Performance optimization / server-side caching

Let's talk about another important factor for being recommended by large language models like ChatGPT. In Google's AI.

Mode. This one's a little more technical, but I'm gonna try to make it simple for you.

That is improving your server response time. And the metric we really care about here is your TTFB, or time to first byte.

You may have heard about this factor in traditional SEO, but it's much more important in AI. Search.

This is how long it takes from the request made to your web server till the first byte is downloaded from the web page. When someone enters a prompt in a large language model, that large language model then performs a search.

It will only wait so long for your website to load before it gives up. Now, just like traditional search engines, AI.

Search is index based. Meaning when it performs its initial search, it's not searching across the live web.

It's searching its own index of saved information on these web pages. To get that initial set of results, it doesn't need to access your website.

However, AI search goes one step further...

around the world, limiting the distance between your website and the server making the request. The second is enabling server side caching.

This is where a copy of your pages are saved on the server so the page doesn't have to be rendered in real time. This is also a service offered by Cloudflare.

There are other micro optimizations you can make, but those are the big ones. This is the number two AI search factor from Kevin Index growth memo.

We have nine more to go. If you missed the first one, I recommend going back and watching it.