For years, "visibility" meant one thing: rank well on Google, and the traffic follows. That world is largely extinct.
The emergence of AI has fractured search and discovery into a dozen different surfaces: ChatGPT, Google’s AI Mode & Gemini, Perplexity, Claude to name just a few, each with a different way of sourcing, interpreting, and citing brands. And for a lot of brands, that's exactly where the strategy breaks down: they're chasing pure visibility: "which platform do we need to crack next", without asking the harder questions:
Are we showing up in the sources that actually matter? Are we actually being interpreted correctly on these platforms? Are we being recommended, or just mentioned?
Different models source and cite information for different reasons. What earns you a citation on one answer engine won't necessarily earn you one on another. Treating AI as a single channel, rather than a fragmented ecosystem, undermines a lot of otherwise well-intentioned strategy before it even gets started.
So what does it actually take to dominate answer engines? In my view, it comes down to three things: being Found, being Understood, and being Chosen. Miss any one of the three, and the others won't save you.
Being found isn't about being everywhere. It's about being present in the places your brand should already be part of the conversation - earning mentions in the media, at the sources, and at the moments that actually matter to your audience.
Here's what surprises a lot of people: even though a brand's own website now accounts for only around 10% of the sources AI cites, that 10% is disproportionately important. Your site is still your storefront, your HQ, your postal address. It's your customer service department, your voice, your company line, your source of truth. AI might not lean on it heavily for citations, but it's still where the machine, and your customers, go to verify everything else.
This isn't a hunch - Moz's own analysis of nearly 40,000 queries found that 88% of citations in Google's AI Mode don't appear in the organic top 10 for the same search, and only around one in five citing sites ranks in the traditional top 10 at all. The mechanics have decoupled, but that doesn't mean the investment can.
That's exactly why the old argument that "AI will find you anyway" is so dangerous. It won't. If you stop investing properly in SEO because you assume AI visibility is a separate game, you don't just fail to gain ground in AI search - you lose your traditional visibility at the same time. It's not an either/or risk. It's both, simultaneously.
Being found gets you into the conversation. Being understood determines what gets said about you once you're there.
This is the part brands can't fake. AI search doesn't just surface your best press release - it surfaces the good, the bad, and the ugly. If you position yourself as a premium brand to justify a higher price point, but your product quality is average, your content is rushed, and your customer service is outsourced and slow, the model will pick up on that mismatch.
You can't out-message a bad experience anymore. The gap between what you claim and what's actually true about you is exactly what the machine is trained to notice.
Every brand says they’re the best - look at the brand you work on and its homepage…. Now similar brand promises, claims and “USPs” are made on your competitors’ homepages too; strong sounding superlatives, cherry-picked reviews and glowing testimonials. In the pre-AI era that was enough to work.
People were funneled through their various stages and every brand put their best foot forward. It took time and effort to compare reviews, deep-dive threads, mine social comments or see what the key influencers in the space recommended. In the AI era all of that is done for you (whether you ask for it or not) and it’s surfaced.
As a result, being recommended by others is absolutely crucial to being chosen. If you’re visible but not you’re not recommended - you’re just background noise.
Proxies we used to use for organic discovery success like brand mentions, links and coverage in key publications still matter but more impactful and effective recommendation metrics; press endorsements, review sentiment and signal clarity (what attributes and associations are people actually making with our brand) are key in being recommended and chosen.
The typical mistakes we’re seeing involve marketers doing the wrong things, often expensively. Some brands pour serious budget into tracking and measurement tools and assume that visibility follows automatically even though measurement doesn't move the needle on its own.
Others fall back on tactics that were already spammy for traditional SEO and are even more transparent to AI: manipulating rankings, buying placements at scale, engineering their own listicles and putting themselves at the top, or pouring all their effort into their own website while ignoring the other 90% of what actually gets cited.
A useful test is: does the tactic sit at the intersection of three things: genuine user experience, discovery by your audience, and protecting your brand's image? If a change serves all three, it's worth testing. If it only serves one - usually "get found faster," it's usually the kind of shortcut that erodes trust rather than building it.
Measurement in AI search is messy, and I'd rather be upfront about that than pretend otherwise. Answers vary between requests, between models, and over time, so a single number is never going to tell the full story.
It’s useful to separate your AI visibility measurement into three types:
None of these alone will provide a clean, defensible number, but together, they do help join the dots and report progress in a way that's credible.
Dominating answer engines was never going to be about a single tactic or a single platform. It's about engineering discovery deliberately, across all three stages: Found, Understood, Chosen - rather than hoping that ranking well in one place carries you everywhere else.
The brands getting this right aren't the ones spending the most. They're the ones being deliberate about all three, at the same time.
The emergence of AI has fractured search and discovery into a dozen different surfaces: ChatGPT, Google’s AI Mode & Gemini, Perplexity, Claude to name just a few, each with a different way of sourcing, interpreting, and citing brands. And for a lot of brands, that's exactly where the strategy breaks down: they're chasing pure visibility: "which platform do we need to crack next", without asking the harder questions:
Are we showing up in the sources that actually matter? Are we actually being interpreted correctly on these platforms? Are we being recommended, or just mentioned?
