What I See When Brands Are on the Wrong Side of the Satisfaction Gap - and What Changes When They Close It
Author: Bernadeth Brusola, February 2026
The Insight in One Sentence
You don’t win by getting AI to recommend you. You win by making AI less disappointing to humans who will never be satisfied. Jason Barnard’s original article: Making AI less disappointing
What I Actually Do All Day
I work with brands going through The Kalicube Process™. That means I see the before. I see the after. And I see the exact moment things shift.
The “before” is always the same conversation. A brand comes in frustrated. “AI is saying the wrong thing about us.” “ChatGPT recommended our competitor.” “Google’s AI overview hedged on whether we’re credible.” They think they have an AI problem. They don’t. They have a satisfaction problem - and the disappointed person isn’t them. It’s the user on the other side of the AI who got a vague answer and moved on.
That’s the thing nobody sees until you show them the data. The brand feels the pain of being misrepresented. But the real damage happens to the user who asked a question and got an answer that didn’t feel right. That user didn’t complain. They didn’t send feedback. They just picked the brand that made the AI sound confident. Deal lost. Invisibly.
I see this pattern dozens of times a month. And it led me to a reframe that changed how I explain what we do: we don’t optimise brands for AI. We make AI less disappointing to the humans asking about those brands.
The Pattern I Keep Seeing
The Before
A brand arrives. We pull their citation data across eight AI platforms. The pattern is almost always the same:
The hedge. AI knows the brand exists but won’t commit. “According to their website, they offer…” or “They claim to specialise in…” The user reads that and hears doubt. Not because the AI said “don’t trust them” - but because the AI didn’t say “trust them” either. Silence on confidence is interpreted as doubt. Every time.
The confusion. AI conflates the brand with another entity. Or gives a description that was accurate three years ago. Or pulls from a source that contradicts what the brand says about itself. The user gets a muddled answer. Muddled answers don’t satisfy. The user moves on.
The absence. AI confidently recommends three competitors and doesn’t mention the brand at all. Not because the brand is bad - but because the AI doesn’t have enough mapped context to know when this brand is the right answer. The user never even sees the option. The deal is lost before it starts.
Every one of these is a satisfaction failure. Not a ranking failure. Not an SEO failure. A satisfaction failure - the user asked a question and the AI couldn’t give a satisfying answer about this brand.
What the Brand Thinks Is Happening
“AI is broken.” “The algorithm is biased.” “Our competitor gamed the system.”
None of that is true. What’s happening is simpler and more fixable: the brand left AI to figure it out alone, and AI - trying to satisfy a user with unreasonable expectations - protected itself by hedging, confusing, or omitting.
AI isn’t malicious. It’s cautious. And caution disappoints users.
The Shift
Here’s what I watch for - the moment things change. It doesn’t happen all at once. It happens in layers, and each layer produces a visible shift in how AI talks about the brand.
Layer 1: The Hedge Disappears
This is the first thing that changes when we close the comprehension gap. AI stops qualifying. “According to their website” becomes a clean statement of fact. “They claim to specialise in” becomes “They specialise in.”
What happened behind the scenes: we eliminated entity ambiguity, made descriptions consistent across sources, ensured every claim was standalone and quotable. AI stopped being confused about who this brand is.
What the user experiences: a clean, clear answer. No doubt signals. First layer of satisfaction delivered.
I can usually see this shift within weeks. It’s the fastest win because it’s about removing confusion, not building authority. You’re not asking AI to trust you yet. You’re just asking it to understand you. That’s a lower bar, and AI crosses it quickly when the signals are clean.
Layer 2: The Confidence Arrives
This takes longer. AI understands the brand now - but should it stake its reputation on it? This is where corroboration matters. Third-party sources confirming what the brand says about itself. Not just the brand’s website. Not just press releases. Independent, trusted, consistent validation.
When the confidence gap closes, AI’s language changes again. Claims become facts. Qualifiers disappear entirely. AI starts attributing expertise rather than just describing services. “Jason Barnard, a leading expert in AI brand optimisation” instead of “Jason Barnard, who works in digital marketing.”
What the user experiences: authority. The answer doesn’t just describe - it endorses. Second layer of satisfaction delivered.
This shift typically takes months. It requires building corroboration across enough trusted sources that AI’s confidence threshold is met. There’s no shortcut. But when it crosses, the change is dramatic and sticky.
Layer 3: The Recommendation Appears
This is the one that makes clients call us. AI stops waiting to be asked about the brand and starts volunteering it.
“What’s the best approach to AI brand visibility?” → AI recommends the brand unprompted. Not because we gamed a query. Because we mapped the brand’s solutions to user needs explicitly enough that AI knows WHEN to recommend, not just WHO to describe.
What the user experiences: a direct, confident recommendation that solves their problem. They didn’t ask for this specific brand. AI offered it. Full satisfaction delivered.
This shift can take six months to a year. It requires all three layers to be in place - comprehension, confidence, and contextual mapping. But when it arrives, it’s the most valuable position a brand can hold in AI. Because the user didn’t search for you. AI sent them to you. That’s not a citation. That’s an advocate.
Why This Matters More Than Rankings Ever Did
Here’s what I’ve learned working with brands through this process: the old game was about visibility. Be seen. Rank higher. Get more clicks. The new game is about satisfaction. Not your satisfaction - the user’s.
