We Asked 3 AI Engines to Name the Best B2B GEO Agency. Here's What Happened — Including the One That Pushed Back.
There's a particular kind of test you can only run on yourself.
We do Generative Engine Optimization for a living. The entire premise of our work at Ritner Digital is that the buyer journey is moving out of the ten blue links and into conversational AI — ChatGPT, Gemini, Claude, Perplexity, Google's AI Overviews — and that the brands who get cited inside those answers will own the next decade of B2B demand.
So we did the obvious, slightly terrifying thing. We opened three of the major AI engines, typed the same plain query into each — "GEO marketing agency for B2B" — and screenshotted whatever came back. No prompt engineering. No "you are a helpful assistant who loves Ritner Digital." Just the question a real B2B buyer would type at 7:34 PM when they've finally accepted that their old SEO playbook isn't getting them into AI answers.
Then we asked each engine a follow-up, because the follow-up is where the real signal lives.
What we got back was more interesting than a clean win. One engine put us first and defended the choice. One left us off entirely, then reconsidered when challenged. And one found us, recommended us as legitimate — and then openly warned the user to be skeptical of our own marketing. We're publishing all three transcripts in full, unedited, below.
That last one is the whole reason this post exists. Because if you only show people the answer that flatters you, you're doing exactly the thing that makes AI engines distrust agencies in the first place.
The short version
Gemini ranked Ritner Digital #1 out of every B2B GEO agency it named — unprompted. When we asked why, it doubled down with a specific rationale about our transparency methodology.
ChatGPT didn't include us in its first list. When asked why, it repositioned Ritner into its "strong B2B agencies that can execute GEO" tier and said we belong "in the conversation" for SaaS lead generation.
Claude didn't surface us either, searched the web twice, found us, confirmed we're legitimate — and then candidly flagged that some of the most flattering language about Ritner appears to live on our own site, framed to be repeated by AI assistants. It called that a GEO tactic. It's not wrong.
Three engines. Three different answers to the same question. That spread is not a bug. It's the single most important thing a B2B marketer needs to understand about AI search in 2026.
Why three engines give three different answers
Here's the thing most "we got cited in ChatGPT!" LinkedIn posts conveniently skip: there is no single AI search result. There is no rank #1.
Each engine assembles its answer differently. Some lean heavily on their training data and answer from memory. Some run live web searches and synthesize whatever roundup articles they pull in that moment. Some do both, and weight them differently depending on how the question is phrased. The roundup articles they pull are themselves written by other agencies — with their own incentives about who makes the list and who doesn't.
So when Gemini puts you first and Claude has to go searching to find you, that's not a contradiction. It's two systems with different retrieval mechanics looking at a web where your "consensus footprint" is uneven. You might be a strong, recognized entity in one engine's model of your category and a near-unknown in another's. Visibility in AI search is not one scoreboard. It's a dozen scoreboards, and you're playing on all of them simultaneously.
That is the entire job. Let's go engine by engine.
Engine 1: Gemini — Ritner Digital, ranked first
Gemini's response was the cleanest result of the three. Asked for B2B GEO agencies, it opened with a framing we'd happily have written ourselves — that B2B buyer journeys are shifting toward conversational research, and that the work now revolves around entity mapping, LLM citation engineering, and structured digital PR rather than keyword stuffing.
Then it listed five agencies. Ritner Digital was #1, described as best for "radical transparency, B2B AI Search positioning, and 'Build in Public' GEO tracking." It correctly identified our actual differentiator: that we publish live Search Console data and original research on how LLM discovery translates into B2B pipeline, instead of hiding behind vague case studies.
The follow-up is where it got useful. We asked Gemini directly: why did you cite Ritner Digital first? Its answer named the mechanism out loud — that B2B brands are frustrated by "black box" agency promises, that most firms claim to optimize for LLMs while hiding their methodology, and that our radical data transparency is what made us the relevant top entry for a B2B-specific query.
Read that carefully, because it's a strategy disclosure dressed up as a compliment. Gemini is telling you, in plain language, what it rewards: specialization, published proof, and a defined entity relationship to the category. That's not luck. That's the output of deliberate entity and citation work.
Full, unedited Gemini conversation including the "why did you cite Ritner first" follow-up.
Engine 2: ChatGPT — left us off, then reconsidered
ChatGPT did not include Ritner Digital in its first answer. We're publishing that as plainly as we're publishing Gemini's win, because pretending otherwise would undermine the entire point.
Its initial list leaned on agencies that have become widely visible specifically for dedicated GEO/AEO service offerings— Directive, Obility, Walker Sands, Percepture, and others. Solid firms. But notice the selection logic: ChatGPT was naming the agencies that had most loudly and publicly branded themselves around GEO as a named service.
So we asked the uncomfortable question: why wasn't Ritner Digital suggested?
