Google says the usual SEO practices still apply to its AI features, and buyers still start with search. What has changed is the goal: you used to aim to rank and get a click, and you now also need to be named and cited in AI answers.
If you run B2B marketing, you have probably heard that AI has killed SEO, yet the work you already do still matters and a small set of additions covers the new part.
This guide separates what stays the same from what has changed, using studies we could check at the publisher. It ends with a 90-day to-do list your team can start on Monday. It is written for marketing heads at B2B firms selling in India, the UK and the Netherlands.
Is SEO dead because of AI?
No. SEO is changing shape, and part of the traffic it used to bring is now answered on the results page itself.
Three things are true at once. People still search. Google says the best practices for SEO remain relevant for its AI features. And studies agree that fewer people click through when an AI summary appears.
So the question to ask is not whether SEO works. Ask whether your pages are good enough to be ranked, and clear enough to be quoted.
What is the difference between SEO and AI SEO?
SEO aims to get your page ranked in a list of links. AI SEO aims to get your company named, described correctly and cited in an answer written by an AI assistant.
You will see other names for the second activity: generative engine optimisation (GEO), answer engine optimisation, or AI search optimisation. They describe the same job. The job builds on SEO and does not replace it.
| SEO | AI SEO | |
|---|---|---|
| Goal | Rank and earn the click | Be named and cited in the answer |
| Where it shows | A list of links | A written answer, often with a few sources |
| Unit of work | The page and its keywords | The fact, the claim and the entity (your company, product, people) |
| What helps | Relevance, links, a fast and crawlable site | Clear answers, evidence, mentions on other trusted sites |
| Main measure | Rankings, clicks, enquiries | Mentions, citations, enquiries that name AI |
In our view, the two share most of their foundations. A team that does SEO well is most of the way there.
How has the way buyers search changed?
Buyers now ask AI assistants for shortlists, and fewer searchers click links when an AI summary is shown. The studies below are the ones we could check at the publisher, and they do not all point the same way.
- G2, March 2026. In a survey of 1,076 B2B decision-makers, 54% said AI chatbots influenced their shortlists, more than any other source listed. G2's report is about software purchases, so treat it as a direction for other B2B categories.
- Pew Research Center, 2025. Tracking 900 US adults, Pew found people clicked a traditional result in 8% of visits when an AI summary appeared and in 15% of visits when it did not. Only 1% of visits included a click on a link inside the summary. This covers general users, not B2B buyers.
- Ahrefs, February 2026. Across 300,000 keywords, the top-ranking page's average click-through rate was about 58% lower when an AI Overview appeared. Ahrefs is an SEO software company, and its method rests on an estimate of what the rate would otherwise have been.
- Google. Google says there are no extra requirements or special optimisations needed to appear in AI Overviews or AI Mode. It also says you do not need new machine-readable files or markup for them.
Where they differ, the studies are measuring different things: general searchers against B2B buyers, clicks against influence, and a company's own statement against third-party data. Together they say that visibility is moving from the click to the answer.
Is SEO still important for B2B?
Yes. B2B buying is research-heavy, and search is where much of that research starts.
AI assistants also draw on pages that exist on the open web. If your pages are thin, slow to load or hard to crawl, you give those assistants less to work with. Google also states that a page must be indexed and eligible to show a snippet to appear in its AI features.
For a B2B firm with a long sales cycle, SEO also does a quieter job. It puts your technical answers in front of the engineer, the finance controller and the procurement lead, each time they check you out before a meeting.
What stays the same in SEO?
Most of the work. If you already do these, keep doing them:
- Content that answers real questions. Specific, accurate pages written by people who know the subject.
- Authority. Other credible sites linking to you and naming you. Topical authority means you cover one subject in enough depth that you are the obvious source for it.
- Technical health. Pages that can be crawled, indexed and loaded quickly on a phone.
- Search intent. One page for one need, clearly labelled.
None of this is new advice. It is the same advice, and it now also feeds the AI answers.
What stays, what changes and what to add?
Use this table as a quick audit. Mark each row with a tick, a partial or a gap.
| Area | What stays | What changes | What to add |
|---|---|---|---|
| Content | Useful pages for real buyer questions | The first 50 words under each heading get quoted more than the rest | A direct answer under each question heading, then the detail |
| Authority | Links from credible sites | Mentions without a link now count as signals too | Presence in the places buyers and AI assistants read: trade media, review sites, partner pages |
| Technical | Crawlable, indexed, fast pages | Little | Check that AI crawlers are not blocked by mistake |
| Structure | Clear headings and internal links | Pages are read in pieces | Tables, short lists and named, dated sources |
| Measurement | Rankings and organic enquiries | Clicks fall even when visibility holds | Mentions, citations and a "how did you hear about us?" option that names AI |
| Brand | A consistent name and description | The assistant describes you from many sources | The same company description everywhere you appear |
The right-hand column is the new work. It is a small part of the total.
What should you add for AI search?
Add three things: mentions, citations and structure.
Mentions. AI assistants build an impression of your company from many sources. Get named accurately in trade press, partner sites, review platforms and industry lists, using one consistent description.
Citations. Give assistants something worth citing. That means original data, a clear definition, a checklist or a table that is better than what already exists. Link to your own primary sources. State dates.
Structure. Put the direct answer first, under a heading phrased as the question. Keep paragraphs short. Use tables for comparisons. Add FAQ sections with plain answers.
Skip anything that promises a trick. Google says no special files or markup are needed, and we have not seen verified evidence that one-off hacks work.
How do you measure AI visibility?
Measure it by hand first. The tools are young and the answers change from one day to the next.
- Write 20 questions your buyers ask, from first research to shortlist.
- Ask each one in two or three AI assistants, on the same day each month.
- Record whether your company is named, whether it is described correctly, and which sources are cited.
- Compare with two competitors.
- Add "ChatGPT or another AI assistant" to the "how did you hear about us?" field on every enquiry form.
In our view, enquiries that name AI are a better signal than any dashboard score. They are few at first, so look at the trend over quarters.
What does a 90-day to-do list look like for a B2B team?
It looks like this. The plan assumes a small team and makes no promise of results.
Days 1 to 30: check the base.
- Confirm your key pages are indexed and fast on a phone.
- Check your robots file does not block AI crawlers by mistake.
- Run the 20-question test above and save the answers as your baseline.
- Write one standard company description and use it everywhere.
Days 31 to 60: fix the pages that matter.
- Pick the five pages closest to enquiries. Add a direct answer under each question heading.
- Add a comparison table or checklist to each.
- Name sources and dates for every claim.
Days 61 to 90: build the mentions.
- Offer one piece of original material, such as a checklist or a small survey of your customers, to two trade publications.
- Update your listings on the review and directory sites your buyers use.
- Repeat the 20-question test and compare with day 1.
Do long buying cycles and local markets change the picture?
We do not know yet. We found no study that shows B2B outcome data, such as pipeline or revenue, from AI visibility, and none that shows how long buying cycles change its effect.
Our reasoning, labelled as opinion: with a six-to-eighteen-month cycle, AI answers are most likely to shape early research and shortlists. Your sales team still wins or loses the deal. So measure enquiry quality, not only volume.
For India, the UK and the Netherlands, check three things before copying a plan across markets.
Which AI assistants and search engines do your buyers use, and in which language do they search? Which data-protection rules apply to any outreach you build on top? Our lead generation guide covers this.