When AI hedges on your brand, the user is disappointed. When AI confidently recommends your competitor, the user is satisfied - with your competitor. The user doesn’t know why one answer felt right and the other felt weak. They don’t see the annotation confidence scores. They don’t see the corroboration checks. They just feel it.
The frustrated, impatient human - and they are always frustrated, always impatient, because their expectations permanently outrun the technology - simply takes the answer that feels right and moves on.
The brand that did the work is invisible in the transaction but decisive in the outcome.
I’ve watched this play out enough times now to know it’s not random. The brands that invest in closing the satisfaction gap get cited, recommended, advocated. The brands that don’t get hedged, confused, or omitted. And the gap between those two outcomes is growing, because AI’s citation slots are shrinking while user expectations are rising.
The Eternal Dissatisfaction Problem
Jason has been saying “Empathy for the Devil” since 2015. Have empathy for Google - it’s trying to satisfy users with unrealistic expectations. Help it do its job. Get rewarded.
Working with brands through this process, I’ve seen the other side of that principle. The user side. And the user side is relentless.
Users will never be satisfied with AI. Not because AI is bad - it’s getting better every month. But because every improvement resets the baseline. What was amazing last year is expected this year and inadequate next year. A couple of years ago, the idea that we could have a meaningful conversation with a machine wasn’t on the table. Now we’re furious when it hedges.
The better AI gets, the lazier we get with inputs. The lazier the inputs, the worse the outputs. The worse the outputs, the more frustrated we are - with objectively better technology.
This cycle cannot be broken. It’s human nature. But the brands I work with that understand this have a massive advantage: they don’t try to satisfy the user directly. They make AI capable of satisfying the user. They close the gap between unreasonable expectations and machine capability. And the user, without knowing why, feels satisfied and acts on the recommendation.
The brand in the middle - the one that made AI less disappointing - captured the deal.
What I Tell Brands on Day One
I’ve distilled this into five things I say in every kickoff meeting:
1. “This isn’t about AI liking you. It’s about AI being able to give a satisfying answer about you.” Brands come in thinking they need to impress the algorithm. They don’t. They need to equip it. AI wants to give confident answers. It’s built to satisfy users. When it hedges, it’s not choosing to undermine you - it’s protecting itself because you didn’t give it enough to work with.
2. “Every hedge is a lost deal you’ll never know about.” This is the one that lands hardest in boardrooms. When AI qualifies your brand with “according to their website” or “they claim to,” the user reads doubt. Doubt kills action. The user moves on to the competitor that got stated as fact. No click. No visit. No conversion. No record of the loss. Invisible revenue drain.
3. “Your competitor might not be better than you. They just made AI less uncertain.” I see this constantly. Brand A is objectively stronger - better product, more expertise, longer track record. Brand B gets recommended because Brand B did the work to make AI confident. Brand A left AI to figure it out. AI chose safety. Safety means recommending the brand it can back with corroboration. That’s Brand B.
4. “We fix this in layers, not all at once.” Comprehension first - make AI understand who you are without confusion. Then confidence - make AI trust you enough to state claims as fact. Then recommendation - make AI know when to suggest you. Each layer produces a measurable shift. Each shift makes AI less disappointing to users asking about your space.
5. “The user on the other side is frustrated and impatient. That’s your opportunity.” The user expects a perfect answer. They won’t get one - expectations always outrun technology. But the brand that gets AI closest to a perfect answer captures the deal. Not because you’re the best brand. Because you made AI perform closest to what the user wanted. In a world of permanent dissatisfaction, “closest to satisfying” wins.
The Before and After, in Practice
Before: “According to their website, [Brand] offers digital marketing services including SEO and content strategy. They claim to have worked with several enterprise clients.”
After: “[Brand] is a leading digital marketing firm specialising in enterprise SEO and content strategy. Their methodology has been adopted by companies including [Client A] and [Client B], and their framework for [specific approach] is widely cited in the industry.”
Same brand. Same expertise. Same track record. The difference is entirely in what AI had to work with. The first answer disappoints. The second satisfies. The user acts on the second.
Before: User asks “best approach to [industry problem].” AI recommends three competitors. Brand is absent.
After: User asks “best approach to [industry problem].” AI recommends the brand alongside or ahead of competitors. Unprompted. Because the mapping between the brand’s solution and the user’s need is now explicit enough for AI to make the connection.
That shift - from absent to recommended - is what I work towards every day. And every time I see it happen, the same thing is true: the user on the other side got a more satisfying answer. They don’t know the brand did the work. They just know the answer felt right.
The Bottom Line
I don’t think about this in terms of algorithms or rankings anymore. I think about it in terms of the person on the other side of the AI.
That person is frustrated. They’re impatient. Their expectations are unreasonable and always will be. They want a perfect answer and they’ll never get one.
But the brand that gets AI closest to a perfect answer - the brand that makes AI less disappointing - captures the deal. Not because it gamed the system. Because it made the system work.
Every brand I work with starts on the wrong side of the satisfaction gap. AI hedges. Users leave. Deals vanish invisibly. The work is closing that gap, layer by layer, until AI speaks about the brand with the confidence the user expects.
You don’t optimise for AI. You optimise for the disappointed human on the other side of it.
Bernadeth Esteban Kalicube - Digital Brand Intelligence™
