How to get recommended by ChatGPT
There is no form to submit and no way to buy a place in ChatGPT's organic answers. What you can do is make it easy for any AI system to understand what you are, who you are for and why you belong on a shortlist, and make sure the sources it reads say so. Then measure, because the answers are the only proof.
Start by measuring where you are
Before changing anything, record a baseline: which buyer questions name you, on which engines, and which competitors appear instead. Without a baseline you cannot tell whether a change helped. A public Shortlist leaderboard, or the free check, gives you that starting point.
Say clearly what you are
Assistants work from text. If your homepage says "the future of customer engagement" but never "WhatsApp API provider for online stores", you are harder to place in a category. State your category, your customer and your key facts in plain words on pages that are easy to crawl.
- One sentence near the top of your homepage: what it is, for whom.
- A pricing page with real numbers, if you can publish them.
- Comparison and "alternative to" pages that are factual, not attack ads.
- Consistent product and company names everywhere.
Be accurate where assistants look
Read the sources the engines cite for your category's questions; Shortlist stores them with every answer. They are often review sites, comparison articles, directories and community discussions. Where you can legitimately be listed or corrected, do it. Where a source describes you wrongly, fix the facts at the source if you can.
This is slower than editing your own site, but it often matters more: assistants tend to name brands that several independent sources agree on.
What does not work
Stuffing pages with "AI" keywords, publishing large numbers of thin pages, or writing fake reviews will not build the consistent, credible picture assistants rely on, and the last one can do real damage. Be wary of anyone who promises placement in AI answers; nobody controls them.
Measure again, on all engines
After changes, re-run the same questions. Look for movement in mention rate beyond the error bars, and check every engine: improvements often show first on engines that use live search, such as Perplexity and Google's AI answers, and later on others. Shortlist re-runs every public category weekly, so the trend is visible over time.
A realistic timeline
Week one: measure the baseline and read the sources cited for your category. Weeks two to four: fix your own pages, starting with the category statement on the homepage, pricing and comparison pages. In parallel, correct inaccurate third-party listings and pursue fair inclusion on the comparison pages the engines rely on. After a month: re-measure the same questions on every engine and compare with the baseline, interval against interval.
Some changes show quickly on engines that search live; others take longer. Keep measuring on the same schedule rather than judging after one run.
See it for a real category
- AI recommendations for crm for small business
- AI recommendations for whatsapp business api provider
- AI recommendations for email warm-up tool
- How Shortlist measures
Questions
Can I pay OpenAI to be recommended?
Not in the organic answer. Visibility depends on how your category is described in the sources and data the model uses.
How long does it take to see a change?
Engines that search the web live can reflect changes within weeks. What models learned in training changes with new model versions.
Does structured data help?
It makes your facts unambiguous for systems that read your pages. It is useful hygiene, not a guarantee.
How do I know if it worked?
Re-run the same buyer questions and compare mention rates with their intervals. Shortlist does this weekly for every public category.