What the September 2026 Google Update Means for Your AI Strategy

What changed in the September 2026 Google Update?

The material change is an AI Overviews interface expansion, not a confirmed September core update. On 27 to 28 August, Google confirmed that some AI Overviews can dynamically expand into a longer answer when its systems decide more detail will help. The manual Show More control is removed for those queries, although the expanded answer can collapse when a user scrolls toward standard results.

That distinction matters. A core update changes how Google evaluates and ranks content. This change alters the amount of generated content a searcher may see before organic listings. A site can hold its traditional position while receiving fewer opportunities to earn a visit, especially where the overview gives a complete answer.

Google also began moving generative interface elements from AI Mode into AI Overviews. Harri Digital reports that these elements can create custom layouts, interactive tools, and simulations within the response. The dynamic expansion and generative interface rollout are related in outcome because both make the answer experience more self-contained. They were confirmed separately, so marketers should not flatten them into one claimed algorithm update.

The important before-and-after is simple: before the expansion, a user could encounter a shorter overview and choose whether to reveal more. After the expansion, Google can show a fuller answer automatically for selected queries. The working implication is not that organic search is over. It is that the visible answer layer has gained more control over the first screen a searcher sees.

1. Before 27 to 28 August 2026: AI Overviews could present a shorter response with a manual Show More path for additional detail.
2. After 27 to 28 August 2026: selected AI Overviews can automatically expand into a longer AI Mode-style response.
3. What to do: audit question pages that drive impressions, then record whether Google cites your domain in the overview, not only where the URL ranks.

Is this a Google core update or a change to the search interface?

It is a search-interface change, based on the available confirmed information, rather than a September core update. Google Search Status Dashboard is the authoritative place to check named ranking-system incidents and rollout status. Harri Digital reported on 5 September that no new September core or spam update had been announced, while the AI Overview expansion was already confirmed.

Treating every visibility change as a core update creates bad diagnosis. A fall in clicks can come from a lower ranking, a different query mix, a richer search feature, an AI Overview, or a change in how much answer text Google places ahead of links. Those situations need different responses. Rewriting pages because of an unverified ranking theory is usually less useful than first checking the search results people actually receive.

This is where old reporting habits break. A position report tells you where a result appears in a list. It does not tell you whether an AI Overview appeared, whether it expanded, which sources it cited, or whether the result was visually pushed down. An AI strategy needs a prompt-level record of all four.

The absence of a named September core update is not proof that nothing moves. Google continuously changes Search, and the dashboard does not list every adjustment. It does mean teams should keep their language precise: describe an observed result-page change as an interface shift unless Google has announced a ranking update.

1. Use Google Search Status Dashboard to distinguish confirmed named updates from observations.
2. Check representative results manually or with approved monitoring before attributing traffic movement to this change.
3. Keep ranking, click-through, AI Overview presence, and citations as separate measures in reporting.

Who does the AI Overviews expansion affect most?

The expansion most directly affects publishers and businesses whose pages answer informational, comparison, and multi-part questions. Those are the query types where a longer generated answer can satisfy more of the searcher's need without a click. If your acquisition model depends on a searcher reading an explanation on your site, the answer layer is now a more important competitor for attention.

Comparison pages deserve special attention. A searcher asking for the difference between two tools, methods, or providers may receive the key distinctions inside the overview. That does not remove the opportunity for your page. It changes the immediate goal: your source needs to be clear enough, specific enough, and current enough to earn a citation that carries your expertise into the answer.

How-to and educational content is also exposed because Google is adding richer generative elements to the overview experience. The source page still has a role when it provides depth, original documentation, a credible method, or a decision-specific answer. Generic summaries are easier for an overview to compress because they has little a searcher cannot obtain in the result itself.

Local and transactional queries should not be assumed safe or doomed. Their result pages include maps, ads, local packs, reviews, and other has that vary by query and location. The right response is to test the actual questions that produce calls, bookings, demos, or qualified visits. Do not transfer a pattern from a broad informational query onto a local service query without evidence.

For B2B SaaS teams, the practical unit is the category prompt and the comparison prompt. For service businesses, it is the service-and-location prompt. In both cases, Citation Share gives a more direct measure of whether the brand is present in relevant AI answers than a conventional average position alone.

