@tjrobertson52

Several people from Google have been going on podcast in the last week telling people that Geo is just SEO. If you don't know Geo stands for Generative Engine Optimization. It's the things you do to get recommended by large language models like ChatGPT or Google's AI mode.

I'm just gonna set the record straight, explain why GEO is different than SEO why that matters and why you shouldn't take SEO advice from Google. Now it's true that everything that matters for SEO also matters for GEO which is why I would call GEO s E o plus. However, large language models have introduced some very new tactics that right now are very effective and the primary new tactic is what I'm calling recommendation building.

One of the pillars of SEO has always been link building. If you want to rank in Google for any kind of competitive term, you need to build links to your website. These links can also help you get recommended by large language models but what's arguably more important is your brand being recommended by the pages that these large language models cite before giving a response.

Now if you've been following Geo advice at all, you've probably heard me and other people saying that you need to get your brand mentioned on as many websites as possible. These large language models are looking all over the internet and the Noisiest brands are gonna get noticed the most and will be most likely recommended in their output. And while that generally is true, my view has evolved some as we've been doing this for clients.

I've been able to sharpen up the strategy a bit. Let's talk about a few surgical approaches that have very high returns for a little bit of effort. First of all, it's important to point out that the strategy is very much industry dependent.

You need to know what kinds of pag

little bit of effort. First of all, it's important to point out that the strategy is very much industry dependent. You need to know what kinds of pages large language models are citing when people ask about the products and services you offer.

To do this, you're going to need a tool. There are about 100 out there. We use peak dot AI.

And when you do this, what you're gonna find is that these large language models are citing the same pages over and over again, which is why a more surgical approach matters. Instead of spraying your brand mentions all over the internet, you can just place recommendations on those pages that large language models are siding over and over again. And for pages where you can't place a recommendation, you can try to replace those pages with your own.

So let's talk about some of the most effective ways to do this. Probably the most effective is listicles. These are gonna be things like best product or service in area or for use case.

And for most industries, there are hundreds of different combinations of products and Services and use cases that you can make listicles for. You want your brand showing up at the top of as many of these listicles as possible. But here's a few ways you can do that.

A lot of these listicles are automatically generated by directories. In this case, you can often pay for top placement or just getting more reviews in that directory will automatically get you into these listicles. Other listicles are going to be on 3rd party sites or on the sites of companies in adjacent industries that aren't direct competitors.

I recommend reaching out to the author of all these listicles and just asking what it would take to be listed at the top of their listicle. Most of the time you probably won't get a response but occasionally th

s and just asking what it would take to be listed at the top of their listicle. Most of the time you probably won't get a response but occasionally they just want like a hundred bucks which is easily worth it for a highly cited listicle. And sometimes there's an opportunity to do a reciprocal mention.

I think this is going to become a very popular GEO tactic once people realise how effective it is. Essentially you recommend me in your listicle and I'll recommend you in my listicle. As long as they're not a direct competitor, it's a perfect trade.

This is very similar to reciprocal linking in SEO. Except there's no risk of a penalty. So this is something we're starting to do more of.

Most brands probably have dozens of high value reciprocal mention opportunities. Another type of page that's commonly cited by large language models are discussion threads, most often Reddit. And you probably noticed more and more brands are just spamming the retweets with mentions of their company.

But don't worry, you don't need to spam Reddit to get the benefit. Cause like I was saying before, it's typically the same discussion threads that are cited over and over again. So you can be very surgical about it.

And most of the time these highly sighted threads are 1, 2, 3 years old, meaning you can go leave a comment recommending your brand and it's unlikely moderators are even gonna see it. And the last highly sighted page type I'll talk about are articles. Now it's unlikely that you're gonna be able to get a recommendation for your brand in these existing articles, but if you look at the topics of the articles that these large language models are citing, often you can create similar content on your own website.

But here's what most people get wrong. Just because the large language model is

are citing, often you can create similar content on your own website. But here's what most people get wrong. Just because the large language model is signing an article on your website doesn't mean they're going to recommend you.

The content they retrieve from the article has to contain a recommendation for your brand. So that's why when we create these articles for clients, we always make sure the introduction of the article at least subtly recommends the brand. The best news about all these tactics is almost no one is doing this well right now.

And I believe there's going to be a huge first movers advantage. If you can get your brand recommended by the large language models now, then when people use large language models to create content, you are more likely to show up in their content. It's never going to be easier than it is right now.