#2: Want to show up in ChatGPT answers? Start here

👋
Everyone’s chasing citations in AI.
We want to be sourced, footnoted & quoted.

But let’s be honest,
how much longer do we expect OpenAI to keep ‘showing’ sources?

Here’s my truth:
Being cited is not the win. Being named is.

Ask ChatGPT, “Where should I eat in Paris?”
✅ If your restaurant is in the answer, you win.
❌ If your blog post is in the footnote, no one clicks it.

So how do you get named?
This is where I’m supposed to say “it depends on a trillion variables”.
But let’s keep it practical, shall we?

Start with something AI actually sees: reviews.

Reviews don’t just help people, they’re a source for AI

ChatGPT, Perplexity, and Google’s AI Overviews are pulling language, sentiment, and structure straight from public reviews.

“AI is turbocharging search by pulling the specific language and data in ratings and reviews.” -Reputation.com

This isn’t hypothetical. These systems extract keywords, features, and tone from review content. They don’t just take the score, they grab the story.

So if your reviews say:

“Highly experienced professionals. Worked with Hannah, she solved our legal issue in two days.”

AI connects that to: experience, Hannah, speed, satisfaction.

Now ask:

“Best experienced law firm in Amsterdam?”

You might no longer be anonymous.

For example:

Reviews = structured signals + social proof

Reviews are dual-purpose now:

• To humans: They’re social proof. 42% of people trust them as much as personal recommendations.
• To AI: They’re structured, sentiment-rich data, exactly what large language models thrive on.

If your site has review schema, even better. Structured markup feeds the AI clean data. Think: star ratings, timestamps, feature mentions. Make sure they’re not buried in a paragraph, but machine-readable.

Fake reviews can backfire, hard

It’s tempting to game the system. Please keep it honest.

AI can spot fake reviews and will get better at it. Repetitive language, suspicious timing, single review accounts or keyword stuffing. Platforms like Google respond fast (if they want to, lol).

• Google blocks millions of fake reviews a year.
• Amazon suspends listings flagged by AI as manipulative.
• The FTC now explicitly bans fake testimonials.

If your trust profile drops, AI could de-rank you.

And here’s a twist: ChatGPT doesn’t read Google Maps

ChatGPT doesn’t have access to live Google Maps data: no real-time star ratings, no local pack rankings, no Business Profile reviews.

Unless someone explicitly shares a review on a crawlable page, or your site embeds them with schema, they’re invisible for ChatGPT.

That means:
If your reviews live only on Google, OpenAI probably isn’t seeing them.

So where is ChatGPT looking?

• 🟢 Bing-indexed sites (e.g. Yelp, Trustpilot, G2, Tripadvisor, Kiyoh)
• 🔍 Schema-marked content on your own site
• 📊 Pages that summarize or compare providers
• 💬 Public reviews in plain text, like Reddit, forums, niche directories

Want to rank in ChatGPT? Then make sure your strongest reviews aren’t trapped inside Google’s walled garden.

This is where to start

Want to be recommended, not just indexed?

Here’s the playbook:

1. 📝 Get real reviews — on Google and Bing-visible platforms (Yelp, Trustpilot, Tripadvisor, G2).
2. 🧩 Encourage specifics — features, outcomes, names. The more detail, the more AI understands.
3. ⏳ Keep them fresh — recent reviews carry more weight, both for people and algorithms.
4. 🌐 Make them public — repost selected (or all!) reviews on your site. Add schema and let AI crawl them.
5. 📉 Monitor sentiment — use tools or scroll manually. Spot negative trends early.
6. 🚫 Flag the fake ones — protect your trust layer. AI reads integrity (and we all want to be honest, right? :)

Subscribe

In summary

If you want to show up in AI answers, don’t chase citations.
Chase trust signals, the kind AI can read, rank, and recommend.

Heaps of great reviews do not mean you will get named, but they improve your chances.

That means fresh, specific, authentic reviews, across platforms, tied to real features, and marked up on your site.

Do that, and next time someone asks ChatGPT who to trust, hopefully your name isn’t in the footnote. It’s in the answer.

See you in two weeks,
Luuk

Thoughts? Questions?
Send them to luuk@offbeatmarketing.nl or drop me a DM on LinkedIn.

PS new to Ai found you? Find last weeks issue on Reddit here.

‍