SEO in the AI Era: How to Get Into Google AI Overview and ChatGPT Recommendations

← Back to Blog

The search results page looks different from what it did two years ago. For a growing number of queries, the first thing a user sees is not a list of blue links — it is a paragraph generated by Google's AI, a ChatGPT answer citing two websites, or a Perplexity summary that pulls from Reddit and a trade publication. The clicks that used to land on page one are increasingly being absorbed by these AI layers before a user ever reaches the organic results below.

This is not a distant forecast. It is the environment that B2B businesses are operating in right now, and the gap between companies that understand how to be cited in AI responses and those that do not is widening quickly.

What Classic SEO Actually Covers

Before getting into what has changed, it is worth being clear on what classic SEO means — because it is still the foundation. Traditional search engine optimisation covers four interconnected areas: technical health (site speed, crawlability, mobile rendering, structured data), on-page content optimisation (keyword targeting, heading structure, internal linking), off-page authority (backlinks from credible external sources), and analytics (tracking which pages drive conversions, not just traffic).

Done well, this framework gets your pages indexed, ranked, and clicked. It still matters. The problem is that ranking on page one no longer guarantees the visibility it used to, because a new layer of results has appeared above the organic listings.

Google AI Overview: What It Is and Why It Changed the Game

Google AI Overview — formerly called SGE — is a generated summary that appears at the top of search results, synthesising information from multiple sources into a direct answer. It does not replace the organic results; it sits above them and, critically, it does the reading for the user.

In 2026, AI Overview appears in roughly 50% of all Google searches. The distribution is highly uneven by query type. Informational queries — the kind that start with "how", "what", "why", "which" — trigger AI Overview in about 36% of cases. Commercial queries, the searches where someone is comparing agencies, software tools, or service providers, trigger it in up to 95% of cases. Transactional and local intent searches still see AI Overview relatively rarely, appearing in just 5–8% of those queries.

The traffic implication of that last statistic is significant. Research by multiple SEO analytics firms has documented that when AI Overview is present, the click-through rate on the first organic result falls by 34–65%. If your business depends on informational or commercial search traffic, some portion of that traffic is now being served directly from the AI summary, without the user ever visiting your site. Being cited inside the AI Overview is no longer a bonus — it is a visibility necessity.

How to Get Into Google AI Overview

Give the direct answer in the first 100 words

AI Overview does not have the patience for a slow build-up. If your article takes 300 words to get to the point it is making, Google's extraction model will often pass over it in favour of a page that leads with a clear, specific answer. The practical implication is that every piece of content should open with the most useful thing it knows, not a warm-up paragraph about the general importance of the topic.

Target question-format and long-tail queries

Queries of eight or more words — the kind that are essentially full sentences — are far more likely to trigger AI Overview than short-tail keywords. "Best B2B lead generation strategies for SaaS companies in 2026" is the format to aim for. Pages optimised for these specific, intent-rich queries are more likely to be pulled into AI summaries than generic pillar pages chasing head terms with thousands of monthly searches but little specificity.

E-E-A-T is no longer just a medical-sector concern

Google's E-E-A-T framework — Experience, Expertise, Authoritativeness, Trustworthiness — was for years treated as something that mostly mattered for health and finance content. In the AI Overview era, it applies across every niche. Google needs to be confident that a source it cites is credible before surfacing it in an AI-generated answer. That means author bios with verifiable credentials, clear company information, external references to your work, and a consistent publishing track record matter far more now than they did when ranking was determined primarily by link count.

FAQ schema raises your odds by roughly 60%

Structured data in FAQ format gives Google a pre-packaged, machine-readable answer. Pages that implement FAQ schema correctly — real questions with direct, specific answers — are measurably more likely to appear in AI Overview than equivalent pages without it. The 60% lift is not guaranteed, but the pattern is consistent enough across studies that FAQ schema on any content-heavy page is now a default, not an option.

