Alex Yarosh Get Free Snapshot

Source record

@tjrobertson52 TikTok profile avatar

@tjrobertson52

2025-12-07

Every SEO agency claims they'll rank you in ChatGPT. Here's what they're actually doing (spoiler: it won't work) ๐Ÿ‘€ #SEO #AISearch #Marketing

Share source record
Tiktokexcerpt onlyen3

Source Text

i

`n`nBut when you look at their plan for doing this, you'll see that almost none of these agencies are actually doing anything that will help you get recommended by large language models. So I wanna talk about what these agencies are calling GEO or AI O. And what you should actually be looking for if you're looking for an agency to help you rank in large language models.

Yesterday, I had someone reach out after watching my YouTube videos and he was kind enough to share about half a dozen proposals he got from other SEO agencies. And I love looking at these proposals. It gives you real insight into how these agencies think.

Sure enough, they all talked about ChatGPT and Google's AI mode and how important it is to be found in large language models. And that's all true. `n`nBut then when you look at the actual services they're proposing to help this business rank in large language models, it's immediately clear that they have no idea how to influence the recommendations made by large language models.

`n`nThese are things like adding FAQs and writing in an NLP friendly format and sure you should be doing that but it's not gonna make a big difference it also recommend things like adding structure data or L 0 m s dot t x t and so far I haven't seen any evidence that those things make an impact in AI visibility truth is those of us that are actually running the experiments to see what working are figuring all this out in real time.

`n`nThe GEO tactics I saw in these proposals are clearly just these agencies parading what people were saying on LinkedIn six months ago I guarantee you if you ask them to show you real data on how these tactics improve visibility and large language models they wouldn't be able to provide it cause it doesn't exist changing the structure of the content on your website is gonna give you 5 or 10% more visibility at most.

So here are a few things an agency should be doing for you to increase your visibility in large language models. Most importantly, they should be tracking what sources the large language models are looking at before recommending you. The software we use for this is called peak dot AI, but there are now many platforms doing this and they're all very similar.

They should be figuring out which directories are most important in your industry. Typically there are only one to three that really matter. Make sure you're in there and then figure out how to climb the rankings so you show up on the automatically generated listicles.

Large language models love these listicles. They should be identifying the most commonly cited discussion threads. These are typically going to be in Reddit or Facebook groups.

You don't need to be posting in these forms everyday. Typically the most cited threads are from one or 2 years ago. You just need to find them and leave some comments talking about your brand.

But most importantly, you need to be creating content, a lot of content published to a lot of different platforms, the platforms that large language models are most often citing. And your content strategy should be informed by the kinds of topics that large language models tend to cite when people are asking about your products or services. And you might also consider setting out press releases, those are the things that actually help you show up in large language models.

Right now, the truth is most SEO agencies aren't ready to completely change their entire model. It's much easier for them to pretend like a few tweaks is all it takes to get you showing up in these large language models. This is actually good news because your competitors are probably using these agencies and that's why there's so much opportunity right now.

Very few agencies are actually doing the work to figure out what works. But if you're serious and want to get a first movers advantage in large language models, you need to be willing to do some experimentation. At least once a month, we run some experiment and get no results from it.

But this is how we find the things that actually do work. And then we just keep doing more of that and less of the things that don't work.

Source Intelligence

i

The creator says many GEO proposals overemphasize FAQs, NLP-friendly formatting, structured data, or llms.txt.

3 related signals ยท GEO agency evaluation / LLM source tracking / AI visibility experiments

  • Treat website-structure-only GEO proposals as insufficient without source tracking and experiments.
  • Build GEO audits around prompt citation tracking, directory presence, and discussion-thread opportunities.
  • Maintain a monthly GEO experiment log with hypothesis, result, and continue/stop decision.

Questions this source answers

i

What is this source mainly about?

The creator says many GEO proposals overemphasize FAQs, NLP-friendly formatting, structured data, or llms.txt.

What should an operator take from it?

Treat website-structure-only GEO proposals as insufficient without source tracking and experiments.

Which topics does it connect to?

This source is connected to GEO agency evaluation, LLM source tracking, AI visibility experiments.

What public evidence supports the record?

Every SEO agency is now claiming they'll help you rank in large language models.`n`nBut when you look at their plan for doing this, you'll see that almost none of these agencies are actually doing anything that will help you get recommended by large language models. So I wanna talk about what these agencies are calling GEO or AI O...