Impressions dropping on Google? Here's what's going on
In late 2025, many marketers and SEO managers found themselves facing the same scenario: opening Google Search Console, watching impressions drop sharply, and wondering whether Google had penalised their websites.
The explanation, however, was more complicated than a potential algorithm update.
In September 2025, Google stopped supporting the num=100 parameter, which previously allowed users to request up to 100 search results on a single page. Several SEO platforms relied on this functionality to analyse search rankings and collect keyword data.
The change disrupted rank-tracking tools and coincided with significant impression declines reported by numerous websites, particularly in desktop search.
SEO specialists debated whether the two developments were connected, although Google did not officially confirm a causal relationship.
Then, in 2026, another piece of the puzzle emerged: Google acknowledged an impression logging error in Search Console, affecting data from 13 May 2025 to 27 April 2026.
These developments raise important questions about how we measure organic visibility. Here's what we know, what remains uncertain, and how to interpret the numbers without confusing reporting issues with actual SEO performance.
Immediate Effects on Search Console Data
Following the change to num=100, SEO professionals began reporting significant variations in Google Search Console.
The most common observations involved declining impressions and apparent improvements in average position, alongside fewer queries appearing in the deeper pages of search results.
One explanation suggested that automated requests from rank-tracking tools had previously contributed to some impressions recorded by Search Console.
It was a plausible hypothesis, but the evidence did not establish precisely how much of the change could be attributed to the parameter.
In 2026, Google also confirmed a separate logging issue affecting impressions and related metrics, including CTR and average position. According to Google's official communication, clicks were unaffected.
This distinction matters when interpreting performance:
- Impressions: a decline may reflect changes in organic visibility, reporting behaviour, or a combination of factors.
- Clicks: these need to be assessed separately to determine whether users are still reaching the website through search.
- CTR: this metric can change even when clicks remain stable, because it is calculated using impressions.
- Average position: changes in the number and distribution of recorded impressions can influence the reported average.
A sudden improvement in average position therefore doesn't necessarily mean your rankings have improved.
To understand what is actually happening, we need to examine these metrics together, comparing them with organic traffic, conversions and the performance of commercially important pages.
Global Data Insights
An analysis of 319 Google Search Console properties, conducted by LOCOMOTIVE Agency and reported by Search Engine Land in September 2025, documented substantial changes following the removal of num=100.
Within the sample:
- 87.7% of websites experienced a decline in impressions.
- 77.6% recorded fewer unique search queries.
- Queries ranking beyond the second page appeared less frequently in reports, while those associated with the first page and top positions accounted for a greater share of the visible data.
These findings demonstrated a widespread pattern within the sample. They did not prove that the parameter change was solely responsible for every fluctuation.
Google's subsequent confirmation of its impression logging error adds another important consideration: historical Search Console data can contain anomalies that affect performance analysis.
For SEO professionals, the lesson is primarily methodological.
Before interpreting a sudden movement in the charts as a ranking improvement or decline, we need to understand what may have changed in the underlying data.
A Blow to Scrapers (and Vanity Reports)
The removal of num=100 also had practical consequences for SEO software.
Several rank-tracking platforms had relied on the parameter to retrieve large numbers of search results through a single request. Once it stopped working, providers had to reconsider how they collected and monitored SERP data, with potential implications for costs and operational efficiency.
For anyone managing automated dashboards, the issue extends to the consistency of historical reporting.
Changes in data collection methods can produce misleading comparisons, especially when they are interpreted without accounting for technical changes in the platforms involved.
The logging error acknowledged by Google in 2026 introduces another complication, requiring particular caution when analysing historical impressions.
And there is an important detail: resolving a reporting problem does not necessarily mean that all the affected historical data will be corrected retrospectively.
Google's 2026 fix addressed the logging issue going forward, while previously affected data may still require careful interpretation.
The underlying principle remains the same: SEO is analysis, not superstition.
Before blaming an algorithm update for declining numbers, investigate what has changed in measurement and whether the business is experiencing any measurable consequences.
How to Interpret (and Use) the New Data
When Google Search Console shows an unexpected change, the first step is to establish whether it reflects a genuine shift in website visibility or a reporting issue.
Here are six useful checks.
- Compare impressions and clicks. A sharp impression decline without a corresponding loss of clicks requires a different interpretation from a genuine drop in organic traffic.
- Check the dates. Pay particular attention to changes around September 2025 and the period affected by Google's logging error, from 13 May 2025 to 27 April 2026.
- Analyse pages and queries separately. Identify which pages have changed, which search terms are involved and whether the affected queries are strategically relevant to your business.
- Interpret CTR and average position in context. Both metrics can be affected by impression reporting anomalies. Avoid drawing conclusions from their movements alone.
- Review dashboards and tracking tools. Check whether data collection methods have changed and document any discontinuities that could affect historical comparisons.
- Connect SEO metrics to business outcomes. Organic sessions, qualified leads, sales and conversions can help establish whether changes in reporting correspond to meaningful shifts in website performance.
Google maintains an official Search Console data anomalies page, which is worth checking whenever unexpected movements appear in performance reports.
The objective is to build more reliable reporting by distinguishing measurement issues from genuine SEO opportunities and problems.
Less time chasing unexplained curves. More attention to what actually generates qualified traffic and conversions.
SEO After num=100: A Reality Check
The num=100 story has become a lesson in the fragility of the metrics we rely on every day.
The September 2025 change disrupted established tools and reporting processes. Google's subsequent acknowledgement of a logging error in 2026 reminded us that even data from official platforms needs interpretation and scrutiny.
For SEO professionals, this brings us back to a few essential questions.
Which pages attract relevant searches? How much qualified traffic do they generate? What contribution do they make to commercial objectives?
Impressions remain a useful indicator of organic visibility, provided they are interpreted in context and alongside other performance metrics.
The charts will continue to rise and fall.
Our job is to understand what those movements actually mean before building a strategy around them.
Updated November 2026: This article has been revised following Google's confirmation of Search Console reporting issues in 2026 and further analysis of the removal of the num=100 parameter.









