Search Console: Why do the data never match between the old and new interface?

Where do these data discrepancies between interfaces come from?

The inconsistencies between old and new interfaces of Search Console are not isolated errors but stem from fundamental differences in the underlying system architecture. The old interface relied on data processing pipelines designed years ago, while the new one is built on modernized infrastructure that aggregates, filters, and presents information differently.

These divergences particularly affect performance metrics: impressions, clicks, average position, CTR. The same site may show a 10% variation in impression volume depending on the accessed interface. The reason? The filter thresholds, sampling periods, and aggregation methods are not identical. The old interface sometimes applied more aggressive anti-spam filters or excluded certain queries deemed irrelevant, while the new one tries to provide a more comprehensive view.

The latency of synchronization also plays a role. Data in the new interface typically refreshes within 48 hours, but some reports in the old interface could take up to 72 hours to incorporate changes. As a result, for a few days, comparing the two sources leads to shaky conclusions.

What types of metrics are most affected by these inconsistencies?

The metrics of impressions and average position are the most vulnerable. Average position, in particular, is calculated differently depending on the interface: the old one could ignore certain positions considered outliers (position 100+), while the new one includes them in the calculation. This technical nuance radically changes the displayed figures for low-competition long-tail queries.

Click data experiences less variation but is still affected by deduplication rules. If a user clicks on the same result twice in a short period, the old interface might count two clicks where the new one counts only one. These micro-differences accumulate and distort analyses of conversion rates or page performance.

Another sensitive point is index coverage data. A site might show 5,000 indexed pages in the old interface and 5,200 in the new. This gap often arises from differences in how canonical URLs, redirects, or AMP pages are accounted for. The new interface aims to be more granular, but this creates discrepancies with the old consolidated metrics.

How can we interpret these discrepancies without falling into paranoia?

The first reflex should be to never directly compare a metric from the old interface with its equivalent in the new one for the same period. That would be like comparing apples to oranges. If you notice a 15% discrepancy in impressions, it does not mean Google has lost or gained real traffic. It simply indicates that the two interfaces do not measure exactly the same thing.

However, medium-term trends remain reliable. If impressions increase by 20% month-over-month in the new interface, this trend reflects an SEO reality, even if the absolute figures differ from the old version. The key is to remain on a single data source to track evolution over time, not to juggle between the two to validate a one-time figure.

Finally, cross-referencing with Google Analytics or server logs helps to put things into perspective. If Search Console shows 10,000 clicks but GA reports 9,500, the 5% discrepancy remains acceptable and consistent with classic methodological differences. If the gap reaches 30%, that indicates a tracking problem or configuration to investigate, regardless of the GSC interfaces.

• Discrepancies are not bugs but architectural differences acknowledged by Google.
• Impressions and average position suffer the most significant gaps between old and new interfaces.
• Never compare a metric from the old interface with the new for the same period.
• Long-term trends remain reliable if you stick to a single data source.
• Cross-reference with GA or server logs to validate the overall consistency of figures.

Is this statement consistent with what we observe in practice?

Absolutely. For years, SEO practitioners have observed unexplained discrepancies between GSC reports, especially during migrations between old and new interfaces. Google is finally formalizing what many already knew: the two systems do not share the same database or the same calculation rules. This is not an admission of failure, but a recognition that perfect data uniformity does not exist in such a massive system.

However, Google remains surprisingly vague about the acceptable magnitude of these discrepancies. Is a 5% gap normal? 10%? 20%? [To verify] No official documentation sets clear thresholds. This opacity forces SEOs to empirically define their own margins of error, complicating audits and client reporting. When a client asks why impressions dropped by 12%, explaining that it could just be an interface artifact becomes difficult without official data to back it.

Another point: the synchronization latency mentioned by Google does not cover all cases. Some sites have noticed persistent gaps for weeks, even months, between the two interfaces. This suggests that discrepancies are not just temporal but also structural, tied to how certain verticals (news, video, images) are treated differently depending on the interface.

What are the practical implications for an SEO audit?