Different models source and cite information for different reasons. What earns you a citation on one answer engine won't necessarily earn you one on another. Treating AI as a single channel, rather than a fragmented ecosystem, undermines a lot of otherwise well-intentioned strategy before it even gets started.
So what does it actually take to dominate answer engines? In my view, it comes down to three things: being Found, being Understood, and being Chosen. Miss any one of the three, and the others won't save you.
Found: earning your place in the conversation
Being found isn't about being everywhere. It's about being present in the places your brand should already be part of the conversation - earning mentions in the media, at the sources, and at the moments that actually matter to your audience.
Here's what surprises a lot of people: even though a brand's own website now accounts for only around 10% of the sources AI cites, that 10% is disproportionately important. Your site is still your storefront, your HQ, your postal address. It's your customer service department, your voice, your company line, your source of truth. AI might not lean on it heavily for citations, but it's still where the machine, and your customers, go to verify everything else.
This isn't a hunch - Moz's own analysis of nearly 40,000 queries found that 88% of citations in Google's AI Mode don't appear in the organic top 10 for the same search, and only around one in five citing sites ranks in the traditional top 10 at all. The mechanics have decoupled, but that doesn't mean the investment can.
That's exactly why the old argument that "AI will find you anyway" is so dangerous. It won't. If you stop investing properly in SEO because you assume AI visibility is a separate game, you don't just fail to gain ground in AI search - you lose your traditional visibility at the same time. It's not an either/or risk. It's both, simultaneously.
Understood: is AI reading your brand the way you want it to?
Being found gets you into the conversation. Being understood determines what gets said about you once you're there.
This is the part brands can't fake. AI search doesn't just surface your best press release - it surfaces the good, the bad, and the ugly. If you position yourself as a premium brand to justify a higher price point, but your product quality is average, your content is rushed, and your customer service is outsourced and slow, the model will pick up on that mismatch.
You can't out-message a bad experience anymore. The gap between what you claim and what's actually true about you is exactly what the machine is trained to notice.
Chosen: Are you more than background noise?
Every brand says they’re the best - look at the brand you work on and its homepage…. Now similar brand promises, claims and “USPs” are made on your competitors’ homepages too; strong sounding superlatives, cherry-picked reviews and glowing testimonials. In the pre-AI era that was enough to work.
People were funneled through their various stages and every brand put their best foot forward. It took time and effort to compare reviews, deep-dive threads, mine social comments or see what the key influencers in the space recommended. In the AI era all of that is done for you (whether you ask for it or not) and it’s surfaced.
As a result, being recommended by others is absolutely crucial to being chosen. If you’re visible but not you’re not recommended - you’re just background noise.
Proxies we used to use for organic discovery success like brand mentions, links and coverage in key publications still matter but more impactful and effective recommendation metrics; press endorsements, review sentiment and signal clarity (what attributes and associations are people actually making with our brand) are key in being recommended and chosen.
Where brands go wrong
The typical mistakes we’re seeing involve marketers doing the wrong things, often expensively. Some brands pour serious budget into tracking and measurement tools and assume that visibility follows automatically even though measurement doesn't move the needle on its own.
Others fall back on tactics that were already spammy for traditional SEO and are even more transparent to AI: manipulating rankings, buying placements at scale, engineering their own listicles and putting themselves at the top, or pouring all their effort into their own website while ignoring the other 90% of what actually gets cited.
A useful test is: does the tactic sit at the intersection of three things: genuine user experience, discovery by your audience, and protecting your brand's image? If a change serves all three, it's worth testing. If it only serves one - usually "get found faster," it's usually the kind of shortcut that erodes trust rather than building it.
Measuring what matters
Measurement in AI search is messy, and I'd rather be upfront about that than pretend otherwise. Answers vary between requests, between models, and over time, so a single number is never going to tell the full story.
It’s useful to separate your AI visibility measurement into three types:
- First-party: direct signals like bot hits at page level, and citations per request.
- Proxy: indirect signals like referral traffic and post-conversion surveys.
- Modeled: visibility by model, tracking citations and sources of influence across the ecosystem.
None of these alone will provide a clean, defensible number, but together, they do help join the dots and report progress in a way that's credible.
Dominating answer engines was never going to be about a single tactic or a single platform. It's about engineering discovery deliberately, across all three stages: Found, Understood, Chosen - rather than hoping that ranking well in one place carries you everywhere else.
The brands getting this right aren't the ones spending the most. They're the ones being deliberate about all three, at the same time.
Written by
Related Insights
Is your brand underestimating TikTok?
Is your brand showing up in ChatGPT? Here's why it matters
How to build a discovery strategy that works across every channel
Should you bank on finding new customers on social?
Creative is the new targeting in Paid Social