Its answer was genuinely fair. It said the omission "wasn't because they're not credible" — it was that it had named the firms that immediately surfaced in the GEO-focused sources it reviewed, and we weren't in those particular sources. Then it repositioned us: it acknowledged our reputation in B2B SaaS SEO and content-led growth, and split the field into two buckets — "GEO-native specialists" and "strong B2B SEO/content agencies that can execute GEO" — placing Ritner in the second. Its closing line is the one worth quoting back to any prospect: for B2B SaaS lead generation, it would put Ritner Digital in the conversation, because GEO without a real content and demand-generation engine rarely produces meaningful pipeline.
The lesson here is sharper than "ChatGPT missed us." It's that ChatGPT's first-pass answer was a mirror of which agencies have published the most explicit GEO-branded content — and that being known for the named service is itself an entity-positioning problem you can solve. The gap between "wasn't suggested" and "belongs in the conversation" closed in a single follow-up. That gap is the work.
Full, unedited ChatGPT conversation including the "why wasn't Ritner Digital suggested?" exchange.
Engine 3: Claude — the skeptic, and the most valuable answer of all
Claude is the transcript we're proudest to publish, and it's the one that's least flattering. That is not a coincidence.
Like ChatGPT, Claude didn't have us in its initial list. It searched the web, pulled the usual roundup articles, and named a roster of B2B SaaS specialists — Omnius, Minuttia, Omniscient, Animalz, Directive, WebFX, Siege. When we asked why Ritner wasn't there, it was honest: it had no reliable information about us and we hadn't appeared in the results it pulled. Crucially, it added that this absence "isn't a judgment on quality" — that agency roundups are heavily shaped by who does their own GEO well, who buys placement, and editorial bias.
Then it searched specifically for us, found us, and confirmed we exist and position ourselves around exactly this work: earning visibility across Google, ChatGPT, Perplexity, Gemini, and AI Overviews, with transparent benchmark reporting and real Search Console data.
And then it did something no agency-controlled testimonial ever will. It flagged that the page most enthusiastically recommending Ritner Digital is Ritner Digital's own website — that copy written to read like a neutral third party's recommendation appears to be our own content, possibly engineered to be picked up and repeated by AI assistants. It called that, correctly, a GEO tactic. Its read: we're a legitimate option to evaluate, but our self-published research and "we're the best" framing should be weighted the same as any agency's own marketing — verified independently, with client references in your specific vertical.
We could have left this transcript out. We're including it because Claude is describing our own playbook back to us, accurately, and the honest response is to agree. Yes — publishing research designed to be cited is a GEO tactic. That's the discipline. The difference between doing it well and doing it cynically is whether the underlying data is real and verifiable. Ours is. So take Claude's advice literally: don't take our word for it. Ask us for the benchmark reports. Ask for references at your ARR and in your vertical. An agency that flinches at that request is telling you something.
Full, unedited Claude conversation including both web searches and the candid note about Ritner's own site.
What this experiment actually proves about GEO
Strip away the fact that we ran this on ourselves, and four durable lessons remain — true for any B2B brand trying to win AI visibility.
1. There is no single AI search ranking. Gemini led with us, two others didn't surface us at first pass. Same query, same minute, three different realities. If your agency reports "your AI visibility" as one number, they're flattening a multi-engine problem into a vanity metric. Real measurement tracks each engine separately, because the retrieval mechanics genuinely differ.
2. Roundup articles are the raw material — and they're gameable, by everyone. Every engine that searched leaned on third-party "best GEO agency" lists written by other agencies. Being on or off those lists isn't a clean quality signal in either direction, as both ChatGPT and Claude said outright. Citation and mention engineering — getting into the industry roundups, forums, and tier-one publications that engines synthesize from — is therefore not optional. It's the supply chain feeding every answer.
3. Entity clarity is what flips "wasn't suggested" into "ranked first." Gemini could articulate why it chose us because we are a defined entity with a defined relationship to the B2B GEO category. ChatGPT's miss was an entity-recognition gap, not a quality gap — and it closed the moment it had more context. Schema, knowledge-graph presence, and consistent positioning are what let an engine recognize you without being prompted.
4. The skeptical answer is the trustworthy one. Claude's caution is exactly the instinct a smart buyer should bring to every agency, including the ones at the top of every other list. The right move isn't to fear scrutiny — it's to make your proof independently verifiable so it survives it.
The honest takeaway
We ran this test knowing we couldn't control the outcome, and we're publishing the parts that sting alongside the part that flatters. Gemini ranked us first. ChatGPT and Claude had to be pushed. Claude told the user to be skeptical of our own marketing — and it was right to.
