How Stella Rising built a Generative Engine Optimization practice on OtterlyAI data, and learned to measure visibility in a world where the click is no longer the point.
A conversation with John Morabito, SVP of Search & Innovation at Stella Rising.
When John Morabito talks about the future of search, he does it the way his colleagues describe him: passionately. “They often use the word passionate,” he says, “but what they probably mean is intense.”
That intensity has powered a twelve-year run at Stella Rising, where he built the agency’s SEO team and now leads its fast-growing Generative Engine Optimization practice. The work: getting brands seen, cited, and recommended inside AI answers from ChatGPT, Google AI Overviews, Perplexity, Gemini, and beyond.
For a marketing leader operating at the frontier of AI search, John’s route in was anything but linear. And that story turns out to say something important about how the best search practitioners think.
A photographer who learned to rank
John “follows the traditional canon of an SEO who stumbled into this via some other general liberal-arts thing.” He went to school for fine-art photography and came out with, by his own account, no obviously marketable skills beyond taking pictures and playing music.
His first real break came from an unlikely Craigslist ad: a Photoshop job retouching before-and-after photos for a dental marketing consultant named Tom Orent. The retouching was the job on paper. The real education happened behind the camera, filming a $50,000-a-year consulting program on direct-response marketing, the psychology of influence, and how to sell. John calls it “a free master’s education in direct-response marketing.”
The SEO itself arrived at a New York camera shop, right as the iPhone was hollowing out the camera business and Beats by Dre was about to take off. His boss asked whether he could build a website for a rack of headphones. A week later (“some sort of Old Testament story about seven days later”) came the follow-up: could he get people to actually come to it?
“Somebody taught me about SEO,” John remembers thinking. “I think I can do that.” So he did. He ranked the store, named it, designed the logo in fifteen minutes (“before AI slop, I was doing logo slop”), and, almost by accident, invented his own version of influencer marketing in 2012. He spotted kids reviewing headphones on YouTube and mailed them pairs with return labels in exchange for reviews. “I was doing influencer marketing in ’12, before people were talking about it, and I was getting it for free.”
The through-line matters, because it is exactly how he now approaches AI search: notice where attention is actually moving, then figure out the play, long before there is an accepted name for it.
Inside Stella Rising
Today John’s days are spent shaping Stella Rising’s product offering and forging partnerships. The agency has been deliberately productizing its SEO work into dedicated work streams: social search, GEO, authority building, and outreach, with a Reddit offering and a set of internal tools on the way.
That structure is a direct response to a changing landscape.
“As much as search is everywhere,” John says, “I don’t think you want to measure it in one big bucket.”
Rather than treating every new surface as generic “growth” bolted onto the SEO team, Stella Rising staffs dedicated specialists against each one. They set the standards of practice, define the deliverables, and make sure the work lands beautifully with clients. John also leads paid search, where much of his energy goes toward making SEO and PPC work together to cut wasted and cannibalistic ad spend inside one holistic strategy.
The client roster reflects a deep specialization in beauty and wellness. Stella Rising works with some of the world’s biggest beauty brands, including major partners like Henkel and L’Oréal, alongside consumer, health, and wellness names. That focus compounds. “We attract a lot of people on the SEO team who want to work on these beauty clients,” John notes. “They are stoked to come to work every day and work on brands they are actually interested in, doing cutting-edge work.”
“We really try to never come back to a client and say, ‘Sorry, we didn’t scope for that.'”
Ask how clients describe the partnership and one word comes back first: flexible. “Our commitment to client results and profitable growth as the North Star can lead us to needing to flex,” he says. Pair that with genuine category expertise, and you get the two things clients name most: flexibility and a relentless commitment to results.
When GEO stopped being a buzzword
John is refreshingly honest about his own arc with AI. He was “instantly enamored” with ChatGPT, using it every day, but he still caught himself doing the “ostrich thing” many practitioners did in 2023: this is cool, but things are not going to change that much.
