Manufacturing executives are beginning to ask me a question I expect we'll hear a lot more: “How do I know if my company is showing up in AI?” With Google, we're pretty good at answering that. Tools can tell you how your website ranks for keywords, how rankings change and which searches bring people to your website.
AI is different.
There isn't one AI. A prospective customer might use ChatGPT, Gemini, Google's AI search experience, Claude or another platform and get different answers to essentially the same question. We also don't have the mature measurement tools we've had for Google search. AI visibility tools are emerging, but I don't think we're at the point where a manufacturer can look at one report and say, “This is exactly how we're ranking in AI.”
For now, some of this has to be done manually. Maybe that changes. Maybe it doesn't. I don't know. But you don't need to wait for a perfect measurement system to learn a lot.
Start With the Questions Buyers Actually Ask
Broad questions matter. An engineer might begin with, “Who are the best manufacturers of X?” That could determine which companies make the initial list. For relatively inexpensive products with shorter sales cycles, a recommendation at that stage could be particularly important.
For complex manufacturing purchases with longer sales cycles, that's probably only the beginning. As the research progresses, questions become more specific. An engineer may need a manufacturer that can work with a certain material, hold a particular tolerance, perform an unusual secondary operation or solve a specific production problem.
So I wouldn't measure AI visibility only by whether your company appears on a “best manufacturers” list. You want to know whether you're visible throughout the questions a serious buyer is likely to ask.
Start with questions buyers ask your salespeople and engineers. Look at RFQs. Then ask those questions across multiple AI platforms. Which companies appear? Which sources are cited? Does your company appear when the question relates directly to something you're particularly good at?
AI Doesn't Have a Number-One Ranking
This is another important difference from Google. We're accustomed to rankings. You're number one for a keyword, number four or number 17. We can track that position and watch it move.
AI doesn't work that neatly. The wording can change the answer. Different platforms can produce different answers. Follow-up questions, personalization and previous conversations may also influence what a user sees.
Don't make a major business decision based on five searches from your everyday AI account. Test a meaningful group of buyer questions and look for patterns. If the same competitor appears repeatedly across important questions and your company doesn't, that's worth investigating.
Look at Citations, Not Just Recommendations
Sometimes AI recommends a company. Other times it answers a technical question and cites a manufacturer's website as a source. I pay attention to both.
If AI repeatedly uses a competitor's website to answer questions about an area where your company has expertise, ask why. What does their website contain that yours doesn't? Maybe they have an in-depth technical article, answered a question you haven't, explain a process more clearly or identify their subject-matter experts. You may not know exactly why AI selected that source, but the comparison can be useful.
And ask AI itself. If it recommends competitors instead of your company, ask why and what those companies demonstrate online that yours doesn't. Don't treat its response as a definitive explanation of the algorithm. It isn't. But it can identify gaps worth investigating.
Separate Visibility From Vanity
The goal isn't to get mentioned by AI as often as possible. I'd rather see a manufacturer appear for 20 highly relevant questions than 200 broad questions that will never produce business.
A buyer may start by asking for the “best aluminum casting companies.” Later, that same buyer may need an aluminum casting company that can perform pressure testing. That's a very different question. If pressure testing is an important capability for your company, that's where you want your expertise to be visible.
Build a Baseline
Build a representative group of questions that matter to your business. Run them across several AI platforms. Record whether your company appears, which competitors appear and which websites are cited.
Keep the testing environment as clean as possible. If you regularly use an AI platform, it may already know about you and your company, which can potentially influence its answers. For testing, I prefer a separate account that hasn't been used for day-to-day work. Where possible, turn off memory or personalization and start with a new conversation.
The objective isn't a perfectly controlled scientific experiment. I don't think that's realistic. It's simply to reduce the chance that AI is recommending your company because it already knows who you are. Whatever method you choose, use it consistently when you repeat the test.
Now you have a baseline. Over time, look for patterns. Are you appearing across more questions that matter? Are you being cited where you have genuine expertise? Are competitors consistently owning areas where you should be credible?
Decide Whether You're In or Out
According to GlobalSpec's 2026 State of Marketing to Engineers research, 69 percent of technical buyers are already using generative AI during the purchasing process. Manufacturers don't need to panic or chase every new AI optimization tactic. But they should understand what's happening.
Your buyers are doing their research online, and increasingly they're using AI as part of that process. So find out what they see. It isn't as clean as looking at a Google ranking report. At least not yet. But imperfect measurement is better than not looking at all.
If you believe this is where your buyers are headed, then you have to decide whether you're in or you're out. And if you're in, the goal isn't simply to give AI more reasons to find you. It's to give your buyers more reasons to trust you.
Bob Engelhart is president of Cobalt Sales Group, an industrial marketing agency.
