Two-thirds of UK senior decision-makers with B2B purchasing power now use AI tools like ChatGPT, Copilot, and Perplexity to research and evaluate suppliers. That's from a 2025 survey of 300 UK professionals by Magenta Associates. Among younger buyers (25-34), the figure is 85%.
If you run an engineering or manufacturing company, that statistic should get your attention. Your next buyer might not start their search on Google. They might ask ChatGPT: "Who are the best suppliers of compact heat exchangers in the UK?" or "Which companies offer hot tapping services for offshore pipelines?" And if your business isn't in the answer, you don't exist in that conversation.
So what does this actually mean if you run an engineering or manufacturing business? What should you do about it? And how much of the hype around "generative engine optimisation" is real versus marketing agencies looking for something new to sell?
Here's what matters.
What's Actually Happening
There are two separate shifts going on, and the difference matters.
The first is Google's AI Overviews. When you search on Google now, many queries trigger an AI-generated summary at the top of the results page. An Ahrefs study of 300,000 keywords in December 2025 found that AI Overviews correlate with a 58% reduction in click-through rates for the top-ranking page. Seer Interactive's analysis of 25.1 million organic impressions found organic CTR dropped 61% when AI Overviews appeared. That's a big shift, but it's still happening inside Google. Your website can still be cited in those summaries, and being cited correlates with 35% higher organic click-through rates compared to not being cited.
The second shift is people leaving Google altogether. ChatGPT now has over 800 million weekly active users. Perplexity is growing fast. Gartner projects that traditional search engine volume will drop 25% by 2026 as users move to AI instead. When a procurement manager asks Perplexity to compare suppliers or evaluate a product category, the platform synthesises information from multiple sources and gives a direct answer, complete with citations. On average, these platforms cite just two to seven sources per response, not the ten blue links you'd get from Google. Getting cited is harder, and the penalty for being left out is absolute: you're either recommended or you're invisible.
Why Engineering Companies Are Better Positioned Than They Think
Almost everything written about AI search focuses on SaaS, tech companies, and consumer brands. That makes sense: those are the sectors being hit hardest by AI Overviews sucking away informational clicks. A query like "what is CRM software" can be answered without anyone visiting a website. The AI gives a summary, the user moves on.
Engineering and manufacturing searches work differently.
When a procurement manager asks ChatGPT about ASME-compliant heat exchangers for a high-pressure application, or a project engineer researches hot tapping procedures for a live subsea pipeline, the platform can't generate that answer from generic web content. It needs to cite sources with real technical authority. Businesses that have been producing detailed, technically accurate content for years — datasheets, application guides, technical case studies, design specifications — are exactly the sources these systems need to reference.
In the SaaS world, ChatGPT can synthesise a perfectly adequate comparison of project management tools from hundreds of review sites and blog posts. In engineering, there's no substitute for the company that actually designs, manufactures, or services the equipment. Your technical content has a depth that AI can't replicate from scratch. It has to cite you.
One study from Whitehat SEO found that AI Overviews appear for 54% of B2B keywords versus just 22% of B2C keywords. B2B informational content sits directly in the path of these changes. But the same study found that traffic from AI citations converts better than traditional search traffic. When someone clicks through from an AI-generated answer, they've already been pre-qualified by the summary. They're further along in the buying process.
For an engineering business with high contract values and long sales cycles, that shift in traffic quality could be worth more than raw volume ever was.
What AI Search Gets Wrong About Your Industry
There's a catch. ChatGPT and Perplexity are only as good as the sources they can find, and engineering and manufacturing businesses have some specific problems here.
A lot of the best technical information in engineering sits behind login walls, inside PDF datasheets that aren't indexed, or on pages that are blocked from AI crawlers by misconfigured robots.txt files. If your technical resources require registration to access, ChatGPT and Perplexity can't read them. If your datasheets are scanned PDFs with no text layer, the crawlers can't extract the information. Your competitors whose content is openly accessible will be cited instead.
