Later’s New Data on How Creators Help Brands Earn Citations in AI Search
Later gave me early access to their new Creator Answer Engine Optimization (AEO) playbook, which includes insights from the almost 500 marketers they surveyed as well as campaign numbers that paint a clear picture on how creators help brands influence AI search visibility.
The numbers below are theirs, but the opinions are mine, and I picked the findings I thought would actually change how you work. This builds nicely on past editions of this newsletter where I’ve made a case for Creator AEO; now we have even more data to further validate this approach!
Let’s do a quick refresher to set the stage before we dive in: McKinsey reported earlier this year that less than 10% of what AI says about your brand comes from content you own (think: what you publish on your company blog, put on the branded LinkedIn page, etc.)
Now pair that with Later's survey of 487 marketers, which found that 1 in 3 believe their brand website carries the most weight in shaping how AI describes them.
There’s a real disconnect happening, and it makes sense: This is an entirely new category of data. We’re all still learning and reverse-engineering what actually gets a brand cited in an answer by an AI tool like ChatGPT or in Google AI Overviews. However, one thing is clear: The industry is largely spending its AI search budget on a surface that matters the least: the company blog.
What is Creator AEO?
Creator AEO is the practice of earning brand citations in AI answers through third-party voices, creators, community members, reviewers, and subject matter experts, rather than through content you publish on your own website.
AEO has "optimization" in the name, so, logically, it has mostly been handed to Search Engine Optimization (SEO) teams, agencies, and practitioners.
The problem here is that SEO teams are used to working on content that’s published on the company blog…and as we just learned, much of Answer Engine Optimization relies on work that happens beyond the company blog.
Here’s the idea I’ve been writing about for the last six months: Boosting AI visibility isn’t just about brand-approved content published on the company website. It requires third-party, external validation from trusted, non-brand-controlled voices.
And that’s where creators come in.
What the Data Says About Creator AEO
The data (thus far) indicates that the majority of what AI knows about your brand was written by someone who doesn't work for you. Later's campaign findings put a number on the creator slice specifically: As many as 1 in 3 AI answers cite creator content.
Asked what worries them about AI search, 85% of marketers said they worry about AI getting their facts wrong and 82% worry about losing narrative control, while only 61% worry about losing to competitors.
Later’s playbook has a story that makes this concrete faster than any numbers I could share with you:
Carly Baron, Chief Commercial Officer at Advantice Health, described searching anything related to one of their brands and watching the same dermatologist surface again and again. Different queries, same person. They had never worked with her. No contract, no brief, no relationship at all. Her narrative was shaping what their brand looked like in AI search, and they found out by going and looking.
That situation is not unusual. Every brand in every category has some version of that dermatologist being the named expert associated with their topic, service/product, and target audience.
The question is: Have you found out who that person is? And if so, why aren’t you working with him or her?
You don't get to opt out of third-party voices defining your brand. You only get to decide whether you have a relationship with them.
Nobody Owns Creator AEO Yet
Later's survey found that 25% of marketers can’t name a clear owner for AEO at their company. The rest splits across a few different functions. That may work for now, but there needs to be a larger strategy in place.
“We've entered a new era where SEO teams and creator teams are now working together. These are two ends of marketing that never had to collaborate before. What we love to do is partner with the SEO team leading on-site content creation, so the creator content ladders into what the brand is already saying on its owned channels.” - Cameron Miskho, Sr. Director of Strategic Consulting at Later
The fix is getting all your internal teams to all look at the same thing: Figuring out how to boost brand visibility in AI answers.
So how do you do that? Here’s what I’d recommend: Figure out which engine you want to focus on (whether that's ChatGPT, Google AI Overviews, etc.) Pick one, as each engine has its own set of quirks and preferences.
From there, build a working list of 10 to 20 prompts across four categories:
Buying intent. Ex: "What's the best project management tool for a 12-person agency?"
Competitor comparisons. Ex: "Asana vs Monday for creative teams"
Features and specs. Ex: "Does Figma have offline editing?"
Broad category. Ex: "Top skincare brands for sensitive skin right now"
Then split the work according to what each team is already good at:
SEO decides which prompts matter, because they can see what language/wording is driving new customers.
