Brand building has one rule that never changes.

Read on

People buy brands they think of first.

Decades of brand science established it: memory structures drive market share. Brands grow through mental availability, by coming to mind easily, in more buying situations, for more people.

No technology repeals this rule. What technology changes is how memories get made.

Soon, most digital brand discovery will happen through AI models. People will still develop brand preferences. But their memory of brands will be fed, every day, by what AI models remember.

So your brand now lives in two memories. The human one you have always built. And the AI one, assembled from what everyone else says about you.

The AI brand architecture.

Brand science explains how human memory works: category entry points, distinctive assets, mental availability. We built the architecture that maps those same mechanics onto AI memory.

REPRESENTATION Discoverability Accuracy Consistency SIGNAL Narrative coherence · Channel architecture Earned authority · Review alignment BRAND SUBSTANCE Distinctive assets · Creative density · Category courage Cross-category reach · Risk signature

Our AI insights shape our strategy.

There are three rules we observe.

AI memory follows Double Jeopardy.

Brands AI rarely mentions also get recommended less warmly when it does. Small brands stay small, in both memories.

The way to build a strong digital brand is the same as it has always been: show up in more situations, on more platforms. Breadth wins.

AI memory trusts other people, not you.

What reviewers, communities and independent voices say counts for more than anything you say about yourself.

You cannot buy your way in. You can only earn it. This introduces a new dynamic where brand strategy and community are interconnected at the strategic level, not a channel plan.

AI memory runs behind.

What people say about your brand today becomes what AI says about you later. AI training data is cached like search engine data. It just has more impact on decisions than search engine data has.

We believe brands can build stronger digital brands now to prepare for AI mediation. Build the advocacy infrastructure now and you will be ready for the future. Our brand science engine exists to benchmark your brand infrastructure today and build for the future.

So where do you start?

You start where the future is already visible. Right now, digital brand memory shows itself in one place: the answers AI gives when someone asks about your category.

But the answer is the AI memory, not the strategy. The memory consists of your reviews, your communities, your coverage, and how consistent your story is across every channel an AI model can read.

We score what AI says about your brand, where, and why. We trace every digital brand signal. And we connect those signals to a brand science framework to translate it into a future facing strategy. Sometimes an amazing brand just needs some tech support. Most of the time the community is absent and there are years of neglected digital brand assets.

Five frameworks behind the work.

Framework 1
The ATTENT10N Score
TECH SIGNAL STRENGTH
Three weighted layers — Technical Foundation, Brand Signal and Brand Strength — combined into your Brand & technical health score, 0–10.
Framework 2
Brand Bravery
Five-component score of structural distinctiveness. Cohort-relative — placement matters as much as the absolute number.
Framework 3
AI Brand Architecture
REP SIGNAL SUBSTANCE
The framework that replaces traditional brand architecture for AI: representation (30%, top) over brand substance (70%, the foundation).
Framework 4
Community Lens
COMMUNITY CANCELLED FUNCTIONAL
Three-state diagnostic of how the brand shows up in conversation: Community, Cancelled or Functional. The state implies the lever.
Framework 5
Structural Surface Audit
L1 L2 L3
Three-layer audit of machine-readable presence: brand-owned, brand-adjacent and discovery surfaces.

That’s the methodology.
Here’s how it works.

What We Actually Measure (and Why It Goes Far Deeper Than Short-Term AI Rankings)

The Attention Score

A single diagnostic score (0–10) that reveals how strongly AI surfaces your brand today across three layers of specificity.

Mass
Broad, high-volume questions that drive everyday discovery.
Segment
Occasion or audience-specific triggers (e.g. students, value seekers, family groups).
Niche
Precise questions about your exact offering.

We don't sample a handful of prompts. Every report puts around 100 questions to each AI model — roughly 400 real conversations about your brand across ChatGPT, Claude, Gemini and Perplexity — spread deliberately across these three layers, from broad category demand down to your exact niche. So your score reflects the full breadth of how customers actually ask, not a lucky or unlucky few. If you're benchmarking us against other tools, ask how many questions sit behind their score.

We reverse-engineer exactly why AI behaves the way it does by scoring the raw material it has to work with.

