clicked but fail to convert, versus pages converting well that deserve more content investment or additional internal links.

Building a Practical SEO KPI Dashboard

A KPI dashboard only earns trust if it can survive someone asking why about every number on it. Build it around four categories instead of one long list of metrics.

• Traffic: organic sessions and engaged sessions by month, plus a rolling ninety-day trend rather than noisy month-over-month swings, split into branded versus non-branded where possible using a regex-based segment on landing query or source, since GA4 does not split this natively.
• Visibility: Search Console clicks, impressions, and average CTR trended monthly, plus tracked rankings for a curated set of twenty to forty commercially important terms rather than the entire keyword universe, which is mostly noise.
• Conversions and revenue from organic: key events attributed to the Organic Search channel, plus ecommerce revenue or assigned lead value where a value has been set on those events.
• Technical health: Index Coverage errors, Core Web Vitals status distribution, crawl stats trended from Search Console’s crawl stats report, and any manual actions on record.

Keep the whole thing to one page. A stakeholder should be able to read it in ninety seconds and know whether SEO is working, not spend fifteen minutes parsing forty tabs of a spreadsheet.

Using GA4 Explorations for Custom SEO Analysis

The standard reports answer the questions Google anticipated. Explorations answer the questions specific to a given business. A free-form exploration with session default channel group as a dimension, landing page as a secondary dimension, and engaged sessions, key events, and total revenue as metrics, filtered to Organic Search, produces the single most useful SEO table in GA4: every organic landing page next to exactly how well it performs, sortable by whichever metric matters that month.

A funnel exploration is worth building for the real organic buyer journey, a landing page view step, a product or pricing view step, then a key event step, to see precisely where organic visitors drop off compared with visitors from other channels, a sharper diagnostic than engagement rate alone. A cohort exploration segmented by first-session Organic Search traffic, tracked across subsequent weeks, is useful for content-led or non-ecommerce sites where the first visit rarely converts; it shows whether organic-acquired users return and eventually convert, something a single-session view misses entirely.

Segmenting by Landing Page, Query, Device, and Country

Aggregate numbers hide almost everything useful. The same dashboard broken down four ways answers four different questions.

• By landing page: which specific URLs actually carry the organic program, and whether a page-level change, a title rewrite, a content refresh, a new internal link, moved clicks and key events for that URL specifically.
• By query, inside Search Console: whether growth is coming from head terms already ranking well, or from long-tail queries expanding as topical authority grows, the latter being a healthier growth signal than the former.
• By device: mobile usually carries the majority of organic clicks for most content sites, but engagement rate and key event rate are often meaningfully lower on mobile, a gap worth investigating rather than accepting as normal, since it frequently traces back to a mobile-specific UX or Core Web Vitals issue.
• By country: important for any site with international content, since a page can look average in aggregate while performing very differently across search markets; a page ranking number 2 in the UK and number 35 in the US behind local competitors needs a different fix than the blended number suggests.

Measuring Content Performance and Catching Decay

Every content library decays. A page published two years ago and never touched will, on average, lose rankings and clicks as competitors publish more current material and as the query’s intent shifts, a best-of-2024 title alone becomes a decay signal by 2026. Catching this before it becomes a crisis means checking it on a schedule rather than waiting for someone to notice a traffic drop.

Build a simple decay report: pull Search Console clicks and impressions for all URLs over the trailing six months versus the prior six months, using the Performance report’s built-in date comparison, sort by the largest percentage decline in clicks, and cross-check each candidate against GA4 engagement rate for the same page to rule out pages that are simply seasonal rather than genuinely decaying. Pages showing a real decline, not seasonal, not a SERP feature change stealing the click, go on a refresh list: updated facts and examples, a rewritten intro where the framing has gone stale, added sections for related queries the page is now getting impressions for but not ranking on page one, visible in the Search Console query list for that specific page, and in some cases a canonical merge with a newer competing page on the same site.

Attribution: When Organic Assists Conversions Elsewhere

Organic search’s job is not always to close the conversion. A meaningful share of organic sessions research a topic, leave, and convert later through branded search, direct traffic, or email, a pattern GA4’s default channel grouping under-credits organic for if the only lens used is last-click key events. Two things help correct for this without over-engineering an attribution model.

First, compare last-click to data-driven attribution for the Organic Search channel inside GA4’s Advertising, Attribution reports. Data-driven attribution, the default for new key events since Google phased out last-click as the standard option in October 2025, frequently shows organic contributing meaningfully to conversions credited elsewhere under a simpler model. Second, for long-consideration purchases, build a view on assisted conversions or a multi-session cohort rather than declaring organic not converting based on a single session’s final action.

Common SEO Measurement Mistakes

• Reporting rankings for a huge keyword list instead of the twenty or thirty terms tied to actual revenue, which buries real movement in noise and rewards chasing vanity terms.
• Treating average position as a single number without segmenting by query, which hides that a page can be excellent for its money term and terrible for ten tangential ones.
• Ignoring engagement quality and reporting only session counts, which credits a viral but irrelevant listicle the same as a converting product page.
• Never linking GA4 and Search Console, forcing analysts to reconcile two disconnected exports by hand every month, which introduces errors and gets skipped whenever time is short.
• Treating a traffic dip as a ranking problem before checking Index Coverage, Core Web Vitals, and manual actions; a real chunk of ranking losses are actually indexing or technical issues in disguise.
• Measuring content performance only at publish plus thirty days and never again, which misses decay entirely until a page has lost most of its value.
• Crediting organic only on last-click, which undervalues its real role in assisted conversions for longer buying journeys.

A Practical Monthly SEO Reporting Template

A monthly SEO report that survives scrutiny needs the same sections every time, in the same order, so trends stay comparable month over month.

1. Visibility snapshot: Search Console clicks, impressions, average CTR, and average position, trended, with month-over-month and year-over-year context, for the site overall and for the tracked keyword set.
2. Traffic and engagement: organic sessions, engaged sessions, engagement rate, and average engagement time, segmented branded versus non-branded.
3. Conversions and revenue: organic key events by type, and revenue or lead value from the Organic Search channel, compared against the site’s average conversion rate.
4. Technical health: Index Coverage errors and trend, Core Web Vitals status distribution, any manual actions, and crawl stats anomalies.
5. Landing page movers: the top five gaining and top five declining landing pages by clicks, each with a one-line diagnosis, new content, algorithm update, decay, or technical issue.
6. Action items: three to five specific next steps tied directly to what the data showed, not a generic to-do list carried over from the previous month.

Keep the report itself to one page plus a linked dashboard for anyone who wants to dig further, and end every section with what happened and why it happened, not just what happened.

Related

• generative engine optimization

Written by Shahzaib Ul Hassan, senior AI SEO consultant and founder of ShazzSEO. Ranking sites since 2009.