A fall in organic traffic shortly after a Google update can look alarming, but timing alone does not prove that the update caused the decline. Search demand changes throughout the year, customers alter their behaviour, competitors gain visibility, search result layouts change and individual pages can lose positions for reasons unrelated to a major ranking change. In 2026, the most reliable way to diagnose a decline is to separate demand from visibility. That means looking beyond total sessions or clicks and comparing impressions, positions, queries, pages and historical search interest. A seasonal decline normally affects the size of the available audience while rankings remain broadly stable. A genuine ranking loss changes how prominently the site appears for queries that previously generated traffic. The difference becomes much easier to see when the analysis uses comparable periods and several indicators rather than one traffic graph.
The first mistake in post-update analysis is assuming that two events occurring at roughly the same time must have the same cause. Google makes many changes to Search during the year, including broad core updates and smaller changes that may not receive a public announcement. At the same time, search demand can fall because a holiday has ended, a buying season is finishing, schools have reopened, weather has changed or interest in a particular product has simply cooled. If organic clicks fall during an update, treat the timing as a reason to investigate rather than proof of an algorithmic loss. The Google Search Status Dashboard provides the confirmed start and completion dates of announced ranking incidents and updates, which gives the analysis a much more reliable reference point than SEO industry speculation.
Google’s current guidance is particularly useful here. When a site appears to have been affected by a core update, Google recommends waiting until the update has finished rolling out and then allowing at least one full week before carrying out a detailed comparison. A sensible test is therefore to compare a complete week after the rollout with a comparable week before the rollout began. Daily data during an update can be misleading because rankings may fluctuate while systems are being adjusted. For seasonal businesses, this before-and-after comparison should not be the only test. The same dates from the previous year are often more informative because they reveal whether the decline follows an established annual pattern rather than beginning with the update itself.
Site-wide totals should then be broken into smaller groups. Check which queries lost clicks, which URLs lost impressions and whether the change is concentrated in a particular country, device or search type. A site can show a 20% decline overall even when most pages are performing normally if a small group of high-volume seasonal queries has lost demand. The reverse can also happen: total traffic may look relatively stable while an important commercial category loses positions and another section temporarily compensates for the decline. Looking at individual sections prevents averages from hiding the real issue. The objective at this stage is not to decide immediately that Google has rewarded or downgraded the site, but to establish exactly where the missing traffic came from.
Four Search Console metrics are particularly useful: clicks, impressions, click-through rate and average position. Clicks show visits from Google Search, while impressions indicate how often a result from the site appeared in eligible search results. CTR measures the share of impressions that resulted in a click. Average position needs more careful interpretation. Google calculates it using the topmost position of the property for each impression and then averages those observations. It should therefore be treated as a directional diagnostic metric rather than a perfect record of one fixed ranking. This is especially important for sites ranking for thousands of queries, because changes in the query mix can move the site-wide average even when many individual keywords remain stable.
A seasonal decline usually produces a recognisable combination of signals. Clicks fall because fewer people are searching, and impressions often fall at a similar time because there are fewer searches in which the pages can appear. However, positions for the important queries remain close to their previous levels. For example, if a page continues to average around position four for a group of summer travel queries but its impressions fall sharply from September onwards, weaker demand is a more plausible explanation than a ranking loss. If the same curve appeared during the equivalent period a year earlier and Google Trends also shows declining interest, the evidence for seasonality becomes considerably stronger.
A ranking problem tends to look different. Important queries may move from positions three or four to positions eight, twelve or lower, followed by a reduction in clicks and often impressions. The decline may be concentrated in particular directories, topics or types of pages rather than following the same pattern across all seasonal searches. Another possibility is that clicks and CTR fall while rankings and impressions remain relatively stable. That pattern deserves separate investigation because the page may still rank well but receive fewer clicks due to a changed title or snippet, stronger competing results, new search features or a shift in user intent. Treating every click decline as a ranking loss can therefore lead to unnecessary changes to pages that still perform adequately.
Seasonality becomes easier to identify when recent performance is compared with a sufficiently long historical period. Google recommends viewing up to 16 months of Search Console performance data when investigating traffic declines. This range normally includes the equivalent period from the previous year and makes recurring peaks and troughs visible. Instead of asking whether traffic fell 25% compared with last month, ask whether a similar fall occurred at roughly the same point last year. A month-to-month comparison may exaggerate the problem when the previous month represented the annual peak. Year-on-year comparisons provide a more realistic baseline for industries such as travel, retail, education, gardening, tax services, sporting goods and event-related businesses.
Google Trends provides a second source of evidence because it reflects broader search interest rather than the performance of one particular website. Start with the queries or topics that account for a meaningful share of the site’s lost clicks. If impressions for “garden furniture” decline after summer and Trends shows a comparable fall in public interest, reduced demand is a credible explanation. If search interest remains stable while the site’s impressions and positions deteriorate, the evidence points more strongly towards a visibility problem. Trends data is relative rather than an exact search-volume counter, so its value lies mainly in showing direction, recurring patterns and changes in interest over time.