Technical Foundation 20%
Machine-readable signals your brand controls directly: schema markup, llms.txt files, sitemap freshness, structured data, and depth of authoritative sources. Engineering-shaped fixes that deliver results in weeks.
Brand Signal 30%
Consistency, volume, and accuracy of information about you across the open web and your owned channels.
Brand Strength 50%
The deepest and most strategic layer. Distinctiveness, cultural presence, community health, brand bravery, and most importantly, Category Entry Points (CEPs).

CEPs (a foundational concept from Ehrenberg-Bass brand science) are the real-world triggers that make someone think of your category in the first place: a long day at work, a student budget on Monday, a craving for something fresh and customizable. Brands grow by becoming mentally available at the most relevant CEPs. AI is now the largest single amplifier of those memory structures. If AI does not link your brand to the right CEPs, those mental shortcuts never form, or they weaken over time.

This is where we diverge sharply from short-term GEO or AI-ranking tools. We harvest and model longitudinal data across AI platforms to build a living picture of your brand's position.

AI conversation share
Over time, tracked month over month.
CEP ownership
Which buying moments AI consistently links to you versus competitors.
Community health
Transactional mentions versus genuine cultural conversation.
Brand drift
How AI perceptions evolve as new content enters the training and retrieval layers.

The result is a brand world model that shows how AI-mediated discovery is actively building, or eroding, your memory structures and market position.

How We Measure (and Why the Signal Holds Steady)

Ask any AI the same question twice and you can get two different answers. That is exactly why most AI-visibility tools feel unreliable — they watch a single, fluctuating output and present the wobble as a metric. We are built the other way around. Five things keep an ATTENT10N reading dependable:

Measured at the foundation
We score the durable signals AI actually draws on — your channels, content, reputation, distinctiveness and Category Entry Points — not just the answer it happened to give today. Those foundations move on a strategic timescale, so the reading is steady by design.
The same evidence, the same score
AI is a research instrument in our process, never the judge. The score is produced by a fixed, deterministic engine, so identical inputs always return an identical result — we never add measurement noise of our own.
Structural findings locked between runs
Your channels, reviews and reputation themes are pinned and reconciled from one run to the next, so a model phrasing something differently on a given day cannot masquerade as real change.
Inconsistency becomes the diagnosis
Whatever variation remains is the point, not a flaw. When ChatGPT, Claude, Gemini and Perplexity disagree about you, that gap measures how clearly your brand reads to AI — we surface it as a finding instead of hiding it.
Read as a trend, not a snapshot
We assess your position across recent runs rather than any single one, so genuine movement stands out and momentary blips do not.

How Often Should You Run a Report?

Because we measure brand foundations and strategic position — not a daily ranking — ATTENT10N is a periodic diagnostic, not a dashboard to check every morning. The full AI Brand Health Report works best quarterly, with the lighter Brand Health Snapshot as a monthly pulse-check in between.

The full report, quarterly
A quarter is long enough for content, PR, community and positioning work to actually register with AI — and to read the trend rather than the noise.
A snapshot, monthly
The Brand Health Snapshot is a lighter, faster read — ideal as a regular monthly check on movement between the deeper quarterly reports.
After you act, and on big changes
Re-run when you have implemented the recommendations to confirm the levers worked, or when something material shifts: a repositioning, rebrand, major campaign, or a new competitor.

Daily and weekly AI-visibility monitors are built for a different job — an SEO team watching output move in near real time. That cadence suits tracking a number that changes by the hour; running a deep brand diagnostic that often would only re-measure foundations that shift over months. Think of ATTENT10N as the strategic read on where your brand stands in AI — the moves it surfaces are usually creative, positioning and brand-building plays that take a quarter to land, not overnight technical tweaks.

From Diagnosis to Action

Our process turns the black box of generative AI into a measurable, actionable part of your brand strategy.

Every time AI recommends a competitor instead of you, it reinforces that competitor's memory structures and quietly weakens yours in the minds of the next wave of customers.

We combine rigorous measurement of AI visibility with deep brand science to give you a strategic advantage in the age of AI-mediated discovery. We pair CMOs with our leading AI brand science engine to produce all of our reports. Self-service tools can't go deep enough to help brands develop impactful strategies that move the needle, especially when the solution is often creative or positioning versus technical gaps.

Book a call to discuss our approach →