A prospective customer asks ChatGPT, Gemini or Google a question about a manufacturing problem. Your company doesn't show up. One of your competitors does.
Why?
I wish I could give manufacturers a formula. I don't think one exists—or at least not one that anyone outside these companies really knows.
Google has published guidance for succeeding in AI search, and one of the most important takeaways is that the fundamentals aren't radically different from what it has been telling website owners for years: create useful, original content for people, demonstrate real expertise and make sure your website can be understood and accessed by search engines.
That sounds simple. In practice, a lot of manufacturers aren't doing it.
AI Can't Walk Through Your Facility
One of the biggest challenges manufacturers have is that much of what makes them exceptional is inside their building. If I walked through your plant with your best engineer, I might learn about a process you've perfected over 20 years. I might see a machine you've modified to do something your competitors can't. I might learn that you can hold a difficult tolerance, work with an unusual material or solve a problem that causes other suppliers trouble.
A prospective customer touring your plant could learn those things, too. AI can't. Neither can a buyer sitting at a desk researching potential suppliers. The only way they can know about that expertise is if you make it available to them.
That's where I think many manufacturing websites fall short. They tell visitors what the company does, but they don't demonstrate what the company knows. There's a big difference.
Listing a Capability Isn't Demonstrating Expertise
Look at enough manufacturing websites and you'll see many of the same claims: quality, service, innovation, decades of experience and state-of-the-art equipment. Then there is usually a list of capabilities. That's useful information, but it doesn't necessarily give a buyer—or an AI system—much reason to distinguish one manufacturer from another.
Suppose two manufacturers both say they can perform a particular process. One has a paragraph on a capabilities page saying it offers the process. The other has answered the questions an engineer would actually ask about it:
- What materials work best?
- What tolerances can be achieved?
- What are the limitations?
- What affects cost?
- When should a different process be considered?
- What commonly goes wrong?
- How does the company solve those problems?
Which company has done a better job demonstrating expertise? I think the answer is pretty obvious. It really isn't about what you say. It's about what you can demonstrate.
Start With the Questions Buyers Are Asking
One of the most useful things manufacturers can do is stop thinking primarily about keywords and start thinking about questions. What does a buyer or engineer need to know before deciding whether to contact you?
Those questions may be surprisingly basic. These aren't necessarily the questions companies like answering on their websites. Pricing is a good example. Manufacturers often resist discussing price because every project is different. That's understandable. But buyers are still asking the question.
You don't necessarily need to publish a price list. You can explain what drives the price. You can show the variables. You can give examples. You can build a calculator.
We built a pricing calculator for Pulse Drying Systems, which makes spray drying systems that run between $250,000 and $5 million. Obviously, there can be some modifications that make the price vary that are not factored into the cost calculator, but we make it very clear that this is a budgetary number, not something that they could submit a PO with. It just gives them something so they know whether it's a good fit. The objective is to become genuinely useful during the buyer's research process.
Make the Expertise Identifiable
Who wrote the information matters, too. Google's people-first content guidance encourages clear authorship and information about the people behind content when readers would reasonably expect it.
That makes sense beyond search rankings. If I'm reading a technical article about a difficult manufacturing process, I'd rather know that the information came from an engineer who has spent 20 years working with that process than from an anonymous marketing department.
Manufacturers have tremendous subject-matter expertise inside their companies. Often, none of it is visible online. You don't need to turn your engineers into writers. Interview them. Capture what they know. Put their names and backgrounds on the resulting content. Let the expertise that's already inside your company become visible outside the building.
I see a lot of discussion about optimizing content specifically for AI. Some of it may prove useful, but I'd be very cautious about anyone claiming to have discovered a secret formula. Google itself says its generative AI features are rooted in its core search ranking and quality systems and that established SEO best practices remain relevant.
That doesn't mean nothing has changed. The way people search is changing dramatically. Instead of typing a few words into Google and opening ten blue links, buyers can have a conversation with AI. They can describe a specific problem, add requirements and ask follow-up questions.
That creates opportunities for manufacturers with deep expertise. A manufacturer doesn't have to be the biggest company in its industry to provide the best answer to a very specific technical question. But the expertise has to exist somewhere AI can find it.
The companies I think are best positioned for this aren't trying to create “AI content.” They're creating useful resources for customers and prospective customers. They're answering real questions. They're demonstrating expertise. They're identifying the people behind that expertise. And they're making sure the technical fundamentals of their websites are sound.
We don't know exactly how every AI system decides which companies to cite or recommend, and I would be skeptical of anyone who says they do. But we do know something much more useful.
If the knowledge that makes your company exceptional exists only inside the heads of your employees and inside the walls of your plant, neither your prospective customers nor AI can use it. The first step is getting that expertise out of the building.
Bob Engelhart is president of Cobalt Sales Group, an industrial marketing agency.
