The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Share
AI Measurement: The Practical Layer
Published about 3 hours agoΒ β’Β 6 min read
β
Use Data Or Be Used By Data!
The June 29 issue of Seotistics is here for you!
The last time we talked about AI measurement and what changed.
Turns out a lot changed indeed!
I will go deeper on some concepts that the last time weren't covered with due credit.
P.S. Next week the last part of this series.
Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users can read this tutorial to do it.
Use the code "MARCO" to get 25% on the ticket at checkout, see you there!
And... I also recorded a podcast about LLMs and Analytics today. You can listen to the full episode here.
β
Similarweb Study - Pt. 2
We briefly mentioned Similarweb's study last time and this is one of the best studies out there.
To recap, people who found your website via AI are more likely to google it later.
The only caveat in this study is that identifying the same user is not that simple in real life.
Regardless, it looks like a reasonable condition.
Also, remember that branded terms matter and ARE used by Google to gauge your domain too.
If you think about it, most of our journeys now involve LLMs.
I often use AI Mode or Claude to do research and get opinions if I have to purchase something.
This means that AI is already part of our lives and not as marginal as some still claim.
The downfall of measuring branded searches today is that most of your queries will be anonymized... β and that also includes branded ones!
The Measurement Hierarchy
The framework I propose is very simple and intuitive.
It doesn't have to be sequential or perfect. Get some numbers to estimate how visibile you are and then start running comparisons.
First, you measure if you even appear in LLMs. You can achieve this with software like Waikay, DataForSEO, BWT and GSC.
This means seeing if you get citations, mentions and calculate your "visibility".
Then, you can use tools like GA4 (or alternatives) to estimate the contribution of AI to your conversions.
Finally, you can estimate the incremental change in revenue or other metrics... which is not so easy to accomplish.
The real issue with this approach is that attribution is very questionable and hard to trace back.
The perfect attribution model doesn't exist. β This is more of a philosophy to drive your decision making.
Citations and then?
I measure citations and visibility with some examples to show we exist in LLMs.
Actually attributing these results to AI is a whole different story for the next issue.
My advice is to measure variations over time in conversions, branded searches and organic.
This isn't easy because you work on multiple changes at the same time π
Prompt Tracking & Other Experiments
I am working a lot on this topic recently, especially from a measurement perspective.
It's all but solved and there are a lot of open questions...
so you need to test consistently and give reasonable output to your stakeholders.
If you are not visible, people don't remember you and are less likely to buy.
This is a very simplified version of what happens but you know it's true.
I've been tracking prompts as an experiment to see if there is variability, along with the sample outputs.
Waikay shows that for this prompt across 3 different models, the website I am tracking is the most visible!
(Will post more about this once I have examples I can show in public).
What I noticed is that yes, there is variability but I often noticed that some websites were always mentioned.
Remember, LLMs generate a new reply every time BUT there are some common patterns and names. β I highly doubt it brands like Apple aren't getting mentioned every time if you want a good smartphone.
Once you analyze the patterns, you should see which passages you miss or what information you don't have.
As you may have noticed from the picture above, not every model is equal... as we will see the next time!
Content Cycle & Automation
Before we start, most of this stuff was always true but not many cared because you could get results more easily.
Content is just the gateway with which people can find your brand, not even website.
As you may have heard multiple times, your brand isn't just your website but the entire environment:
social media (Reddit incl.)
your community
3rd-party websites citing you
Which is also why you need to go all in with improving your processes, especially in B2B.
What takes the most time is researching topics and well... making sure what you write makes some sense.
Now a good chunk of the research phase you can outsource to AI and review the output.
I try making my automation as simple as possible because otherwise it's a mess to debug/update over time.
My n8n automation is kinda done and now I am improving it and making sure it's even better.
Will tell you more next week ;)
While this has nothing to do with measuring visibility... it's an important part to work on your content in 2026.
And maybe, just maybe, have some agents review your content and give you advice from different angles π
I am currently using openrouter.ai to play with several models at the same time, will tell you more later!
Google Not Giving Much
Google added on some GSC properties AI visibility... but honestly no one cares.
Apart from the fact we don't have exports, we don't have actual visibility (like Bing) either.
Impressions per se is a pointless metric and you can only use this report to have a vague idea of your visibility.
Sort of brand awareness... but a citation or an impression don't mean the person saw your website.
For the website above, we are actually comparing 2 different timeframes: Bing = Jan to June, GSC = May to June.
By comparison, we can infer (obviously) that Google brings more visibility!
Then we can't say much because GSC doesn't have anything to say.
So It's All Speculation?
Yes because no one has the crystal ball BUT you can act on it.
Don't waste gazillions on tools, pay what you need and focus on action next:
having original content AND in the right passages
shift from just posting to distribution-first
build a process to review existing content and write new
It looks easy, right? Trust me, in reality this is incredibly complex due to politics and time allocation.
This is why you need to figure out a smart way to prepare more content and evaluate existing one.
But Marco... more content isn't always go- β Yeah no but if your competitors are faster than you, be it articles or a catalog, you know how it ends.
Now that you have some more notions, prepare for next week and my LinkedIn posts, I will talk a lot about some points I missed!
Until next week!
π₯ Join Our Community
Our Discord community offers a small place where we can talk business and web data.
If you hate all the noise of social media, then this place is for you.
I will start posting more there as we have a forum channel now.
This is the best way to stay updated in real time on Seotistics:
The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Read more from Seotistics - Web Analytics + Business + Strategy
Use Data Or Be Used By Data! The June 22 issue of Seotistics is here for you! LLM/AI measurement is on the mouth of everyone... but is it actually possible? I don't want to give you the secret to all problems because it doesn't exist. In fact I will show you why most of this is not as easy as you imagine. This is just an intro for the next content. P.S. I will also write a longform article on the topic, this is just the starter. Please move this email to your Primary inbox or reply to it....
Use Data Or Be Used By Data! The June 16 issue of Seotistics is here for you! The more I work in Web, the more I realize the shift that is happening is more real. SEO, PPC, Social media, you name it, it's just a different game now. Yet, people wage their marketing campaigns with outdated tools and tactics. It's always fun to come back to old problems with new perspectives and tools! Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by...
Use Data Or Be Used By Data! The June 8 issue of Seotistics is here for you! After one of my best weeks in terms of Web Analytics events, I am here to share what we learned. From Wednesday to Saturday I attended 4 different events in Copenhagen organited by the Brand Leadership Community. Naturally I must inform my readers about what I learned and what's going on in Web Analytics! Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by...
