Where Middle-Market Companies Can Win AI Recommendations

The Signal Harbor Journal

Research and perspectives from Signal Harbor

By Sebastian Miller
Co-Founder, Signal Harbor

Where Middle-Market Companies Can Win AI Recommendations

Large companies bring familiar names, larger websites, more reviews, and years of public coverage into AI search. Middle and upper middle market companies often have real strengths that are less visible online.

A buyer, however, rarely needs “the biggest company.” The buyer needs a provider that fits a particular problem.

That creates an opening for a company with a clear specialty and credible evidence to support it. It does not erase a larger competitor’s advantage. It makes the choice of buyer questions especially important.

Traditional Search Authority Does Not Transfer Perfectly

SEO, or search engine optimization, helps companies earn visibility in search results. Brand recognition, content budgets, links, and established domains have often helped large companies build that position.

AI answers do not always draw on the same sources. In a 2026 study of 11,500 questions, Riley Grossman and colleagues compared traditional Google results with Google AI Overviews and Gemini. The systems drew on substantially different sources, and traditional search was more likely to surface popular or institutional websites in that benchmark.

This does not prove AI favors middle-market companies. A less prominent source is not necessarily a specialist supplier. It does show why a traditional search ranking and an AI recommendation should be measured separately.

The study also found that AI Overview sources were less consistent across repeated results. One favorable answer does not establish a durable position.

Familiarity Still Gives Incumbents an Advantage

In a controlled 2026 product experiment, Xi Chu and Yupeng Hou tested recommendations involving a familiar brand and fictional alternatives. When the products had the same specifications, the familiar brand had a strong advantage. When the alternatives had meaningful differences, recommendations could change.

That was a product experiment, not a formula for B2B purchasing. For a middle-market company, the useful question is: Where do our actual strengths make us a better fit, and can a buyer verify them?

Broad claims such as “high quality” and “industry-leading” give a buyer little to evaluate. Specific capabilities, requirements, and proof make the distinction clearer.

The Best Opening Is Usually Specific

Imagine a middle-market manufacturer that specializes in short production runs with demanding documentation. Asking AI “Who are the top manufacturers?” invites a broad shortlist of familiar names.

A buyer may ask something more useful: “Which Midwest suppliers can handle a small run of custom precision parts and provide the documentation our procurement team needs?”

That is an unbranded buyer-intent question. It describes a need without naming a provider. The specialist has a reason to enter that conversation if its capabilities and constraints are clearly explained and supported.

The same approach can help a private equity firm assess its portfolio. Each company has different customers and strengths. The goal is to find the buying decisions where those strengths matter, then test whether AI recognizes them.

Evidence Has to Support the Claim

The market calls work aimed at improving visibility in AI answers GEO, or generative engine optimization. In a foundational GEO study, Pranjal Aggarwal and colleagues found that changes to source content affected visibility in their benchmark, with results varying by subject. Those results are not a promised improvement for a live company.

Useful evidence is concrete: documented capabilities, specific use cases, legitimate customer proof, relevant certifications, current product information, and credible descriptions beyond the company’s own site.

Signal Harbor reviews the sources cited in tested answers. A citation tells us which page appeared in a response; it does not reveal every step behind a recommendation. We examine whether a company’s strengths are visible, whether a competitor’s evidence is clearer, and whether the AI’s claims are accurate.

Measure the Opening Before Investing in It

An AI visibility score describes results within a particular set of questions. It cannot tell an executive whether the company is winning every buyer conversation in its market.

We test questions tied to customer fit, compare the answers with competitors, check accuracy and sources, and repeat the tests. Then we give marketing teams priorities for content and evidence. After changes are published, we retest to see whether the answers improve.

For middle and upper middle market companies, the goal is targeted visibility: being considered when a qualified buyer has a need the company is equipped to serve. Signal Harbor identifies those buying situations, measures where the company stands, and gives the team a clear place to act.

Sebastian Miller
Co-Founder, Signal Harbor
signalharborai.com
sebastian.miller@signalharborconsulting.com
402-306-2213