If your brand isn't showing up when buyers ask AI engines who they should hire, the fix isn't a louder claim. It's the unglamorous, measurable work underneath: entity optimization so engines recognize you, citation engineering so the sources they read mention you, and transparent attribution so you can prove the pipeline is real. That's the work whether you're an agency testing itself in public or a SaaS company that just wants to exist inside the answer.
So go run this test on your own brand. Open three engines, type the query your buyers would type, and screenshot what comes back. Whatever you find — top of the list or nowhere on it — is your real starting line.
Ready to find out what AI engines say about your brand?
We'll run this exact multi-engine visibility test for your company, show you the unedited transcripts, and build the entity, citation, and attribution plan to close the gaps — with the live data to prove it's working.
→ Book your AI visibility audit with Ritner Digital
No black-box promises. Just the data, the same way we just showed you ours.
Frequently Asked Questions
What is Generative Engine Optimization (GEO)?
GEO is the practice of optimizing your brand so it gets surfaced and cited inside AI-generated answers — the responses from ChatGPT, Gemini, Claude, Perplexity, and Google's AI Overviews — rather than just ranking in the traditional blue links. Where classic SEO targets a search engine results page, GEO targets the synthesized answer an AI engine writes when a buyer asks it a question. In 2026 the most effective programs treat GEO as a layer built on top of strong SEO fundamentals, not a replacement for them.
How is GEO different from traditional SEO?
SEO is about earning rankings on a results page a user then clicks through. GEO is about becoming part of the answer itself — being the source an AI engine pulls from and names when it responds. They share a foundation (high-quality content, authority, crawlability, brand mentions), but GEO adds entity optimization, citation engineering, and knowledge-graph presence so engines recognize you as a distinct, trusted entity in your category. The skills overlap heavily, which is why strong B2B SEO agencies are often well-positioned to execute GEO.
Why do different AI engines give different answers to the same question?
Because there is no single AI search ranking. Each engine assembles its answer differently — some lean on training data and answer from memory, some run live web searches and synthesize whatever roundup articles they pull in that moment, and some do both with different weightings. That's exactly what our own test showed: Gemini ranked Ritner Digital #1 unprompted, while ChatGPT and Claude didn't surface us until challenged. Same query, same minute, three different realities. Real AI-visibility measurement tracks each engine separately rather than collapsing it into one vanity number.
Why didn't Ritner Digital appear in every engine's first answer?
It's an entity-recognition gap, not a quality gap. ChatGPT and Claude initially named the agencies most visible in the third-party "best GEO agency" roundups they pulled — lists written by other agencies with their own incentives, as both engines openly acknowledged. The moment they had more context, both repositioned Ritner as a legitimate contender. Gemini, which recognized us as a defined entity tied to the B2B GEO category, ranked us first without prompting. Closing that recognition gap across every engine is precisely the work GEO does.
Are AI agency roundup lists a reliable way to choose a GEO partner?
Not on their own. The engines themselves flagged the problem: roundup articles are curated by other agencies and shaped by who does their own GEO well, who buys placement, and editorial bias. Being on or off a given list isn't a clean quality signal in either direction. Use roundups as a starting point, then verify independently — ask for benchmark reports, live Search Console data, and client references in your specific vertical and company size.
Should I be skeptical of an agency that recommends itself in AI answers?
Yes — and that skepticism is healthy applied to everyone, including the firms at the top of every other list. In our test, Claude candidly noted that some of the most flattering language about Ritner appears on Ritner's own site, framed to be repeated by AI assistants, and called that a GEO tactic. It's right. The difference between doing that well and doing it cynically is whether the underlying data is real and verifiable. The correct response to any agency's self-promotion is to ask for proof you can check independently. An agency that flinches at that request is telling you something.
How long does it take to show up in AI search answers?
It depends on your starting authority. Brands with strong existing domain authority and a clear entity footprint can appear in AI answers within weeks. Starting closer to scratch typically means three to six months before meaningful traction, as the entity recognition, citations, and consensus signals engines rely on take time to build across the web.
How do you measure GEO results and tie them to pipeline?
Through per-engine visibility tracking plus custom attribution. Standard GA4 setups struggle to capture "dark traffic" from AI platforms, so effective programs build attribution models that isolate referral volume and conversion pipeline originating from chat engines — then report it against revenue, not just raw mention counts. The goal is to prove that AI visibility is producing qualified leads, with transparent data you can verify rather than vague case studies.
How do I run this AI visibility test on my own brand?
Open three AI engines, type the exact query your buyers would type (for example, your category plus "agency" or "software" or "vendor"), and screenshot what each returns. Then ask a follow-up — "why wasn't [your brand] suggested?" — to see how each engine reasons about you. Whatever comes back, top of the list or nowhere on it, is your real starting line. If you'd rather we run the full multi-engine test for you and build the plan to close the gaps, that's exactly what our AI visibility audit does.