Can You Trust Your Arabic Rank Tracker? A 38-Day Check of Semrush Against Search Console

Google has been making it harder for tools to read its results, and rank trackers are the tools Arab agencies report from. We checked what that looks like in one real Arabic campaign: 200 Arabic keywords for a Saudi medical lab, tracked daily in Semrush, set next to Google Search Console for the same site over the same 38 days.

For the wider question of why the two numbers differ, read our earlier guide, Rank Tracker vs Search Console: Which Position Number Should an Arabic Site Trust? This post tests one campaign over 38 days.

What has Google changed for rank trackers?

On 12 October 2026 Search Engine Roundtable reported research by Nozzle: since September 2026 Google blocks large residential proxy networks, adds more captchas, and, in Nozzle’s measurement, returns altered results on about 8.0% of the pages it serves to suspected trackers. The same report names Sistrix, Semrush, SE Ranking, DataForSEO, Advanced Web Ranking and Morningscore as tools that posted notes or lagged on some changes. Search Engine Roundtable assumes the target is AI bots and scraping services, not SEO tools, and that is an assumption. On 16 October it also noted a ranking-volatility spike with no confirmed Google update and warned that tracking tools are struggling. Sources: Search Engine Roundtable, 12 October and 16 October.

What did one Arabic keyword campaign show?

The campaign tracks 200 Arabic keywords on Google Saudi Arabia, desktop, for a medical lab site. Between 10 September and 17 October 2026 (38 days) the average position Semrush reported ranged from 4.33 to 7.995, a spread of almost 3.7 places. In 37 day-to-day steps, 6 moved by a full place or more. The largest was on 24 September, when the average went from 4.755 to 7.25 in one day (2.5 places), the day Google’s September spam update began. Visibility moved much less, between about 57% and 77% across the period.

Search Console for the same site tells a quieter story. The average position stayed between 6.1 and 6.4 on all 38 days, including 24 September and 8 October, the day the update finished. Clicks did fall from about 20,000 a day in mid-to-late September to about 16,000 to 18,000 a day from 1 October, with impressions down about 8% and position flat, so the site lost some traffic that position alone does not explain. We have not tested why (seasonality is one candidate).

Does this prove Google is poisoning the data?

No. The two numbers measure different things: 200 hand-picked keywords on one device and location against every query the site appears for. A smaller keyword set is expected to swing more. A one-day jump of 2.5 places on 24 September could be a real move in a few keywords, a tracker reading an altered page, or a normal fluctuation. We did not check the 200 keywords one by one in an incognito window, so we cannot say which. What the data does show is that a day-to-day change in a tracker on a campaign this size is not enough, alone, to say a site was hit by an update.

How should an Arab agency read its rank reports now?

Start from Search Console, because it comes from Google and covers every query. Read the tracker second, and treat a one-day swing as a reason to check, not a result. Before reporting a drop or a recovery to a client, spot-check the keywords behind it in an incognito window from the client’s country, Riyadh or Cairo for example, and write down the date and the device. Report a range or a seven-day average instead of a single day. Annotate the dates when you know the tool or Google changed something. If the tracker and Search Console disagree for more than a week, say so in the report.

What does this not cover?

It covers one site, one country and one device, 38 days, and one tracker. We did not test Ahrefs, Sistrix or SE Ranking, and our other client sites are not in this Semrush account, so we have no second Arabic campaign to compare. Treat it as a worked example of how to check your own reports, not as a measurement of how often trackers are wrong.

FAQ

Which number should an Arabic site trust for rankings, Semrush or Search Console?

Use Search Console as the base because it is Google’s own data across all queries. Use a tracker for the specific keywords you sell and for competitors, and confirm important changes by hand.

Did the September 2026 spam update change this site’s rankings?

In Search Console the average position did not move (6.1 to 6.4 throughout). Our spam update analysis covers this in more detail.

How many keywords should I spot-check?

Twenty chosen from the keywords behind the change you are about to report is enough to see whether the tracker and a live search agree. It is a sanity check, not a statistical sample.

Related: Search Console’s AI report for Arabic websites and AI visibility tools audit.