A serious SEO audit can no longer simply capture screenshots of GSC metrics without specifying which interface was used, on what date, and with what filters. Reports must now explicitly mention the source of the data. Otherwise, comparing an audit conducted in January with another in June becomes impossible if one used the old interface and the other the new.

Monthly performance tracking also needs to be standardized. If you switch mid-course from the old interface to the new, trend lines may show a false drop or rise. The only solution is to export historical data from the old interface before a complete migration and never touch it again. Any new analysis should be based exclusively on the new interface.

Finally, automated alerts configured via the Search Console API can trigger false alarms if they compare periods straddling an interface migration. An alert system based on an impression variation of +/- 10% must be recalibrated to account for these artificial fluctuations, or risk drowning out real signals in noise.

Can we still trust Search Console as a source of truth?

Yes, but with some methodological caution. Search Console remains the most direct tool to understand how Google sees and indexes a site. No third-party tool can provide such granular data on queries, indexed pages, or coverage errors. However, it should never be the sole decision-making source.

Cross-referencing GSC with Google Analytics, server logs, and third-party tools (SEMrush, Ahrefs, etc.) helps detect inconsistencies. If GSC reports a 20% drop in impressions but organic traffic in GA remains stable, it’s likely an artifact of the interface or a change in how Google counts certain queries. Conversely, if both sources align, the trend is real.

The real issue is not the reliability of GSC itself, but the lack of transparency regarding calculation methodologies. Google could publish precise technical guidelines on the differences between interfaces, acceptable deviation thresholds, and best practices for data consolidation. Without this, every SEO has to reinvent the wheel and develop their own heuristics, which is time-consuming and prone to error.

How can we standardize data tracking to avoid inconsistencies?

The first rule is to choose one interface and stick with it. If you are still using the old Search Console for some reports, fully migrate to the new one and archive historical data in a separate file. Never mix sources in the same dashboard or reporting. This discipline helps avoid faulty comparisons and false alerts.

Next, automate data export via the Search Console API and store it in an external database (BigQuery, Google Sheets, SQL database). This allows you to freeze metrics at a given moment and compare them consistently even if Google changes its interfaces or calculation methods. The API itself may undergo variations, but these are generally better documented than those of graphical interfaces.

Finally, document every methodological change in your audit processes. If you transition from the old to the new interface in March, note it in your reports and inform clients that March figures are not directly comparable to those from February. This transparency avoids misunderstandings and enhances the credibility of your analyses.

What indicators should be prioritized when data diverges?

If you notice a significant gap between old and new interfaces, always prioritize the new one. Google is directing its resources toward this modernized version, and it will receive future updates. The old interface is under minimal maintenance and will eventually be completely deactivated. Basing a strategy on outdated data poses an unnecessary risk.

For average position metrics, never take them at face value. They are too sensitive to methodological variations and long-tail queries. Focus instead on actual clicks and CTR, which are less prone to calculation artifacts. If CTR increases even with a stable average position, it indicates that on-page optimizations are working. This is more actionable than an average position fluctuating from 5.2 to 5.8 without apparent reason.

Finally, index coverage data deserves separate treatment. If the new interface reports more indexed pages than the old one, manually verify via site: queries on Google. If both sources diverge significantly from the site: result, it indicates a potential canonicalization or crawl issue that needs investigation, irrespective of GSC interfaces.

What checklist should be applied to secure reporting?

• Use exclusively the new Search Console for all new reports and performance tracking.
• Export and archive historical data from the old interface before complete migration.
• Automate extraction via the API and store data in an external database to ensure consistency.
• Document the date and interface used in every client or internal report.
• Cross-reference GSC with Google Analytics and server logs to validate trend consistency.
• Never compare metrics from different interfaces for the same period.

These optimizations in tracking and data consolidation may seem straightforward in theory, but their rigorous implementation requires time, technical skills, and constant vigilance. If your team lacks internal resources or you manage multiple sites, partnering with a specialized SEO agency can be wise. An experienced partner can configure robust data pipelines, automate reporting, and interpret discrepancies with the necessary perspective. This allows you to focus on strategy instead of spending hours reconciling inconsistent figures.