Seasonality should also be judged at query level because different parts of the same business can follow very different calendars. An online retailer may see falling searches for outdoor furniture while demand for winter clothing begins to rise. A travel publisher may lose searches for Mediterranean beach holidays while ski-related interest increases. If all traffic is combined, these movements can partly cancel one another and make the site’s performance appear more stable than individual categories really are. Separating major topic groups therefore provides a cleaner diagnosis. Brand queries should also be reviewed separately from generic searches because a decline in branded demand may reflect weaker brand interest or marketing activity rather than a Google ranking change.
A practical seasonal baseline does not require complex forecasting software. Start by selecting the pages and queries responsible for most organic traffic, then record their clicks, impressions and average positions for the current period and the equivalent period one year earlier. Compare both the absolute figures and the percentage changes. Suppose a category lost 30% of its clicks this August compared with July. On its own, that looks serious. If the same category lost 28% between July and August last year while its important rankings stayed nearly unchanged, the current decline is probably close to its normal seasonal pattern. If last year’s traffic remained stable but this year’s positions have fallen significantly, further ranking analysis is justified.
The calendar also needs context. Year-on-year comparisons are useful, but equivalent dates are not always commercially equivalent. Easter moves between March and April, major sporting events change dates, school holidays vary between markets and exceptional weather can alter demand for travel, clothing, home improvement and leisure products. Promotional campaigns can also distort the baseline. A site that received a large television, email or social campaign last August should not automatically expect the same level of branded search this August. Add these business events to the analysis before attributing an unusual curve to Google. Good SEO diagnosis combines Search Console data with an understanding of what customers were actually doing at the time.
By this point, most cases can be placed into one of several practical categories. Falling clicks and impressions, stable positions and lower market interest usually indicate seasonality or changing demand. Falling clicks accompanied by clearly weaker positions across important queries is more consistent with ranking loss. Stable positions and impressions combined with a weaker CTR suggest that search result presentation or user behaviour may be responsible. A sudden site-wide collapse that does not match demand patterns requires checks for technical, indexing, security or policy problems. These categories are not mutually exclusive. A site can lose some rankings during a period when demand is also falling, which is why the diagnosis should estimate how much of the decline each factor explains instead of looking for one universal cause.

If seasonality does not explain the decline, identify the pages and queries with measurable ranking changes. Compare the periods before and after the update and focus first on URLs that previously generated substantial impressions or business value. Small movements should be kept in perspective. A query moving from position two to position four may reduce clicks without indicating a serious site-wide problem, while a large group of important pages moving from the first page of results to much lower positions deserves closer attention. Google itself distinguishes between small and large position declines when advising site owners after core updates and warns against making radical changes simply because rankings have moved slightly.
Before rewriting content, rule out problems that can imitate an algorithmic loss. Check whether important pages are still indexed and accessible to Google, whether canonical settings have changed, whether robots directives block content and whether a recent migration, redesign or URL change coincided with the decline. Search Console should also be checked for manual actions and security issues. These checks do not need to become an extensive technical audit unless the data points in that direction. The key question is whether Google can access, index and serve the same pages that performed before the drop. If rankings disappeared immediately after a deployment rather than gradually around a search update, the website change deserves at least as much attention as the algorithm.
Ranking losses can also occur because other results have become more useful for the same search intent. Core updates are broad reassessments of search results; they are not penalties aimed at individual websites. A page moving down does not automatically mean that Google has identified a violation. Competing pages may provide clearer answers, stronger first-hand evidence, more appropriate formats or information that better matches what people now want from the query. Search results themselves also evolve. Shopping elements, videos, local results and other features can change how much attention traditional organic listings receive. For this reason, review the actual queries and the type of results appearing for them rather than judging performance entirely through historical ranking numbers.
When the evidence indicates seasonality, avoid making major SEO changes merely to reverse a normal traffic curve. Use the quieter period to prepare for the next demand cycle instead. Update genuinely outdated information, make important seasonal pages easy to reach before interest begins rising and use historical query data to determine when users start researching rather than when purchases reach their peak. Publishing or refreshing a useful seasonal resource shortly before demand normally appears can give Google time to crawl and process it. The success measure should then be year-on-year visibility during equivalent stages of the season, not an unrealistic attempt to maintain peak traffic throughout the entire year.
When a substantial ranking loss remains after seasonal, technical and reporting factors have been excluded, review the affected content from the user’s perspective. Google’s current core update guidance emphasises helpful, reliable and people-first material rather than quick ranking fixes. Check whether pages provide original information, sufficiently complete answers, clear authorship where users would expect it and evidence of relevant experience or expertise. Compare the affected pages with the needs behind the queries, not simply with the word counts or keyword usage of competitors. Large batches of superficial edits, unnecessary deletion of useful material or changing publication dates without meaningful updates are unlikely to solve an underlying quality problem. Improvements should have a clear reason for the reader.
Finally, measure recovery with the same discipline used to diagnose the decline. Keep a record of the update dates, the periods compared, the affected page groups and any substantial changes made afterwards. Review query and page performance over several weeks rather than reacting to individual days. Google notes that some improvements can be reflected relatively quickly, while broader reassessment of a site’s helpfulness may take several months, and a noticeable recovery is never guaranteed. A reliable diagnosis therefore does more than label a traffic drop as “seasonal” or “algorithmic”. It establishes a repeatable evidence trail showing what happened to demand, visibility and user response, allowing future traffic changes to be assessed without unnecessary panic or speculative site-wide edits.