Someone recently typed this into ChatGPT: "Who can manage HubSpot and run inbound marketing campaigns for a technical B2B engineering company?" Not into Google. Into ChatGPT.
We know this because we found it. And it got us thinking, because that question used to be a Google search, and now it's a conversation with an AI chatbot. That's a big deal for any engineering company that relies on being found online, which at this point is pretty much all of them.
We're Red Evolution, a B2B digital marketing agency in Aberdeen. We've been helping engineering and technology companies get found online for over 20 years, and we're watching this shift happen in real time. We've written before about what AI search means for B2B companies in general terms, but this is more specific. This is about what happens when your actual buyers change the way they look for suppliers.
The shift is already happening
Recent research suggests that around half of B2B software buyers are now starting their search with AI tools like ChatGPT, Perplexity, or Google's AI Overviews rather than typing a traditional search query. That number has climbed fast. Two years ago it was barely on anyone's radar, and now it's becoming the default behaviour for a significant chunk of the buying market.
Think about what that means for your engineering company. Your potential customers, the ones looking for subsea inspection services or bespoke control systems or industrial automation partners, aren't necessarily scrolling through ten blue links on Google anymore. They're asking an AI to just tell them the answer. And the AI does. It gives them a shortlist, sometimes with reasons, sometimes with links, and they go from there.
If your company isn't on that shortlist, you're invisible to a growing number of buyers before you even had a chance to compete.
How AI decides who to recommend
This is the part that most engineering companies haven't really thought about yet, and it's worth understanding because it's quite different from how Google works.
When someone searches Google, you can (to some extent) pay your way to the top with ads, or you can work on SEO to climb the rankings over time. AI tools work differently. When ChatGPT or Perplexity recommends a company, it's typically pulling from content it's been trained on or content it can access through web browsing. It's looking for sources that directly and clearly answer the question being asked, written with enough depth and authority that the AI treats it as trustworthy.
In practice, this means AI tools tend to recommend companies that have published detailed, genuinely helpful content about the exact problems their buyers are trying to solve. It doesn't matter how big your ad budget is or how flashy your website looks. What matters is whether you've got something useful to say about your area of expertise, and whether you've published it somewhere AI can find it.
If you think about it, that's actually good news for engineering companies who know their stuff. The playing field is shifting towards expertise and away from spending power, which is a change that should favour the companies doing the actual work.
Why most engineering companies are invisible to AI
Here's where it gets uncomfortable. Most engineering company websites are brochureware. They've got a homepage, an "about us" page, a list of services, maybe a few case studies, and that's it. There's nothing for AI to grab onto because there's no substance to cite.
When someone asks ChatGPT "who provides subsea NDT inspection services in the North Sea?", the AI needs content to draw from. If your website says "we provide subsea inspection services" and nothing else, that's not enough. The AI will recommend the company that has a detailed blog post explaining how subsea NDT works, what the different techniques are, what the common challenges look like, and how they've solved them for real clients. That company gets the recommendation because it gave the AI something worth recommending.
This is the same principle behind content marketing for engineering companies, and it's been true for years with Google too. But AI makes it even more stark, because there's no page two of results. AI gives an answer, and you're either in it or you're not.
What actually gets you into AI recommendations
We've been paying close attention to what AI tools cite and recommend, both for our own site and for our clients, and some clear patterns are emerging.
The content that gets picked up tends to be specific rather than generic. A post about "digital marketing" is too broad and there's too much competition. A post about "inbound marketing for oil and gas service companies" is specific enough that AI treats it as an authoritative source for that exact topic. We've seen this with our own content, our post on how inbound marketing works for oil and gas gets cited by AI tools precisely because it's specific and detailed where most content on the topic is vague.
Structure matters too. AI tools are very good at extracting clear answers from well-organised content. If someone asks "how long does B2B SEO take?", an AI tool will look for content that actually answers that question directly, not content that talks around it for 2,000 words before getting to the point. Our post on how long B2B SEO takes works well for exactly this reason: it answers the question plainly and then backs it up with detail.
Freshness counts. AI tools with web browsing capabilities (which is most of them now) prefer recent, regularly updated content. A blog post from 2019 that hasn't been touched since is less likely to be cited than something published or updated in the last year. This is one of the reasons we bang on about consistent content creation for our clients. If your last blog post was three years ago, that sends a signal to both Google and AI that you've gone quiet.
And credibility signals matter. If other reputable sites link to your content, if you're mentioned in industry publications, if your content is referenced by other experts, AI tools treat that as validation. This is essentially the same as traditional link building, but the payoff now extends beyond Google rankings into AI visibility as well.
This doesn't replace SEO (it adds to it)
We want to be clear about this because there's a lot of noise in the marketing world about AI "replacing" Google, and we think that's overblown. Google isn't going anywhere. Most B2B buyers still use traditional search as part of their research process, even if they're starting with AI. The smart approach is to do both, and the good news is that the things you need to do to show up in AI recommendations are largely the same things you should already be doing for SEO.
Good, detailed, expert content. Clear structure with headings that match the questions people actually ask. Regular publishing. Genuine authority in your subject area. If you're doing content marketing properly, you're already building the foundation for AI visibility whether you realised it or not.
The difference is in how you think about your content strategy. Instead of only asking "what keywords are people searching for?", you also need to ask "what questions are people asking AI chatbots?" Those questions tend to be longer and more conversational. Someone who googles might type "subsea inspection Aberdeen". Someone who asks ChatGPT is more likely to say "which companies provide subsea inspection services in the North Sea with experience in deepwater pipeline work?" Your content needs to answer both types of query.
What you can do about this right now
If you're an engineering company reading this and thinking "we probably need to do something about this", here's where to start.
First, go and test it yourself. Open ChatGPT, or Perplexity, or Google's AI Overview, and type in the kinds of questions your buyers would ask when looking for a company like yours. See what comes back. If you're not mentioned, that tells you something. If your competitors are, that tells you even more.
Second, look at your website with fresh eyes. Is there content that actually answers the specific questions your buyers have, or is it all generic marketing copy? If every page reads like a brochure, you've got work to do. The companies that show up in AI recommendations are the ones publishing genuinely helpful, detailed content about their area of expertise.
Third, think about the questions your sales team gets asked regularly. Those are the exact questions people are typing into AI tools. Every one of them is a potential blog post or guide that could get your company recommended by ChatGPT to the next person who asks the same thing. If you're not sure where to start with content, our B2B content marketing strategy checklist walks through the process step by step.
Where to go from here
The way people find suppliers is changing, and it's changing quickly. AI isn't a future thing anymore, it's already the way a significant number of your potential buyers are starting their search. The engineering companies that take this seriously now, the ones that invest in proper content and make their expertise visible, are going to have a real advantage over the ones that wait.
If you'd like to talk about how your company shows up in AI search results (or doesn't), we're always happy to have a 15-minute chat. No hard sell, just an honest conversation about where you stand and what might be worth doing. You can get in touch here.