1. Informational pages: test whether the overview answers the page's main question without a click.
2. Comparison pages: test whether your brand, evidence, or product category is cited.
3. B2B category pages: track citation presence on relevant recommendation prompts.
4. Local pages: test service-and-location prompts separately because result layouts differ by market.

How should you respond to the September AI Overview change?

Respond by making citation visibility a reporting discipline, then improving the pages that answer the questions AI Overviews are absorbing. Do not chase a supposed ranking fix for an interface change. Start with the prompts tied to commercial outcomes and inspect what Google shows today.

First, build a baseline. Group prompts by intent, such as category research, comparisons, how-to questions, service-and-location searches, and branded questions. For each prompt, record whether an AI Overview appears, whether it expands, which domains receive citations, what facts are used, and whether your brand appears. This produces a defensible before state rather than a collection of impressions.

Second, inspect cited competitor pages. Look for the underlying answer structure, not a trick to copy. Strong cited pages often make a specific claim, explain it in plain language, show relevant evidence, and maintain the page when facts change. The goal is quality, coverage, and freshness. It is not to manipulate the models.

Third, repair gaps by publishing pages that answer a single important question cleanly. Lead with the answer. Use descriptive question headings. Add real sources for factual claims. Include useful comparisons where the prompt requires a choice. Explain limits instead of burying them. These choices make a page more useful to readers and easier for an answer engine to evaluate.

Fourth, watch outcomes over time. Citation Count per day measures volume. Answer Presence measures how broadly a brand appears across the question set. Citation Share measures the percentage of relevant answers that cite the brand. Share of Voice compares that presence with named competitors. Each metric answers a different question, so a single score should not hide the underlying evidence.

Finally, protect the work that still earns clicks. An expanded overview may handle the basic answer, but users still leave Search for proof, implementation detail, pricing context, distinctive research, and decisions that require trust. Build those reasons into the page instead of relying on the existence of a blue link.

1. Create a prompt inventory tied to revenue, leads, calls, bookings, or product research.
2. Capture a dated baseline of AI Overview presence and cited domains.
3. Prioritize pages where your brand is absent but the query matters commercially.
4. Update factual pages with source-backed, question-first answers and clear comparison context.
5. Report Citation Share alongside rankings and organic clicks.

What should your September monitoring report include?

Your September monitoring report should state what changed on the result page, what happened to citation visibility, and what action follows. It should not claim that a core update caused a movement unless Google has confirmed one. This gives leadership a clearer explanation of risk and prevents unnecessary content churn.

Use a repeatable prompt sample. The sample should include high-intent questions, queries that already produce impressions, competitor comparison prompts, and any local combinations that matter to the business. Retain the exact query text, country or city, device context where relevant, observation date, AI Overview state, cited URLs, and notes on whether the answer expanded.

A useful report separates rank from citation. If your URL remains visible in traditional results but is not cited in the AI Overview, the next job is not necessarily technical SEO. It may be content coverage, stronger sourcing, clearer page structure, or an unanswered question in the topic cluster. If your domain is cited but clicks decline, investigate whether the overview now satisfies the basic query and whether the cited page provides a compelling next step.

The strategic shift is direct. Rankings got you found. Citations get you chosen. September does not prove that every click will disappear. It does show why a visibility program built only around blue-link positions is incomplete when Google can expand the answer before the searcher reaches those links.

1. Record the query and observation date so changes are comparable.
2. Save cited domains and URLs, not only a screenshot or a position number.
3. Mark whether the AI Overview was short or automatically expanded.
4. Connect every content action to a measured citation gap or a commercially important prompt.

Sources

Source: Harri Digital

Google Search in September 2026 had no confirmed new core update as of 5 September, while AI Overviews could dynamically expand into a longer answer. Harri Digital, 2026-09-05

Source: Search Engine Land

Google is dynamically expanding AI Overviews for some queries. Search Engine Land, 2026-08-28

Source: Google Search Status Dashboard

Google's Search Status Dashboard is the official status destination for Google Search ranking updates and incidents. Google Search Status Dashboard, 2026-09-05