Creator teams decide which creators/influencers to hire, because they know how to structure those deals and have existing relationships.
If you feel stuck at this point: that’s fairly common.
Building these lists requires knowing what your customers are actually asking AI and which creators to hire based on who already owns that topic. Later now does both.
Your Creator Roster Is Sorted by the Wrong Number
Later’s campaign data shows that 57% of the YouTube videos AI cites come from channels with fewer than 100,000 followers. This is surprising, because it runs counter to the sorting logic almost every creator program runs on.
The roster you built for reach is not the roster that gets you cited.
Reach and citation are different outcomes with different inputs. Models weigh authority, specificity, and crawlability. None of those correlate cleanly with audience size. Most brands have only ever optimized for reach and impressions, because those numbers are good for reporting on the brand visibility front.
Marketers already sense this. Asked which creator types matter most for AI search, they ranked everyday product users first at 39%, then industry specialists and product reviewers. High-reach social influencers came in last. The instinct is right; the rosters haven't caught up to it.
Later’s playbook sorts creators three ways, and I've started using this framing with clients because it’s really easy to understand:
Attention. Entertainment, short-form explainers, infotainment.
Authority. Thought leadership, expert POV posts, long-form articles and guides.
Answers. Technical depth, reviews and comparisons, community Q&A, real-world use cases.
Look at your current creator roster and sort it into those three buckets. If you're like most brands, it's stacked with Attention and nearly empty on Answers.
Answers is the category AI models reach for.
This matches what I found reporting on employee-creators earlier this year. Follower count barely predicted citation, but demonstrated expertise and specificity did. The subject matter expert with 4,000 followers and a detailed, data-backed opinion outperformed the generalist with 400,000 followers and no real expertise on a topic.
Picking for citation likelihood isn't a taste call, though, and this is where I think most teams will need help.
Later's EdgeAI identifies which creators, platforms, and formats are most likely to get picked up across a 20M+ creator network on YouTube, Reddit, SubStack, Instagram, and more.
These are places AI actually goes looking for answers.
The Content You Already Paid For May Be Invisible
This topic is touchy for people in social and creator industries, because it comes with a painful reality: Instagram Reels and TikToks frequently sit behind walls that block AI crawlers.
A video on a platform like Instagram can do enormous numbers, drive real sales…and still never reach the models forming answers about your brand. It has its place and purpose! It’s just not yet in a format where AI models can access it, too.
The retrofit to fix this is three steps:
Pull the video’s transcript.
Upload it to YouTube as a .srt file.
Build a landing page that links back to the original post, so the content is discoverable on both ends.
Retrofitting works, but the smarter move is briefing creators for this added use case from the start moving forward. When transcription, captioning, and cross-platform distribution are in the brief instead of in a repurposing plan, you stop paying twice for the same visibility and can move faster at accomplishing both the AI visibility and social reach initiatives.
The report's short-form video checklist is worth stealing, too: put the question and the answer into your on-screen text, upload real captions instead of auto-generated ones, and write titles and descriptions around the exact prompt the video answers rather than a clever hook.
Which Metrics to Measure for Creator AEO
Alignment falls apart when two teams track two different scoreboards, so Later proposes one: Share of Model, or how often and how accurately your brand shows up across your highest-priority prompts.
It has three layers: how often you appear at all, how accurately you're described, and whether the mention comes with an actual citation or is just a passing reference. A mention without a link is awareness, but a mention with a citation is traffic.
Later's team has found a healthy Share of Model starts around 30% for top-priority prompts. Treat that as a working benchmark rather than gospel, since the category is young, but it beats measuring against nothing.
Where to Start with Creator AEO
Here are the three steps Later recommends taking to get started with Creator AEO.
Align your SEO and creator teams around one shared prompt list.
Rebuild the creator mix around authority and specificity instead of reach.
Activate and measure monthly, so citations compound instead of resetting every campaign.
Later's full Creator AEO playbook has the survey data, the frameworks, and a monthly tracking checklist I'd have charged money for. It's their research and it's free to read.