What changed his mind was doing the work. “To truly understand the nature of the unique work and defend it as its own practice, you need to really understand the work at hand,” he says. His advice to any skeptic is a simple exercise:
“Take a bunch of prompts you care about for your client, run them, and then look at the sources of citation and where they are coming from. Then think about your play. What are you going to do? It is different than what we have been doing.”
Stella Rising did not wait for the industry to catch up. “I don’t think it took us that long when everyone was saying, ‘Where’s all our traffic going? ChatGPT,'” John says. “We were getting ready far before that.” When they went looking for tools to track AI visibility and could not find anything mature, they started building their own, and jumped on OtterlyAI early, while most vendors were still announcing.
On the client side, the conversation has evolved fast. The “what is this and should we care?” moment has passed; executives are convinced, often by “the guy on LinkedIn they have been following,” and some even ask whether they should abandon their SEO budgets. John’s job is to navigate that. The happy surprise: “I have never seen such an interest in SEO by executives asking questions about GEO.” Pressure from the top, essentially “what are we doing about AI visibility,” has renewed appetite for core SEO, content, and brand work all at once.
The playbook: PDPs, content, and the rise of social search
So what does a typical GEO play actually look like? John breaks it into pockets.
For bottom-of-funnel, transactional intent, the product detail page is now center stage. “AI systems are aiming to become the category page, the aggregator and curator of the PDP,” he explains. That flips a career’s worth of instinct. For most of his time in SEO, the category page was the prize and PDPs were an afterthought. Now the optimization work moves to the product page itself: how-to-use, ingredients, “nevers,” and structured question-and-answer content. Crucially, it is not about schema tricks. “It is not the schema that is driving it. It is the presence of the structured question-and-answer content on the page.” The tactics rhyme with classic SEO; the why maps back to specific prompts where the product should surface as the solution.
Upstream, informational content gets a renewed mandate. “‘So you mean to tell me we should be putting blogs on our website that answer questions users have?’ Yes, that is exactly what I mean. Let’s do it.” There is a new reason for a very old best practice.
But the biggest shift is social search. AI engines lean heavily on social platforms – Reddit threads, YouTube videos, creator content – when assembling their answers, which turns social presence into a direct visibility lever. Stella Rising has spent years folding social strategy into its search plans; what used to require convincing is now self-evident. AI search didn’t create the case for social – it just made it impossible to ignore.
What 4,500 prompts and 1.8 million citations revealed
This is where John gets specific, using, as he puts it, “some fun numbers.” Stella Rising ran an analysis on OtterlyAI data across a portfolio of brands for one of the world’s biggest beauty conglomerates, tracking roughly 4,500 prompts through the OtterlyAI API.
The first question was blunt: can we actually influence citations with social content? The answer was yes.
The study, by the numbers:
- Roughly 1.8 million total citations surfaced by OtterlyAI across the tracked prompts.
- About 279,000 of those citations came from social platforms.
- Around 5% of the social citations belonged to the brand itself.
- The brand’s videos appeared in thousands of individual AI answers.
The pattern under the numbers is intuitive but underexploited. “If you want to go after something like how to do your eyebrows, and you produce a really well-done video about how to do your eyebrows, you are going to end up showing up for it.” For beauty brands, John says, social typically accounts for anywhere from 7 to 20 percent of all citations.
Reddit is hard. YouTube and TikTok want you there.
Which channels? It depends on the engine, and that is the point. “Reddit is the dominant source of social citation within ChatGPT,” he says, “but ChatGPT is not the only game in town.” Across Perplexity, Google AI Mode, Google AI Overviews, and Gemini, the mix is far more homogenous, with YouTube and TikTok showing up in the 20 to 40 percent range.
And unlike Reddit, where, as a self-described Redditor, John is candid that “they firmly just do not want brands there,” YouTube and TikTok actively welcome brand content. “If you are doing a good job, the ecosystem will reward you.”
“You are asking me what the Reddit play is. I will ask you what your YouTube play is, because that is where you probably ought to be from a citation perspective.”