Multimodal content and regular updates

AI Overview increasingly weighs whether a page pairs its written content with supporting video and images, particularly when the query involves a process or comparison. This does not mean producing video for every article, but it does mean that a written explanation of a six-step process that also includes a diagram or walkthrough video will outperform the same article in text-only form. Separately, recency matters: pages that are visibly updated — with a clear dateModified in their schema and fresh content additions — are preferred by Google over static pages, even when the original publication date is older.

You do not have to rank first

One of the more counter-intuitive findings from analysis of AI Overview citations is that Google does not exclusively pull from the top three organic results. Pages ranking anywhere from position 4 to 20 appear regularly in AI summaries, particularly when they give a more direct, authoritative answer than the pages above them. This means the strategy for AI Overview optimisation is about answer quality and authority signals, not just position.

GEO: Getting Cited in ChatGPT, Claude, and Perplexity

Beyond Google, a parallel shift is happening across the AI assistant layer. ChatGPT, Claude, and Perplexity are increasingly where users go for research, comparison, and vendor discovery — and each of these systems has its own source logic.

ChatGPT draws on Bing's index for its web-connected responses. It has a strong affinity for content structured as ranked lists — "top 10 agencies", "best tools for X", "leading providers of Y" — and it heavily weights aggregator sites and industry roundups. Being included in those lists is more valuable than most direct homepage rankings when it comes to ChatGPT visibility.

Claude (which uses Brave Search) tends to select one or two sources with the most precisely worded, directly relevant answer. It is less interested in comprehensiveness and more interested in specificity. A tightly written, well-sourced article with a clear thesis will outperform a longer, more meandering guide.

Perplexity crawls in near real-time, and its source selection has a visible bias toward forum content — Reddit threads, niche communities, Q&A sites. For industries where professionals discuss tools and vendors on Reddit or Slack communities, that conversation is a direct input into Perplexity's answers. Managing your brand's presence in those spaces is increasingly a part of search strategy.

All three systems share one underlying pattern: they look for what might be called consensual authority — sources that are mentioned across multiple independent outlets rather than a single highly optimised page. Being cited in a trade publication, a review roundup, and a practitioner's blog simultaneously carries more weight than ranking well in one channel alone.

One important caveat: AI recommendation lists are not stable. The same query about "top B2B SEO agencies" run on ChatGPT at different times can return meaningfully different results. The models are probabilistic, and the sources they pull from change as the web changes. Consistency of presence across multiple sources — not a single highly ranked page — is what produces durable visibility in this environment.

What to Actually Do: A Practical Checklist

Getting into AI-generated recommendations requires work across several areas simultaneously. Listing them as isolated tactics misrepresents how they function — they are mutually reinforcing — but for clarity:

Directory and ranking presence means actively working to appear in the "top agencies" and "best tools" lists that aggregator sites publish, because that is where ChatGPT frequently pulls its lists. Independent reviews on G2, Clutch, Google Business, and similar platforms create the third-party validation that all three AI systems look for when assessing credibility. PR and editorial mentions in trade publications — not press release syndication, but genuine editorial coverage — produce the cross-source signal that AI systems weight heavily.

NAP consistency — the uniformity of your business name, address, and phone number across every directory, listing, and social profile — is a basic trust signal that AI systems use to verify that a business is legitimate. Inconsistencies are disproportionately penalising in AI contexts compared to classic SEO. Content formatted as Q&A, step-by-step guides, and checklists is more easily extracted and cited than long-form prose — which does not mean abandoning depth, but it does mean thinking about how answers are structured within longer content. And schema markup across all content types — not just FAQ schema, but Article, Organization, and LocalBusiness where relevant — remains one of the clearest signals you can send to an automated system about what your content is and who produced it.

SEO Is Not Dead — It Has Expanded

The narrative that AI has killed SEO misses the actual situation. Classic SEO — technical health, content quality, link authority — is still the entry fee for being in the game at all. What has changed is that winning now requires operating across three layers at once: SEO (organic ranking), GEO (AI answer visibility), and brand authority (the cross-source presence that AI systems use as a credibility proxy).

The companies we work with that are growing their inbound pipeline fastest are not the ones abandoning SEO — they are the ones who understood early that the search landscape expanded, not replaced, and built strategy accordingly.