The most important metric you’re not tracking
A new mental model for marketers navigating the shift from attribution to influence
👋 Hey, I’m George Chasiotis. Welcome to GrowthWaves, your weekly dose of B2B growth insights—featuring powerful case studies, emerging trends, and unconventional strategies you won’t find anywhere else.
This note is brought to you by Above Apex
Above Apex is a SaaS link building partner focused on earning high-quality links and brand mentions that impact organic results and AI Search.
They’re trusted by teams at Synthesia, Ketch, Improvado, Action1, Influencity, and more, and help SaaS brands get featured on sites like Forbes, HubSpot, GoDaddy, Zapier, and Wix.
Will Homden (Head of SEO at Synthesia) calls the team “professional, transparent, and excellent communicators,” specifically highlighting the clarity of their process.
Explore their approach in the link below.
What are the questions we’re not asking?
The assumptions and notions we’re not questioning?
The insights we aren’t paying attention to?
The metrics we’re not tracking?
These are the questions that keep coming up when I think about marketing.
Especially the last one—metrics—has been on my mind for a while.
And it’s not just me; the need to update our measurement system is a common theme in my conversations with other marketers.
I don’t think there’s a definitive answer to this question.
And obviously, the answer is case-specific and will likely change in 6, 9, or 12 months, given how fast everything moves.
In this note, I’ll share my thoughts on one of the metrics that is becoming—or, to be more accurate, will become—really prominent when it comes to marketing.
Let’s go.
Why do we need new metrics?
We were on a call with a Minuttia client a couple of months ago to review the results of our work. We had:
Top 10 rankings for key Google search terms
Visibility in AI search for several relevant prompts
Visibility in AI Overviews for many high-intent queries
Organic impressions trending up
Organic traffic steadily growing
All good, until our client asked...
Things have changed. Shouldn’t we rethink what success looks like?
… our team didn’t have a good answer, and this was the signal I needed to start working on improving our approach to measurement and impact.
Let’s zoom out and look at the bigger picture for a moment.
Things have changed a lot in the last 3 years. (Since the launch of ChatGPT in November 2022.)
And it’s not just ChatGPT—in the last 3 years, we’ve all seen a massive wave of change across multiple fronts.
Author’s Note: Click on the image to zoom in.
Which means many of the success metrics and signals we relied on no longer serve us.
We need new ways to understand—and win in—this ever-changing, increasingly complex landscape.
One where visibility and influence come first, and attribution follows. As I shared a few weeks ago:
The future of marketing will center around visibility. We’re moving from attribution-led to influence-led models. The marketers who’ll win are the ones who can influence the customer journey and clearly communicate that impact.
Marketing used to be about influence, but somewhere along the way, it shifted towards attribution-led models. There are several reasons for this transition:
Technology has come to a stage that allows for attribution-led models
Marketing started getting equated (and unfortunately, still is being equated in several cases) with revenue operations (RevOps)
Investors started demanding more transparency and performance clarity, especially after the honeymoon period of the latest ZIRP period (March 2020 to March 2022)
Channels with easier-to-prove ROI influenced the perception regarding attribution for channels with hard-to-prove ROI (e.g., paid search vs organic search)
During this transition, we moved towards a more ROI-first, scrutinized approach to marketing activities, where every marketing dollar spent should be matched by an equivalent marketing dollar earned through return on investment.
Needless to say, this shift hasn’t been good for marketing.
The fact that there’s software that promises customer journey clarity doesn’t mean that a) the insights these tools provide are accurate, and even if they are, b) that we should obsess over attribution.
The problem is that marketing teams should, by default, have the space to execute on initiatives where ROI isn’t even part of the conversation.
Otherwise, marketing becomes an attribution center where creativity and experimentation suffer, and only easy-to-prove initiatives and channels prevail.
That’s not good for anyone, because in the long run, companies with this mindset will miss out on the benefits of all these initiatives and channels with obscure attribution, but that play a big role in building attention, building authority, and mental availability for the company’s products and services.
To be clear: I’m not suggesting that companies shouldn’t use attribution models and software.
On the contrary, they should, but always under a balanced system where both performance, creativity, and experimentation can co-exist and thrive.
Having said that, emerging marketing channels over the past three years have highlighted the need for a new understanding of the customer journey, including all the touchpoints that influence decision-making.
One such channel is AI search.
AI search: A new (somewhat familiar) channel
AI search and the practice (or craft, depending on how you see it) of optimizing for it (aka AEO, GEO, LLMO) is a relatively new channel that came into the mainstream in the past couple of years, following the launch of ChatGPT.
There are two types of search in AI search engines, such as ChatGPT, Perplexity, or Google’s Gemini:
No Web Search: When you ask a question, ChatGPT may answer using only its training data, which is based on information from the open web up to the model’s knowledge cutoff. In this case, the response is generated without checking live sources and works best for general knowledge or topics where freshness is not critical.
Used Web Search: If a question requires up-to-date information or if ChatGPT detects gaps in its existing knowledge, it may use web search to enrich the answer. It does this by running multiple search queries, reviewing top results from traditional search, and using those sources to produce a more current and informed response.
