Jesus Requena spent more than ten years in a rock band before spending fifteen years marketing software to people with highly tuned nonsense detectors. The careers sound unrelated until you notice the shared predicament: talented individuals, complicated gear, strong opinions, and an audience that leaves when the performance feels false. At Sanity, where he is chief marketing officer, Requena has taken that lesson into a web increasingly mediated by machines. His wager is simple enough to make a century of search advice feel suddenly antique. If an AI is doing the reading, a brand must be clear enough for the machine to reason about it.
This is a natural destination for a marketer who calls his method “anti-marketing.” Requena does not mean silence, or an allergy to selling. He means dropping the upholstered language that accumulates around technology products. Developers, he has observed, dislike being marketed to. They want to see how something works, test the claim, and decide without a trumpet voluntary playing over the documentation.
“Ask yourself why three times.”Jesus Requena on reaching the plain meaning beneath campaign language
His instruction to writers is almost childlike: ask why, then ask why again, until the idea underneath the idea can be written in ordinary words. It is a useful piece of craft and a revealing piece of temperament. Requena’s public conversations return to the same conviction: marketing works when it removes friction between what a product does and what a person needs to understand. A flourish may attract the eye. A clear claim survives the meeting.
The free user is not a rounding error
Requena’s technology career has moved through products that people can begin using before a procurement committee appears. At Unity in 2016, the scale of the opportunity was already visible: close to one million active users a month, many on personal plans. His job was to convert more of that activity into paid subscriptions. The challenge was less like filling a funnel and more like listening to a city. Millions of actions created noise. Somewhere inside it were signals of intent.
The team built a product analytics stack to learn which behaviors correlated with an upgrade. Rather than tossing every signup toward sales, they looked for a small set of actions that indicated an account was approaching value, pressure, or both. In later work, Requena described a model with roughly ten data points. When two or more appeared, the likelihood of an upgrade rose enough to make the account interesting. Product data became a way to improve timing, not an excuse to automate impatience.
One experiment opened live chat on high-intent pages. Prospects who had once faced an email form could have a real-time conversation. The channel quickly generated revenue and later appeared on the cancellation page, where a timely exchange helped save some subscriptions. There is a miniature philosophy in that placement. Conversation should arrive at the moment of uncertainty, not chase everyone around the internet with the digital equivalent of a mall kiosk.
The product-led signal path
That practice followed him from Unity to Algolia, then Figma. Each company had a different audience and market, so the technique had to bend. Figma’s growth already owed much to direct traffic, word of mouth, and community. Its product moved through organizations because designers used it together. Marketing’s task was not to manufacture affection in a spreadsheet. It was to help existing affection travel from an individual to a team, and from a team to an enterprise agreement.
He has been unusually candid about the point where easy slogans stop. Product-led growth is not merely offering a free plan. It requires a product that can reveal value without a guided tour, an addressable market large enough to support the motion, and a company willing to measure the right behavior. At Unity and Figma, which he joined around the time each was crossing roughly the hundred-million-dollar revenue neighborhood, the next investment was depth: richer experiences, sharper segmentation, and sales work grounded in what users had actually done.
Every company changed the arrangement
The itinerary covers game developers, search APIs, designers, data practitioners, and content teams. Requena’s rule is to follow the audience into its natural habitat. Designers gather in visually fluent social spaces. Developers spend time on YouTube, Reddit, and Discord. The channel is a cultural choice before it becomes a media-plan row. “You need to market where your audience is,” he says, a sentence so plain that an industry has managed to ignore it in several thousand elaborate ways.
At Hex, he returned to a smaller and more specialized market. The motion could not simply be copied from Figma. Requena talks about bringing a user to a pain state: the moment when the free or individual experience has proven useful, but a limitation now matters enough to support a commercial conversation. That requires attention to the product, the account, and the buyer’s timing. It also requires the discipline to abandon a hypothesis. His advice is blunt: stop what is not working and double down on what is.
How the central growth question evolved
From detecting purchase intent, to scaling community-assisted adoption, to making a company legible to human and machine readers. Bar length represents breadth of audience surfaces, not financial performance.
The reader who never clicks
Sanity puts Requena inside the product category his work has always depended on: content infrastructure. He joined first as an advisor, then as a senior marketing leader, and became CMO. In May 2025, the company unveiled a new brand, website, and positioning as a Content Operating System while announcing an $85 million Series C. Requena celebrated it as the largest release in Sanity’s history. The work tied an ambitious phrase to a technical argument: content treated as structured data can be reused by websites, applications, workflows, and AI agents.
By 2026, he was describing a more provocative marketing consequence. Sanity had stopped producing conventional SEO content and begun optimizing parts of its information architecture for large language models. The program started with an audit of roughly 100 to 200 important pages, hunting for contradictory terms and concepts. If a company cannot describe itself consistently, a model has poor material from which to reason. Poor reasoning means weak or absent recommendations.
The next layers were stranger. The team converted a large collection of frequently asked questions into answers intended to be easier for machines to parse. It built tightly focused microsites and examined outdated third-party descriptions that could muddy the story. Requena reported that Sanity moved from the middle of the pack in AI visibility into the top five, and that self-reported discovery showed a large share of recent signups coming through language models.
A website once waited for a visitor. Now its argument may travel without the page.The strategic turn in Requena’s work at Sanity
The claim deserves care. Attribution is especially slippery when someone discovers a tool in a chat, installs it through a command line, or connects through an MCP server without a trackable web session. Requena’s team fell back on the decidedly unglamorous question, “Where did you find us?” The answer is imperfect but revealing. Marketing, after two decades of increasingly intricate surveillance, has rediscovered asking.
This shift does not erase humans from the audience. It changes the route by which information reaches them. A model may assemble the shortlist, but a person still lives with the decision. Requena’s anti-marketing principle becomes more demanding in that environment. Claims need enough structure for retrieval, enough consistency for reasoning, and enough substance to withstand inspection after the answer arrives.
The drummer still gets a vote
Long before growth became his specialty, Requena toured as a professional musician. In a 2018 essay, he mapped band life onto management. A team is a second family. Every person needs room. Trust and communication are prerequisites. Patience matters. Recording an album taught him that the finished mix cannot lean entirely toward one member’s taste. His summary was compact: “The group before the individual.”
It is tempting to treat the band as colorful prologue, but the metaphor explains the operator. Product-led growth asks marketing, sales, data, product, and community to hear one another. AI discovery asks brand, documentation, third-party commentary, and structured content to stay in tune. The work fails when one section plays a different song and calls the resulting noise omnichannel.
Requena is still drawn to the commercial result. Colleagues describe a marketer who moves comfortably between customer insight, analytics, creative work, and revenue. He advises companies and investment portfolios alongside his operating roles, and remains on Green Hat’s board. Yet his public advice keeps returning to ensemble virtues: listen actively, make context visible, give each contributor a useful channel, and persist long enough for the group to improve.
In September 2026, he is scheduled to sit down at Sanity’s Everything NYC event with journalist Gina Chua for a conversation titled “Information is liquid now.” The phrase fits his current chapter. Information no longer stays politely inside the page where a company published it. Agents extract it, recombine it, and place it wherever an answer is requested. The marketer can complain about the acoustics or improve the composition.
Requena has chosen the latter. His career began with the practical problem of helping free users become paying customers. It has arrived at a larger version of the same puzzle: how does useful information become a trusted decision? The instruments changed. The audience multiplied. The standard remains recognizable from any small club with a skeptical crowd. Be clear. Be worth hearing. Know when to stop playing.