Semalt Analytics: the SEO data GA4 hides (and how to get it back)

Why Google Analytics 4 stopped being enough for SEO

When Google killed Universal Analytics in July 2023 and forced the migration to GA4, most marketing teams thought they were simply getting a new interface. A year later the reality is different: GA4 was not designed to answer SEO questions. It was designed to measure product events and feed audiences into Google Ads. Everything that was useful for SEO — traffic by landing page, organic keywords, behaviour by source — ended up broken, hidden behind anonymisation thresholds, or removed entirely.

If you work SEO at any scale, you feel it every week. You open GA4, land on Acquisition / Sessions, and half your organic queries arrive as (not provided). You cross-check with Search Console and the numbers do not match. You try to see what users did after landing on the «moving cost Santiago» article and you get an aggregated funnel that does not split by channel. Time spent on a monthly SEO report went from 45 minutes to more than two hours — and it ends up being a report nobody trusts.

The Semalt Analytics platform was built precisely for that gap. Instead of recycling a general-purpose product, it is an engine designed by SEOs, for SEOs, that consumes raw data from Search Console, your server logs, and your CRM, and blends everything in one place. No event limit, no sampling, no (not provided).

What Semalt Analytics is and where the data comes from

Semalt Analytics is an SEO data layer connected to three sources:

1. Google Search Console v3 API — using the new endpoint that returns up to 16 months of daily-granularity history, not just the top 1,000 terms.
2. Server logs — processed in real time via a one-line tag or through a Cloudflare Worker. This captures sessions GA4 loses because of ad blockers, cookie rejection or private browsing.
3. Semalt's own crawl — the same rank-tracking engine cross-references your traffic so every organic visit is tagged with the exact ranking position of the term that brought the user in.

That cross-reference is the difference no other tool provides. GA4 tells you how many sessions arrived. Search Console tells you the average position of the keyword. Semalt tells you, for each individual session, the exact position of the term in that specific minute, which SERP feature was displayed on top (People Also Ask, video, map), and how that affected real CTR by device.

5 things you see here and never in GA4

1. Real organic traffic per URL, unsampled

GA4 starts sampling once you cross two dimensions and exceed 10 million events per month. For a mid-sized ecommerce that happens quickly. In Semalt there is no sampling. Every URL is processed in a columnar index (built on ClickHouse) that answers under a second, even for questions like «show me the top 500 landing pages by organic traffic in January, compared with January the previous year, mobile only».

2. Keywords without (not provided)

The infamous (not provided) from GA4 does not exist in Semalt because the source is Search Console raw data and Semalt's own crawl. For every URL you see the 50 keywords that drive the most traffic, with clicks, impressions, CTR and average position. And most importantly: you see the keywords sitting in positions 4-10 — the ones a small content tweak could push into top 3 and double the traffic. That view is called Striking Distance and it is the first screen we open on any new client audit.

3. Sessions GA4 loses to ad blockers

In Latin America the desktop ad-blocker rate hovers around 22% according to GlobalWebIndex 2025, and in users under 35 it climbs to 34%. All of them are invisible to GA4. Semalt's log-based tracking sees them because it does not depend on JavaScript: when the server returns the page, it is counted. In our client base we have seen the gap swing between 12% and 41% of extra sessions, higher in tech and lower in traditional retail.

4. Full cross-device journey

GA4 promises to unify cross-device sessions via Google Signals, but that requires users to be logged in to Google on both devices and not have rejected consent mode. Consent acceptance rates in most markets sit below 40%. Semalt uses a technique called server-side fingerprint fusion that respects privacy law (no advertising profiles) but still lets you see the user read your article on mobile in the morning and buy on desktop at night.

5. Organic revenue per keyword

The crown: if you connect your CRM or store (WooCommerce, Shopify, Vtex, Magento, Prestashop) via webhook, Semalt maps each sale to the first organic keyword that brought the user in. You can see that «santiago office furniture price» generated CLP 14 million in revenue last month while another keyword with much higher traffic generated zero. With that data you stop chasing vanity and start prioritising content by real revenue.

Case study: Chilean ecommerce, 40k organic sessions/month

A client of ours in Providencia, a pet accessories store, had this picture in January 2026:

• GA4 reported 41,200 organic sessions and 128 conversions (0.31% CVR).
• Search Console showed 87,000 clicks for the same period.
• The difference (more than 2x) was impossible to explain. The internal team blamed GTM, Cloudflare, the consent banner.

We connected Semalt Analytics in 14 minutes. Numbers the next day:

Real organic sessions: 76,400 (85% more than GA4). Conversions correctly attributed to first organic touch: 314 (2.45x). Verified organic revenue: CLP 38.7M vs 11.2M the team was estimating — a gap of 27.5 million that was invisible in their monthly report.

The most valuable insight came when we crossed keywords with revenue: nine long-tail terms (all between 40 and 300 searches/month) accounted for 63% of revenue, while the head term «dog accessories» with 4,400 searches/month accounted for barely 4% of sales. They reshuffled the content strategy in two weeks and by March organic revenue was up 71%.

12-minute setup

Semalt's advantage over homebrew stacks (BigQuery + Looker) is setup speed. Four steps:

No devs, no BigQuery, no dashboards to build. First reports are live within the hour.

Semalt Analytics vs GA4 vs GSC vs the classics

More in this series: Semalt AutoSEO: optimisation that runs 24/7, Competitor analysis without leaving a trace, Content marketing: from cluster to results.

Frequently asked questions

Does it replace Google Analytics 4?

If your focus is SEO and content, yes. If you also run Google Ads and need to export audiences, keep GA4 in parallel — Semalt does not replace the native Ads connector. In 8 out of 10 clients we keep GA4 as the «Ads source» and Semalt as the «SEO source».

Is it privacy-law compliant?

Yes. Log-based tracking does not need persistent cookies or advertising fingerprinting. Server-side fusion operates with pseudonymised data and sessions anonymise after 26 months by default (configurable). Semalt provides a DPA that can be e-signed from the account.

How fast until it delivers value?

Search Console historical data is ready in 5 minutes. Live tracking shows complete data after 24 hours. The Striking Distance insight (keywords in positions 4-10) is actionable the same day you connect. Our record time between connection and first profitable decision was 90 minutes.

Does it work for small sites?

Yes. A personal blog with 2,000 sessions/month uses Semalt the same way as an ecommerce with two million. The engine processes the same data. For small sites the value is spotting which article is about to rank on page 1 and tuning it before Google crystallises the position.

Does it integrate with Semalt AutoSEO?

Yes, and that is the synergy. Analytics identifies which URLs have keywords in positions 4-10 and AutoSEO executes the fixes on those URLs without a dev in the loop. Detail in our AutoSEO guide.