What is an AI Website Analysis?
An AI website analysis is the practice of reviewing a website using AI tools to assess its quality and value, as well as analyzing the website to understand its AI visibility score and ability to appear in AI search results.
Using AI tools to analyze a website for different purposes
There are thousands of reasons someone would want to use AI tools to analyze a website. In essence, AI tools promise the ability to understand and interpret a much broader range of problem sets.
Before AI, website analysis was limited to those with more technical abilities. Website crawlers mimicked how search engines would crawl and evaluate a site, for SEO and paid search services. Social media sites would crawl websites for contextual information about brands. Traders and hedge funds acquired data from alternative data aggregators and providers to get additional signals not typically used in trading.
Analyzing a website for AI visibility
Business and marketing leaders want to know how well their website is showing up in AI chatbots and LLMs like ChatGPT, Claude, and Gemini. There are developing visibility tools for understanding AI visibility for brands, such as Profound, AirOps, Peed, Gumshoe and Amplitude, among others.
Google Analytics can still be powerful for understanding which referral traffic is coming from AI search, via referral UTM codes, like ChatGPT, adding parameters on every link.
Google Search Console is not yet helpful for analyzing this side of things. If Google and Gemini want to take the lead, they will start. If they do this before OpenAI, they could win over SEOs, content creators, and other builders.
Speculating further, Google learned a lot in their loss in social media, yet they still grew thorug search and YouTube. They built out a strong YouTube creator program and actually won video. I expect them to rapidly redeploy their YouTube growth model to the Gemini suite and push all of their AI tools across their entire organization.
We could get to the point soon where optimizing for Google means optimizing for their entire stack of products – both using their generative AI products as well as building, creating, and optimizing for their platforms across Google Search, Maps, Ads, YouTube, Gemini, and more.
One threat to Gemini is Sora. While Google has Nano Banana Pro + Veo 3, they have not built a creative AI-generated video feed like Sora has. Meta has tried to launch in this space but it has not caught on.
While we may not want this right now, this may be where the next big action is: who owns the generative AI video feeds.
You thought TikTok was intense; think about a non-stop flood of AI-generated videos on any subject. Go create with Grok and see how fast they are generating.
As Elon recently said, real-time is the next big thing and cannot be defeated.
Jensen is talking about what “generative” actually means. It’s not retrieval-based.
See this video for context:
Connecting SEO keyword rank-tracking with AI visibility
One approach we’ve observed and have developed internally is mapping keywords tracked for traditional SEO and search purposes and expanding those to be a series of prompts that might be used in chat LLMs and AI search.
At the most practical level, a brand should take the top 50 keywords they typically test, and do a pilot discovery project to analyze what prompts map to those keywords, and how they might show up in AI search engines. Applying the 5Ws: who, what, where, when, why and how – is a great way to get started.
But this should be customized to the context of the company and the industry and their customers’ own journeys. This approach is an evolution of past SEO, content, PR and other marketing channels rather than a wholesale rewrite.