The bigger thesis: brands win by building a hybridized organic social approach that blends viral, of-the-moment brand content with evergreen, genuinely educational material. The job is to bring audiences your expertise, exactly as brands were long told to do on the blog.
Measuring what matters when the clicks dry up
GEO forces search marketers to borrow from the brand marketer’s playbook, and that is uncomfortable for a discipline raised on direct response. “Nobody loves how difficult it is to measure,” John admits.
He still insists on rigor. Referral traffic and any revenue it drives from AI remain worth tracking, but he treats them as correlative rather than the headline metric. More important is a question clients keep asking: not just how often do I show up, but how often am I the answer, the one being recommended? That makes share of voice central.
Stella Rising built internal tooling to answer it. One tool, SearchTrak, unifies share of voice across all of search, paid, organic, and AI, pulling OtterlyAI data in alongside everything else. Another, Brand Track, looks at the correlation between social views on YouTube and TikTok and branded search. His logic is that visibility should ultimately show up as demand. “If the brand awareness is there, it ought to manifest at some point in increases in brand search.” Like a brand marketer, he is learning to live with correlative measures and lift-style thinking instead of pure last-click attribution.
Why OtterlyAI
John is discerning about the tools his team invests in, and early in the market, plenty of options did not inspire confidence. Some vendors, he says, felt almost “predatorial,” quoting tens of thousands of dollars per client and “rushing into California to make money on the gold rush.” What he was looking for was not the cheapest option, but the one that paired serious capability with fair, transparent terms.
OtterlyAI stood out for exactly that.
“These guys are chill. They are search marketers. They come from the space.”
He points to transparent, competitive pricing and flexible commitments. “OtterlyAI fit the bill. But then when I got into the tech, it is as good as the other stuff people want to spend a lot more money for.”
The API is what turned a good tool into a strategic one.
“The API has been a really transformative tool for us.”
It is what powers the big-data-style analyses like the 1.8-million-citation study, work that would otherwise mean “going into the tool for fifty million things.” He also credits the pace of the product: “The continued evolution of the features is great. They are always releasing new stuff.”
Reporting, reinvented
That API sparked what John calls his biggest aha moment, one that goes beyond OtterlyAI entirely. Tasked with building a quarterly competitive-landscape dashboard, he had been manually exporting data when he asked whether OtterlyAI happened to have an API. It had just launched. Piping the data straight into a live dashboard reframed how he thinks reporting should work. “Instead of a PowerPoint or a Word doc, what if reporting is a live web experience, completely customized to that client, their branding, the questions they have, with Claude Code as your UI control?”
“I never want to go into Data Studio again.”
For a practitioner who now believes static dashboards are on their way out, it is a fitting punctuation mark: the reporting itself becomes as dynamic as the search landscape it describes.
Closing the gap on business impact
What keeps John up at night, sometimes literally, is closing the gap on business impact. Recent studies suggest around 20 percent of beauty purchases are now influenced by AI, yet marketers still lean on last-click attribution and simply are not asking buyers what led them in. He is chasing something like a post-purchase survey that asks customers to recall the prompt that sent them, tied to revenue on the page. “Figuring out complex problems, and then decent solutions for how we look at them, that is the stuff that has me excited.”
For a practitioner who taught himself SEO from a headphone rack and spotted influencer marketing half a decade early, that instinct for the next unnamed thing is the whole career in miniature. AI search is simply the latest surface. The play, as ever, is to understand the work deeply enough to defend it, and to bring the receipts.
See where your brand stands in AI search. John and the Stella Rising team run free audits that show brands exactly where the opportunity is inside ChatGPT, Gemini, and Google’s AI surfaces. Stella Rising also built this cool tool called TokTrak for TikTok Video Analytics. And if you want to track it yourself, OtterlyAI’s data is what powers the analysis.
John Morabito is SVP of Search & Innovation at Stella Rising. Find him on LinkedIn, on X at @JohnMorabitoSEO, or at jmorabito@stellarising.com.
