One day very soon, the clients who get referred to you aren’t going to bother checking your website. They’re going to ask their favorite AI about you. And if you’re not sending out any online signals, you’re basically invisible.
But we’re not working with a black box here. You actually have real influence over what AI has to say. But some signals are more powerful than others so here’s what you need to do to make sure you’re not wasting your time.
Your SEO Playbook Won’t Save You Here
This is different than getting Google to notice you. The SEO game was about key words and whole page relevance.
Generative engine optimization, or GEO, works differently. It doesn’t reward pages. It rewards ideas – chunks of information a model can select and summarize when it builds an answer.
So the EEAT – experience, expertise, authority and trust – content still matters. But the packaging needs to be different.
Try this: Type your own name into ChatGPT or Claude (incognito mode) and ask, “What do you know about [YOUR NAME] Would you recommend working with her to [WHAT YOU DO]?”
When I ask it about Laura Zavelson for helping solopreneurs integrate AI, it recommends me.
That’s not luck. Every piece of content I’ve published for the last year has been built to tell that story.
About six months ago, I did a deep dive on what the AI engines actually look for: what the platforms say publicly, plus the academic research on generative engine optimization.
The most surprising thing I learned (but maybe I shouldn’t have been surprised) was that even the people who build these things don’t know 100% how they decide what they’re going to say.
But the evidence converges on five signals:
- name recognition
- brand alignment
- expertise focus
- external signals
- content recency
Signal 1: Name Recognition
When your referral asks, “Should I work with Laura Zavelson to do XYZ?” you want you to come back. Not someone who shares your name. And not “I have no idea who she is.”
So if you’re starting from ground zero, here’s what I would do:
- connect all your properties: website links to socials (insta, linkedin, etc), socials connect back to website or landing page at your domain
- add your name as the author on your blog posts. I didn’t do this for years — it’s my website, of course I wrote it. But AI doesn’t know that
- if you have a common name, put your title in your meta descriptions and bios: “John Doe — GTM consultant.” Give the model a way to tell you apart
Signal 2: Brand Alignment
This one caught me when I tested the Online Signal Report on myself. It actually shows up as a “fix” for most people who run it.
Our businesses are organic. So your website says one thing, your LinkedIn says another, and your podcast guest bio is different again.
To the prospect checking you out – and to the AI synthesizing all of it – you look like Frankenstein. Stitched together from parts that don’t match.
The item on your todo list is to find a consistent way to talk about the problem you solve, who you want as a client, and the results you deliver. And be as consistent as possible across every place that you show up.
Signal 3: Expertise Focus
I know you have a wide range of skills and interests. It make you interesting to people. But it doesn’t make you interesting to AI. The way AI synthesizes information, it doesn’t read “expert” if you post about wine one day, logistics the next, and cats on Friday. Your entire body of work has to connect around a fairly narrow niche.
So pick about three topics and post on them over and over. Boring? Only for you. Your audience doesn’t consume your content the way you create it. You keep it interesting by changing the angle, not the subject.
Think of a photographer shooting the Eiffel Tower: wide landscape, close-up on a single bolt, from the ground looking up, from the top looking down.
Here’s a test you can run in ten minutes: pull your last 20 LinkedIn posts or newsletters. Could a stranger (or an AI bot) name your expertise in one sentence from that content alone? If not, you need to narrow it down.
Signal 4: External Signals
This is the only signal you don’t fully control. It’s also the one with the most impact on your recognition and your revenue.
AI trusts what others say about you more than what you say about yourself. Sound familiar? It’s the same reason referrals work.
Third party mentions might include podcast guest appearances or best of list mentions or backlinks.
For me, the best return for the effort has been podcast guesting. Here’s the secret — they don’t have to be big and you only have to do a few. Any podcast creates an online artifact that AI can find and use as a marker of your expertise.
Signal 5: Content Recency
A stale website reads as a question mark to AI, to prospects, and to your own referral network. I’ve checked out someone’s website before making a referral, seen a blog that died two years ago, and wondered: is she still running her business? Did she go back to corporate?
What does “recent” mean to an AI? It’s not a simple count of days. It’s whether the content was written recently enough to reflect current information. Newton’s laws don’t need a refresh. But if your industry is even tech-adjacent, it’s moving, and your content needs to move with it.
Your Next Move
You can try to scope out how AI might interpret what you’ve got out there. Or skip the guesswork: my free Online Signal Report takes about five minutes, reads your links, and gives you a 360-degree view of how you look to client prospects and AI engines across all five signals.
The bottom line: getting AI and your referral network to introduce the right people to you means creating content. Not for followers. But to showcase your expertise, support your fees, and build the trust that people need to sign your engagement letter.
