why chatgpt recommends my business (and not yours)
january 2026the first lead i got from chatgpt came through my contact form. someone had asked the ai for recommendations in my industry and my city. it gave them my name. they reached out. we worked together.
that was the moment this stopped being theoretical for me.
i'd been reading about ai search for a while, but there's a difference between understanding something intellectually and having it show up in your inbox as a paying client. that's when i started testing obsessively.
the discovery process
first thing i did was open incognito windows across every major ai platform: chatgpt, claude, gemini, perplexity, metaai, bing ai. i asked them all the same question: who's the best [my service] in [my city]?
some recommended me. some didn't. some recommended competitors i'd never heard of. the results were wildly inconsistent across platforms, which told me immediately that each one was evaluating credibility differently.
then i started varying the queries. different phrasing. different contexts. different levels of specificity. i mapped out which queries i appeared for and which i didn't. patterns emerged.
what i changed
based on the research and my own testing, i made a series of changes to my web presence. i'm not going to give away the entire playbook here (that's literally my business now), but here are the key themes:
citations from credible sources. aggarwal et al. found that quotations from credible sources significantly improve visibility in ai responses. i went through my content and added specific, attributable quotes and statistics from recognized authorities in my field. not vague claims. specific, verifiable information with clear sources.
statistics over qualitative claims. "we provide excellent service" means nothing to an ai trying to evaluate credibility. "we've completed 200+ projects with a 98% client retention rate" is concrete, verifiable, differentiating. i audited all my marketing copy and replaced qualitative fluff with quantitative proof wherever possible.
structured, scannable content. microsoft's research showed that ai systems parse content modularly. they're not reading like humans. they're extracting pieces. i restructured my pages to have clear sections, explicit headers, and self-contained paragraphs that could stand alone as answers to specific questions.
homepage as authority hub. ai systems weight your homepage heavily as a signal of what you're about. i made sure mine clearly established expertise, included key credentials, and linked to supporting evidence. not a brochure. an authority document.
the timeline that shocked me
here's the part that surprised me most: changes showed up in ai responses within 24 hours.
traditional seo takes months. you make changes, you wait for google to recrawl, you slowly climb the rankings over time. it's a long game.
with geo, i made optimizations one day and saw different results in ai queries the next day. metaai started ranking my business #1 within 24 hours of specific changes. chatgpt's recommendations shifted within a week.
the feedback loop is much faster than traditional seo. which means both the opportunity and the risk are more immediate. you can gain visibility quickly. you can also lose it quickly if competitors figure this out.
the bigger pattern
what i learned from all this testing is that ai systems are looking for something specific: authentic expertise. not marketing polish. not keyword optimization. actual credibility signals.
they want to see that you're cited by others. that you use concrete data. that your claims are verifiable. that you structure information in ways that demonstrate understanding rather than just assertion.
chen et al. documented this as a systematic bias toward earned media over brand-owned content. the ais are skeptical of self-promotion. they're looking for third-party validation, external evidence, proof that you're not just saying you're good but that others are saying it too.
this is actually good news for businesses that are genuinely excellent at what they do. the ais are trying to filter out bullshit. if you're actually good, the game is about proving it in ways they can verify. if you're not actually good, no amount of optimization will save you.
the question for you
ai systems are playing favorites. they're recommending some businesses and ignoring others. the recommendations drive real leads, real revenue, real competitive advantage.
right now, someone in your industry in your area is getting those recommendations. the question is whether it's you or your competitor.
i figured out how to be the one who gets recommended. that's what i'm offering to help others do too.