Marketers should mostly be interested in the second type of search.
When the AI search engine uses web search to answer a particular question, then as a marketer, you’d want to know whether your brand is being mentioned (if this is a question with commercial intent), in what position, whether any resources on your website are being cited, and so on.
Generally speaking, the metrics and insights we’re interested in at this point for this emerging channel are:
Visibility Score: The percentage of responses where your brand is mentioned.
Share of Voice: Your brand’s mentions compared to all company mentions.
Average Position: The average position your brand is mentioned in AI-generated answers.
Citation Share: How often your brand is cited by AI-generated answers.
Sentiment: How positively AI responses reference your brand and what its overall perception is.
Query Fanouts: The queries AI conducts on search to satisfy a question by a user.
The team at Minuttia has built its own AI search-tracking tool—you can request a free AEO report via the link below.
So, visibility monitoring and tracking is—to a certain extent—solved.
It’s also a commodity, as we’ve established in a report we published in December 2025.
Essentially, what I’m saying is that tracking your visibility on AI search is covered by all these solutions that exist in this software category, or one you can build yourself.
That’s level one when it comes to this new and emerging channel.
Level two is gaining insights into how to act on your visibility status and opportunities, which some of these tools are already doing.
Before delving any further, let me just say that there’s a lot of snake oil salesmanship in all this, as the specifics of what drives visibility on these platforms remain a black box, at least for the time being.
To add to that, there are so many differences across platforms, which make us realize there isn’t a universal optimization solution or strategy.
Even within the same platform (Google search vs Google AI Mode), there are significant differences—you can imagine how much wider that spread is across different platforms.
So optimization is level two, and the industry is trying to solve it.
When I say industry, I mean the SEO industry, which “adopted” AI search due to the overlap between SEO and AEO, plus the fact that, regardless of the query length and what results look like, for the most part, we still have a search box and an answer.
Just like traditional search. (I’m oversimplifying, but you get the point.)
We can debate on how big that overlap is, but there’s certainly a big overlap here.
Okay, so at this point, you may wonder, “What’s level three?”
Let’s take a look.
Introducing: Perception Deviation
What’s interesting about AI search is how answers are formed from a variety of sources.
If we were to visualize how perception is being formed, it would look something like this:
For those of you who prefer text over visuals:
Brand Experience: How people experience your product and brand shapes perception before content, links, or mentions ever matter.
Owned Perception: Your own content and channels let you control the narrative and clearly communicate positioning, expertise, and value.
Multi-Channel Perception: Your brand’s reputation across platforms like G2, Trustpilot, Reddit, YouTube, and communities reinforces or weakens credibility signals.
Brand Mentions and Backlinks: Mentions and links still matter, especially in listicles, comparisons, and third-party content that AI systems frequently reference.
When it comes to forming a perception, we have two important concepts:
Ground Truth: Ground Truth is the brand’s own understanding of itself—how it believes it creates value, who it serves, and what it stands for, before any AI system or external audience interprets those signals.
Formed Perception: The perception that emerges from how AI systems—and, as an extension, people—synthesize signals about your brand across owned content, third-party platforms, brand mentions, and broader multi-channel reputation.
These two concepts give rise to what I call Perception Deviation, which can be defined as follows:
The gap between Ground Truth and Formed Perception. The extent to which your brand’s reality diverges from how it’s understood, represented, experienced, and inferred across AI search and the wider digital ecosystem.
The mental model that led us to the concept of Perception Deviation is the following:
Attribution-led marketing models
Shift to Influence-led marketing models
AI search engines influence the customer journey
They influence it based on the perception they’ve formed
The perception is formed based on the following attributes:
Brand Experience
Owned Perception
Multi-Channel Perception
Brand Mentions
That perception may differ from the ground truth
Which means we need a way to identify and quantify that gap
Here’s a visualization of the mental model:
Even though there might be some solutions that attempt to solve the Perception Deviation challenge, I’m pretty sure none of them have figured it out yet.
And not just because of personalization in AI search (which makes universal Formed Perception almost impossible), but because Formed Perception and Ground Truth aren’t easy to contextualize.
This is one of my next big tasks for Minuttia’s AI search tracking tool—helping companies measure the Perception Deviation relative to their Ground Truth and identify all the sources that might be contributing to an AI search engine's perception that's off.
I’ll keep you updated, of course.
In the meantime, I just wanted to introduce this new concept and stress the importance of not just tracking brand mentions and optimizing for AI search, but also of understanding whether your brand's perception aligns with your ground truth.
Let’s wrap this up.
Final thoughts
A while ago, one of my favorite entrepreneurs, renowned real estate investor Barry Sternlicht, said in an interview:
Managing your perception is as important as reality.
I agree with Barry, and I believe that we’re moving towards a more influence-led future for marketing, where visibility and perception become more important.
In that context, managing your perception—online and offline—may soon matter as much as reality itself.
Thank you for reading today’s note, and see you again in two weeks.