What is AI saying about your app?
People ask ChatGPT, Gemini and Perplexity which app to download. AppHeard tracks whether they say yours — and shows you the web pages that decide the answer.
Free check · no account needed · See pricing
You optimised for the App Store. Your buyers are asking an assistant instead.
Nobody scrolls a category chart to pick an app any more. They ask, they get three or four names back, and they download one of them. If yours isn't in that answer you never enter the comparison — and no ranking report will tell you, because none of them read the answers.
How it works
Three engines, every day, and a specific page to go change.
We ask what your buyers ask
Around 30 real recommendation questions for your category — the ones people actually type — asked every day across three AI engines with web search on.
We read every answer
Every app named, its position in the list, and every source the AI cited. Sampled repeatedly, because one answer is noise and a rate is a signal.
Influences 2 questions where AI names someone else.
We tell you what to fix
The listicles, threads and gaps driving the answers you lose, with the outreach already drafted. You send it. Nothing posts on your behalf.
Every figure on these cards is from this week's real runs, not a mock-up.
One score, and everything underneath it
A number on its own can't be acted on. Every tile here is a real view of real answers — the questions, the apps named beside you, the pages the engines cited.
How often AI names you when someone asks for an app like yours.
Your share of the answer, against the apps you actually lose installs to.
Which questions get you named, how often, and which ones never do.
The domains the engines cited to build it, ranked by how much they steer it.
The same question on all three, because they disagree more than you'd think.
Built for shipping, not for staring at
Five things we decided on purpose.
- Every number is a measurement, not an estimate — the sample count sits right next to it
- ChatGPT, Gemini and Perplexity, web search on, re-asked every single day
- Tasks name the exact page or thread to change, with the outreach already drafted
- You send it yourself: nothing posts on your behalf, and we never touch the Reddit API
- Cancel any time from Settings, and your history stays readable for 90 days after
Measured across tracked categories. Updated daily.
One plan
Cancel anytime. Run the free check first — that's the trial.
- 1 app tracked
- 5 prompts, re-asked every day
- Answers from 3 AI engines: ChatGPT, Gemini, Perplexity
- 5 competitors + category leaderboard
- Instant change alerts
- Improvement tasks with drafted outreach
- Weekly digest of what AI said
Questions
Is this the same as what I see in ChatGPT?
Close, but not identical. We measure through each platform's official API with web search enabled. Consumer apps add personalization and memory, so treat the score as directional intelligence and the trend as the signal.
How often do you check?
Every day. Your highest-demand intents get a daily pulse, and every tracked intent completes full three-engine coverage each week, spread across the days so your dashboard freshens continuously.
How long until I see the score move?
Perplexity typically moves in 4–8 weeks; ChatGPT in 2–4 months.
Do you post to Reddit for me?
Never. We show you which threads the AI cites and draft a helpful reply, then you post it yourself. We don't touch the Reddit API and nothing is automated on your behalf.
Do you support Android?
Not yet. AppHeard is iOS and App Store only.
Is there a free trial?
The free check is the trial. It asks the engines your category's real recommendation questions and shows you exactly where you stand, without an account. Tracking is $19.99/month from the first day, and you can cancel any time from Settings.
Find out what AI said about your app
The check is free, takes about a minute, and needs no account.
Free check · no account needed