There's another limitation. Ask ChatGPT to recommend a heat exchanger supplier and it will give you a list, but it can't evaluate whether a specific design is appropriate for your operating conditions, your material compatibility requirements, or your maintenance access constraints. The answer is a starting point, not an engineering recommendation. Buyers know this. But they'll use that answer to build a shortlist, and if you're not on it, you won't get the chance to demonstrate your technical capability.
There's also a concentration problem. Research from Magenta Associates found that just five brands capture 80% of top AI-generated responses in any given B2B category. That's more extreme than Google ever was, where at least you had ten organic results and multiple pages to appear on. It's winner-takes-most: the top few cited companies get the attention and everyone else gets nothing.
What You Should Actually Do
I need to be straight with you, because there's a growing industry of consultants selling "generative engine optimisation" (GEO) as though it's a completely new discipline. For most engineering businesses, what works for AI visibility is the same thing that works for traditional SEO done properly. The fundamentals haven't changed. The emphasis has shifted.
Make your technical content accessible. If your best technical content is locked behind a registration form or buried in unindexed PDFs, it's invisible to ChatGPT, Perplexity, and Google alike. Get your datasheets, application guides, and technical specifications into crawlable HTML pages. You can still gate premium content like full design manuals, but the information that helps buyers evaluate you during the research phase needs to be open. This is the single biggest lever most industrial companies can pull.
Check your robots.txt. ChatGPT's crawler is called GPTBot. Perplexity uses PerplexityBot. Google uses Googlebot. If your robots.txt file blocks any of these, your content doesn't exist as far as those platforms are concerned. A surprising number of engineering websites block AI crawlers without realising it, sometimes because a CMS update or security plugin added the rules automatically.
Structure your content so it can be extracted. ChatGPT and Perplexity pull fragments from pages, not whole articles. Each section of your content should make sense on its own. Use clear headings that state what the section covers. Include specific data points: dimensions, operating ranges, material grades, compliance standards. Concrete, verifiable information beats vague marketing language every time. If an AI is comparing suppliers and your page says "we offer high-quality heat exchangers" while a competitor's page says "compact fin heat exchangers rated to 450°C and 250 bar in Inconel 625," the competitor gets cited.
Build your third-party presence. AI doesn't just read your website. It looks for consensus across multiple sources. If your company is mentioned in trade publications, industry directories, and partner websites, that creates multiple signals that you're a credible supplier. Businesses with decades of backlinks from trade bodies, industry associations, and technical publications have a built-in advantage here. Make sure your company information is consistent and current across all these sources.
Invest in content that demonstrates first-hand experience. Google's E-E-A-T framework (Experience, Expertise, Authoritativeness, Trustworthiness) matters more now than ever. ChatGPT and Perplexity cite content from sources they consider authoritative. Generic blog posts that any marketing agency could write won't get cited. Technical content that shows you've actually worked on these problems — case studies with real project details, application guides based on your engineering team's experience, technical articles that go beyond what's in a textbook — will. The bar for getting cited is higher than the bar for ranking on page one, and that plays to your strengths.
Don't abandon traditional SEO. Over 92% of AI Overview citations come from domains already ranking in Google's top 10. Traditional search performance and AI citation performance are closely correlated. If you rank well on Google, you're far more likely to be cited by ChatGPT and Perplexity too. The businesses doing well in AI visibility are, almost without exception, the ones that were already doing SEO properly. Think of AI visibility as a layer on top of your search strategy, not a replacement for it. If you want to understand why engineering SEO requires a specialist approach or how long results take to show, those posts cover the foundations in detail.
What You Can Safely Ignore (For Now)
Right now, the noise around AI visibility is deafening. A few things that don't need to keep engineering company owners up at night in 2026.
You don't need to buy AI visibility monitoring tools. These are designed for SaaS companies tracking whether they appear in ChatGPT responses for queries like "best CRM for healthcare." In engineering, the number of relevant queries is much smaller and more specific. You can test your visibility yourself by asking ChatGPT and Perplexity the questions a buyer would ask. Do that quarterly. Don't spend thousands on a platform to automate it.
