You check ChatGPT and your brand is there. You check Gemini and it is not. Perplexity names two competitors and skips you entirely.
That is normal, and it catches people out because it looks like a measurement error. It is not. These tools work differently enough that being visible in one tells you very little about the others, which is worth understanding before you judge whether your AI SEO work is paying off.
What they all have in common
Start with the shared part, because it is most of it.
Every one of these tools reads short passages rather than whole pages. They all favour specific facts over marketing language. They all give more weight to what other sites say about you than to what you say about yourself. They all struggle with anything that cannot be read as text, whether that is a video, an image or a paywall. And they all prefer claims that several sources agree on.
That is why the core work is shared. Fix your product facts, make your answers extractable, sort out the contradictions between your site and your retailers, and you have improved your position everywhere at once.
The differences sit on top of that foundation, not instead of it.
Where they part ways
The biggest split is how often they actually go and look.
Some search the live web on nearly every question. Others answer from what they already know, and only search when the question clearly needs current information. That single difference explains a lot of the inconsistency people notice.
If a tool is answering from memory, your recent work has not reached it yet. If it is searching live, your changes can show up quickly. Same brand, same effort, two very different results on the same afternoon.
Tip: If a tool names you when asked directly but ignores you in a general recommendation, it likely knows your brand and does not consider you a strong enough answer for that particular question. Those are different problems with different fixes.
The other differences worth knowing
Which index they sit on. Tools built on a major search engine’s index inherit that engine’s view of your site. If you rank poorly there, you start behind. Tools that search independently are not bound by that, which is why a brand can appear in one and not the other despite identical content.
How much weight forums carry. Some lean heavily on community discussion, where real people answer real questions with no obvious incentive. For brands with genuine word of mouth, that is an advantage. For brands with none, it is a gap that content on your own site cannot close.
How visibly they cite. Some show their sources clearly. Others give an answer with little indication of where it came from. That matters practically, because the ones that cite openly are the easiest place to see whether your work is landing.
How much they hedge. Some will state a recommendation confidently. Others qualify everything. The same underlying information can produce a firm answer in one tool and a vague one in another.
Why a single visibility score is close to meaningless
This is the part that matters if you are buying AI SEO from anyone.
You will see tools and agencies offering one number for your AI visibility. A score out of a hundred, a percentage, a single figure that supposedly captures how visible you are.
It cannot. Visibility differs by tool, by question, and by run. The same question asked twice an hour apart can return different brands. A single blended score hides all of that behind a number that feels precise and is not.
| A single score hides | Why it matters |
|---|---|
| Which tool you appear in | You may be strong in one and absent from another |
| Which questions you appear for | Visible for a niche question, invisible for the commercial one |
| Whether it was live or from memory | One is current, the other reflects an older picture |
| How much it varies between runs | A single check can be a fluke in either direction |
What is useful instead is tracking by tool and by topic. Which questions matter commercially, which tools your buyers actually use, and whether you appear for those specific combinations. That is more work than a single number, and it is the only version that tells you anything.
Tip: Movement over several months is worth more than any single measurement. These answers vary enough that one check tells you almost nothing on its own.
What this means for the work
Three practical consequences.
Do the shared work first. It improves your position across every tool at once, and it is the majority of what matters. Chasing platform-specific tactics before the foundations are right is effort spent in the wrong order.
Expect uneven results. Appearing in one tool and not another is not a sign the work failed. It usually means the tools weigh things differently, or one has simply not caught up yet.
Judge it over time, by tool. A month of data on one platform proves very little. A trend across several months, tracked separately, starts to mean something.
Where the difficulty is
None of this is complicated to understand. What is harder is deciding what to actually track, which questions are worth watching, and how to tell a genuine improvement from normal variation. That takes a fixed set of questions, checked consistently, over enough time to see past the noise.
That is part of the work I do. If you want a clear read on where your brand stands across these tools, my AI SEO page explains how I approach it, or request an SEO audit covering organic and AI visibility together.
For how these tools decide who to name in the first place, see how AI search decides which brands to mention. For what the work can and cannot promise, see what AI SEO can do.
Frequently asked questions
Why does my brand appear in one AI tool but not another?
Because they work differently. Some search the live web on nearly every question while others answer from what they already know. They also sit on different indexes and weigh sources like forums differently. Being visible in one says little about the others.
Should I optimise separately for each AI tool?
Not at first. Most of what these tools reward is shared, so the foundational work improves your position everywhere at once. Platform-specific effort only makes sense once those foundations are right.
Is there a single score for AI visibility?
Not a meaningful one. Visibility varies by tool, by question and between runs of the same question. A single blended number hides all of that. Tracking by tool and by topic is more work and far more useful.
How often should I check my AI visibility?
Monthly is enough for most brands, using a fixed set of questions so results are comparable. Checking more often mostly captures normal variation rather than real change.
Why do results change between two identical questions?
These systems are probabilistic, and the sources they retrieve can differ between runs. That is why movement over several months matters more than any single check.
Aashish Maharjan is an SEO specialist focused on ecommerce and international organic growth across global markets. Learn more about my work or request an SEO audit.