Why ChatGPT doesn’t recommend your product
The usual reason ChatGPT never names you is not that your product is bad. It is that you are missing from the candid sources it reads, mostly Reddit threads where people compare tools, or the verdict there quietly favours someone else. ChatGPT can only repeat what it has read, and right now it has not read enough good things about you.
The short version
- ChatGPT can only repeat what it has read. If it never names you, it has not read enough about you.
- Reason one: you are simply absent from the candid sources it leans on, which for buying questions is mostly Reddit.
- Reason two: you are there, but the consensus favours a rival that people rate more highly.
- It is rarely your website or SEO. Recommendation answers run on what others say about you, not your homepage.
- The fix is slow and honest: be genuinely present and useful in the right threads. subding flags them live; a human writes the reply. We cannot force a mention, and neither can anyone else.
The real reason you are not named
It stings to ask an assistant for the best option in your category and watch it list everyone but you. The instinct is to assume your product fell short. Usually it did not. ChatGPT is not judging your product on its merits; it is repeating what it has read about it, and the answer says more about your visibility than your quality.
Once you accept that, the diagnosis gets simpler. There are really only two things going wrong, and often both at once. Either the model has barely read anything about you, or what it has read points somewhere else. Here is how the symptoms map to their likely causes.
| What you see | The likely cause |
|---|---|
| ChatGPT names three rivals, never you. | Those rivals are discussed in the threads it read; you are not. |
| It describes your category but no brand. | No clear consensus exists yet for it to lean on. |
| It mentions you, then warns readers off. | The candid verdict about you is mixed, and it read that too. |
| It recommends an older, worse option. | That option has years of accumulated goodwill in the sources. |
| It gets your product plain wrong. | It is stitching together stale or thin mentions of you. |
Notice what is not on this list: "your landing page needs more keywords". That is almost never the problem.
Reason one: you are simply absent
The most common case is quiet absence. When people compare tools in your space, they do it in candid places with nothing to sell, and on the open internet that mostly means Reddit. If your name rarely comes up in those threads, the model has almost nothing to pull from, so it reaches for the brands that are discussed instead.
You cannot be recommended out of a conversation you were never part of.
This is oddly good news. Absence is fixable in a way that a genuinely weak product is not. The work is to become part of those conversations, honestly and as yourself, so there is something real for the model to read next time.
Reason two: the consensus favours a rival
The harder case is when you are present, but the mood is not on your side. A rival has years of goodwill banked in the threads, or a recent wave of people quietly steering others away from you. The model reads that too, weighs it, and lands on someone else. It is not malice; it is arithmetic on the sentiment it found.
- Read the actual verdict. Find the threads where your category gets debated and see, plainly, what people say about you.
- Take the criticism seriously. If a complaint keeps recurring, that is product feedback the model is already repeating.
- Add real counterweight. Show up, help, and let satisfied users speak, so the balance of what gets read shifts over time.
- Never astroturf. Fake praise gets caught, and a burned reputation is far harder to fix than a quiet one.
To understand how these models turn a pile of opinions into a single recommendation, our guide to ChatGPT SEO is the companion piece to this one.
What to do about it, honestly
The fix follows straight from the diagnosis: become someone worth reading about, in the places these models read, before the recommendation threads close. Those threads gather their answers fast and then get indexed with whatever got said, so timing is half the battle. Miss the window and the names are already set without yours.
That makes hearing about the right thread early the practical lever. A Reddit monitoring tool watches posts and comments across Reddit for your keywords and pings you while the conversation is still open.
"ChatGPT keeps pushing the same two brands at me and I don't trust it, what's the underrated option you'd back instead? genuinely curious."
→ Join it while it's live
subding points you at the thread; a human writes the comment, discloses who they are, and adds something useful. We do not post, upvote or message for you, and we cannot promise the model will ever name you. There is more on how Reddit shapes these answers in our Reddit AI visibility guide, and the wider picture in AI visibility.
What will not work
Before you spend money on the wrong thing, here is the short list of moves that reliably disappoint. They either do nothing or actively set you back, and the tempting ones are usually the worst.
- Stuffing your site with keywords. The recommendation is not built from your pages, so this changes nothing.
- Buying fake mentions. Communities and moderators catch it, and a caught brand loses the trust it was chasing.
- Anyone guaranteeing a citation. Nobody controls the output, so a promised mention is a promise they cannot keep.
- Doing it once and stopping. Reputation decays; a single good week fades from what the model reads.
Questions, answered straight
Why doesn’t ChatGPT recommend my product?
The usual reason ChatGPT never names you is not that your product is bad. It is that you are missing from the candid sources it reads, mostly Reddit threads where people compare tools, or the verdict there quietly favours someone else. ChatGPT can only repeat what it has read, and right now it has not read enough good things about you.
Is it because my SEO or website is weak?
Rarely the main cause. A clean site helps a model understand you, but recommendation answers lean far more on what other people say about you elsewhere. You can rank well and still go unmentioned, because ChatGPT is repeating a consensus formed in places you do not own, not reading your homepage aloud.
Does this apply to Claude and Perplexity too?
Yes. Claude, Perplexity and Google’s AI draw on similar candid sources when they answer recommendation questions, so if you are absent from those places you tend to be absent everywhere at once. The fix is the same across all of them: be genuinely present and well-regarded where people compare options.
How do I find out what these tools say about me?
Ask them directly. Pose the recommendation questions your buyers actually ask, across ChatGPT, Claude and Perplexity, and note who gets named and how you are described. Then watch the Reddit threads those answers are built from, so you can see the consensus forming rather than only its output.
Can I just pay to get mentioned?
No, and trying tends to backfire. Manufactured praise gets spotted by communities and moderators, and the sources feeding these models are read by humans too. subding will not post for you and cannot force a mention; treat any guaranteed-citation pitch as a red flag. The only durable route is being genuinely worth recommending.
How long does it take to change what ChatGPT says?
Longer than you would like, because reputation is slow to build and slow to surface in an answer. There is no switch to flip. What you can do is start being present and useful in the right threads now, improve your odds steadily, and judge it over months. Better odds, never guarantees.
Turn absence into presence
ChatGPT skips you because it has not read enough good things about you yet. Fix that where it reads, honestly and early, and the odds start to move. subding handles the part that is impossible to do by hand: hearing the right thread while it is still live.
See what every plan includes on the pricing page.