Chelena Peart.
From found to recommended

How to get recommended by AI when it already answers your clients' questions

Ask ChatGPT how to fix a problem and it will happily give you an answer. It might even use your content to do it. Then the person goes off and tries it on their own, and your name never comes up. This guide shows you how to publish so your content gives people a reason to come and find you.

By Chelena Peart Updated 28 Oct 2026 10 min read Includes 5 prompts + content check
The short answer

To get recommended by AI, publish content that answers your clients' questions and then says where that answer stops working. Show what your experience lets you spot and add proof from real clients. Finish with a clear next step, so the language about when to bring in an expert sits right next to your name. Back it up with mentions on other people's platforms and test regularly. Nothing guarantees a recommendation, but these are the signals that give AI a reason to point people to you.

Watch: Why AI Answers Your Clients' Questions But Doesn't Recommend YouWatch the video this guide is based onWhy AI Answers Your Clients' Questions But Doesn't Recommend You

Used, but not recommended

Think about how people actually use these tools. Some of them ask "who's a good LinkedIn strategist?" A lot more of them ask "how do I get more clients from LinkedIn?"

For that second question, AI doesn't need to recommend anybody. It just answers. And if I've published a really clear answer to that exact question, AI can take my tips, tidy them up and hand them over with my name nowhere in sight.

I only spotted this while talking it through with ChatGPT after my video on getting recommended. I'd been telling people to answer their buyers' questions in public, and I still stand by that. The advice just needed a second half.

The bit that worries me most is what happens to the person asking. They get a tidy list of tips and go away thinking they've got the whole solution, when there's a lot more to it.

Being found

AI can use your expertise

People and search tools can find what you know and repeat it when someone asks a question.

Being recommended

AI has a reason to point to you

There's clear evidence that you solve this problem, and a clear moment when someone should bring you in.

You need both. My guide on getting recommended by ChatGPT covers being found. This one is about the step after.

Make the limits part of your content

The answer ChatGPT and I landed on was to make the limits part of what you publish. Alongside the helpful answer, you also say where that advice stops working and what can go wrong if someone stops there.

Here's an example from my own world. I could write a guide on how to improve your LinkedIn profile, and AI could happily summarise it. Or I could also explain why a profile that looks great can still bring in no enquiries at all, because the problem is somewhere else, often in the positioning or the offer. That second part is where my experience shows, and it's very hard for anyone to get it from a list of tips.

AI can still borrow the useful bits. But now the language about when to bring in an expert is sitting right next to your name.

What to do

Run everything you publish through the five-part check below. I use it on videos, posts and guides, and it takes a few minutes once you're used to it. To show you how it works, I'll use my video on the hidden LinkedIn photo setting as the example all the way through.

01

Answer the question simply

Start with the problems your clients are trying to solve and the words they use when they describe them. People type their problem into AI. They very rarely type your job title or the name of your service.

Then give a clear, helpful answer. Holding back the useful part to make people come to you tends to backfire, because AI will simply use someone else's answer instead.

In my example

In the video, I showed exactly where the photo visibility setting is and how to switch your photo on so everyone can see it.

02

Say where that advice stops working

This is the part most of us leave out. Explain when the standard advice works and the situations where it won't be enough. Name the mistakes you see when people try it on their own, and what it costs them when they get it wrong.

In my example

Fixing your photo means people can see your face. It won't make your profile bring in enquiries on its own.

03

Show what your expertise changes

AI can give someone an answer. Your experience is what tells them whether that answer fits their situation. So share how you work out what's really going on before you recommend anything, and the patterns you've noticed from doing this every day. The things you spot that someone without your experience would walk straight past are the most useful of all.

In my example

When I spotted a client's blue avatar, I knew what it was costing her because I look at profiles all day. Spotting things like that is the bit people can't easily do for themselves.

04

Add proof

Case studies and testimonials that describe a real outcome count here. So does explaining the decisions behind a result, because that shows how you think. Make your experience and credentials easy to check, on your website and your LinkedIn profile.

Always ask your client before you share their name, words or results.

In my example

The client with the blue avatar became a one-to-one client. A real result, from a real person.

05

Make the next step obvious

This is the one I see skipped most. We put so much into the helpful answer that we forget to tell people what to do if they want more help.

It doesn't need to be pushy. A line like "if this sounds like you, here's where to start" at the bottom of a guide or the end of a video is enough. Link the content to the offer or resource that fits it, and make it easy to enquire or book.

In my example

For anyone whose profile looks fine but still isn't bringing in enquiries, the next step is my LinkedIn Action Plan, a one-to-one session where we find what's getting in the way.

Before anything goes out

"If AI can give someone everything I've published here, what have I shown that would make them want to work with me?"

If you can't answer that, the piece isn't finished.

Check your own content

Pick something you've published recently, or something you're about to post, and score it here. Yes is 2, partly is 1 and no is 0.

1. Does it answer the question simply?In the words your clients would use

2. Does it say where the advice stops working?Or what can go wrong if someone stops here

3. Does it show what your expertise changes?Something a reader couldn't easily work out alone

4. Is there proof?A client result, case study or real example

5. Is the next step obvious?Someone who wants more help knows exactly where to go

Your score

Answer the five questions

Your score and the first thing to fix will show here.

Your score is saved in this browser only. Nobody else can see it.

Back yourself up beyond your own platforms

Your own content can only say so much about you. When other people mention you, it becomes easier for both AI and buyers to check that you are who you say you are.