You don't need a separate GEO strategy. Good SEO with technically accurate content, proper site architecture, and strong E-E-A-T signals is your GEO strategy. Agencies selling GEO as a distinct service for industrial businesses are, in most cases, repackaging work you should already be doing and charging extra for the buzzword.
You don't need to panic about zero-click searches. Yes, AI Overviews reduce clicks for informational queries. But engineering buying decisions aren't made from a one-paragraph AI summary. Procurement teams are specifying equipment for projects that take months to source. They'll still click through, download datasheets, and request quotes. Buyer behaviour that drives your revenue hasn't been replaced by an AI summary. What has changed is how buyers build their initial shortlist, and that's what you need to focus on.
The Real Risk
The risk isn't that AI search will destroy your business overnight. The risk is slower and harder to spot.
Over the next two to three years, a growing proportion of buyers will start their research in ChatGPT or Perplexity rather than Google. The suppliers that appear in those responses will make the shortlist. The ones that don't will receive fewer enquiries, and they won't understand why because their Google rankings might look fine.
For engineering and manufacturing businesses, this creates a quiet competitive advantage for those who get it right early. Most of your competitors haven't thought about AI visibility at all. If your technical content is accessible, well-structured, and backed by real expertise, you'll be among the small number of companies these platforms cite for your product category. And because the attention concentrates on just a handful of suppliers per query, that early positioning compounds.
This isn't a revolution that requires you to tear up your marketing plan. It's an evolution that requires you to do the things you should already be doing (quality technical content, proper site architecture, consistent third-party presence) and do them with AI discoverability in mind.
Where to Start
If you want a quick sense of where you stand, try this: open ChatGPT or Perplexity and type the query your ideal buyer would ask when looking for what you sell. Not your company name. The product or service they need. See what comes back. See who gets cited. If it's your competitors and not you, that's your starting point.
Then look at the practical list above. For most industrial businesses, the biggest wins come from making existing technical content accessible and properly structured. You probably already have the expertise and the content. It just needs to be findable.
We've been helping engineering and manufacturing companies with their search visibility since 2003, and we're seeing this shift play out across our client base. If you want an honest assessment of where your business stands — in both traditional search and AI search — a free video audit is a good place to start. We'll tell you what's working, what's missing, and what to prioritise.
Frequently Asked Questions
How does AI search affect engineering companies differently from other industries?
Engineering search queries require technical depth that AI tools can't generate from generic sources. AI systems need to cite companies with genuine expertise, which gives engineering businesses with strong technical content a natural advantage. However, AI Overviews appear for 54% of B2B keywords compared to just 22% for B2C, so engineering companies are more exposed to the click-through rate reductions that AI search causes.
Should engineering companies invest in generative engine optimisation?
For most engineering companies, good traditional SEO with technically accurate content and proper site architecture covers the majority of what GEO requires. The foundations are the same: authoritative content, structured data, strong E-E-A-T signals, and third-party mentions. Buying a separate GEO service is unnecessary if your SEO strategy already addresses these areas.
What's the biggest AI search mistake engineering companies make?
Locking technical content behind registration walls or leaving it in unindexed PDF formats. AI crawlers can't access gated content or extract information from scanned PDFs. If your datasheets, application guides, and technical specifications aren't in crawlable HTML, AI tools will cite your competitors instead.
Are B2B buyers really using ChatGPT to find suppliers?
Yes. Research from Magenta Associates shows 66% of UK senior decision-makers with B2B purchasing power now use AI tools to research and evaluate suppliers. Among buyers aged 25-34, the figure rises to 85%. Separately, Forrester found that 89% of B2B buyers have used generative AI during their purchasing process.
Will AI search replace traditional SEO for engineering companies?
No. Over 92% of AI Overview citations come from domains ranking in Google's top 10. Traditional search performance and AI citation performance are closely linked. Engineering companies should treat AI search visibility as an additional layer on top of their existing SEO strategy, not a replacement for it.