  • Guest on podcasts and take part in expert discussions in your field.
  • Contribute to industry publications your clients read.
  • Build credible partnerships and collaborations with people who serve the same clients.
  • Keep your name, title and description consistent on LinkedIn, your website and any bios, so it's clear every mention is the same person.

Each of these should give people a clear way back to your website or LinkedIn profile, for anyone who wants to go deeper.

Keep testing

Every so often, ask AI the questions your clients would ask. Ask both kinds, "how do I fix this?" and "who can help me with this?", and see whether your name comes up and what it says about you.

There's no setting that guarantees you'll be recommended, and the answers change from day to day. So look for patterns over a few months. Each piece you publish with proof and a clear next step gives these tools more reason to point people your way.

If you'd like a way to track this, my guide on getting recommended by ChatGPT has a monthly tracker you can use each time you test.

Watch: I Asked AI to Recommend an Expert. It Didn't Recommend Me.Watch me run the recommendation testI Asked AI to Recommend an Expert. It Didn't Recommend Me.

The prompts

Copy these into ChatGPT or whichever AI tool you use. Replace anything in square brackets with your own details.

1. Check a piece of content

The five-part check
I'm a [your role] who helps [who you help] with [the problem you solve]. Below is a piece of content I'm about to publish.

Check it against these five points. For each one, tell me whether it's there, partly there or missing, and quote the line that does the job if there is one.
1. Does it answer the question simply, in the words my clients would use?
2. Does it say where this advice stops working, or what can go wrong if someone stops here?
3. Does it show something my experience changes that a reader couldn't easily work out alone?
4. Is there proof, such as a client result or a real example?
5. Is there a clear next step for someone who wants more help?

Then answer this honestly: if AI gave someone everything in this piece, what have I shown that would make them want to work with me?

Don't rewrite the piece. Just show me the gaps.

[Paste your content]

Use it before you post, or on your most popular older pieces.

2. Write where the advice stops

Point 2
Here's a piece of advice I give my audience: [paste the tip or section].

Help me write a short section to go after it about where this advice stops working. Before you write anything, ask me questions one at a time about:
- the situations where this tip won't be enough
- the mistakes I see when people try it on their own
- what it costs them when it goes wrong
- what I notice in those situations that they usually miss

Use my answers and my wording. Keep it under 150 words, in UK English, and don't add claims or results I haven't given you.

The questions matter more than the draft. Your answers are the part AI can't make up.

3. Write the next step

Point 5
Write a short "if this sounds like you" block for the end of my [guide / video / post] about [topic].

The person who needs more help is someone who [describe the situation where your help makes the difference].
The next step I want to point them to is [your offer or free resource], and they can get it by [link or how to book].

Keep it under 60 words, warm and low pressure, in UK English. No hype, and no urgency or results I haven't given you.

Write one for each offer you have, then reuse them.

4. The "how do I" test

Keep testing
[Type the question your client would ask, in their words. For example: How do I get more clients from LinkedIn as a consultant?]

Then, in the same chat:
Who could help me with this? Recommend some specialists or experts, and tell me why you chose each one.

Use a temporary chat, or one that isn't signed in, so your own history doesn't skew the answer. Note whether your name or your ideas come up.

5. What's missing?

Keep testing
Here's my LinkedIn profile and website: [your links].

Based on what you can find about me, why might you not recommend me to someone who needs help with [the problem you solve]? Tell me which proof or positioning signals seem to be missing, and be specific.

Treat the answer as a to-do list for your next few pieces of content.

Questions

Can I pay or optimise my way into ChatGPT's recommendations?

No. There's no setting or payment that guarantees AI will recommend you, and the answers change from day to day. What you can do is give these tools clear evidence of who you help and why you're worth recommending, then test regularly to see what's changed.

If I answer my clients' questions for free, am I giving my expertise away?

You're giving away the answer, and that's what earns trust. Your expertise shows in the parts a list of tips can't cover, like where the advice stops working and what you'd spot in someone's situation. Put those in, along with a clear next step, and the free content starts bringing people to you.

How long does it take before AI starts recommending me?

There's no fixed timeline. It depends on how much evidence about you already exists and how consistently you add to it. I'd test every month or two and look for patterns over a few months. One answer on its own doesn't tell you much.

What if I don't have case studies yet?

Start with what you do have. A short story about a client's situation and what changed, shared with their permission, counts as proof. So does explaining the decisions you made on a piece of work. Ask happy clients for a testimonial that describes the outcome, and build from there.

Do I need to go back and change everything I've already published?

No. Start with new content, then pick your most viewed older pieces and add the missing parts to those. Often a few lines on where the advice stops working and a next step at the end is all it needs.

What to do next

If you haven't tested whether AI knows who you are yet, start with this video, where I test my own name live in ChatGPT and walk you through the first steps I took.

Watch: How to Get Recommended by ChatGPT (I Tested My Own Name Live)Where this startedHow to Get Recommended by ChatGPT (I Tested My Own Name Live)
Chelena Peart

Chelena Peart

Visibility & Authority Strategist · Founder, The LinkedIn Strategy Lab

I help established experts, including consultants, coaches, strategists and founders, get found, trusted and chosen by the right buyers. Before building my business I spent around 18 years in corporate IT and project management. I'm based in the UK, and I test what I teach on my own business in public.

© 2026 Chelena Peart · The Vault Virtual Agency · Some links are affiliate links. Back to top ↑