Digital healthcare advertising is running at a projected $26.2 billion against $6.9 billion in traditional spend. That’s not a shift in the marketing mix — that’s a different industry wearing the old one’s clothes. And the thing that broke the dam wasn’t a technology breakthrough. It was the FDA.

Regulation as an accelerant

The conventional story about advertising moving online is about targeting and measurement. In pharma, there’s a more specific driver: intensified FDA scrutiny of television advertising. Pharmaceutical TV ads occupy a peculiar regulatory position — the long recitation of side effects at the end exists because of disclosure rules — and when the regulator leans harder on that format, the economics of the format degrade.

The result is that in 2025, digital channels outpaced linear television in this sector for the first time, and social media has now passed linear TV outright. Regulatory pressure on one channel functioned as a subsidy to another.

There’s an irony worth naming. Advertisers are moving from a channel where regulators have decades of established practice and clear disclosure requirements into channels where the rules are far less settled. Whether that constitutes an improvement in how drug advertising reaches the public is a genuinely open question, and one that regulators appear to be behind on.

The privacy squeeze is creating the specialists

The second force is state-level privacy regulation, which is emerging piecemeal and demands more adaptable, privacy-forward targeting. This is the part that explains why the money is flowing to specialists rather than to the giants.

Health data is the most regulated category of personal data there is. A general-purpose demand-side platform built for retail and consumer packaged goods was architected for a world where you can target based on behaviour with relatively few constraints. Health advertising cannot work that way — you cannot casually infer and act on someone’s medical condition, and the compliance cost of getting it wrong is severe.

That constraint is a moat. Specialised healthcare ad-tech firms — DeepIntent, PatientPoint, PulsePoint, alongside agencies like Branchlab — are taking share from legacy demand- and supply-side platforms precisely because compliance-heavy targeting is genuinely hard to retrofit. The regulatory complexity that makes the category difficult is the same thing that protects the companies who have solved it.

The capital agrees. DeepIntent raised $637 million from Vitruvian Partners in late 2025 — a number that only makes sense against a $26.2 billion addressable pool and a belief that specialists, not generalists, will capture it.

Agentic AI arrives in ad tech

DeepIntent has launched Helix AI, described as an agentic platform allowing natural-language analysis of patient and provider data, audience building, and multi-channel activation. Its CEO says the company is “growing multiple times higher than what the competition is,” with new dollars flowing disproportionately to its platform.

Treat the growth claim as what it is — an unaudited statement from an interested party. The product direction, though, is worth taking seriously, because it’s the same pattern showing up across enterprise software: the interface to a complex data system becomes conversation rather than configuration.

The pitch is that a marketer describes an audience in plain language instead of assembling it through query builders and segment tools. In a domain where the underlying data is dense, the rules are intricate, and the specialists who can operate the tooling are scarce, natural-language operation removes a real bottleneck.

It also concentrates responsibility in an interesting place. If an agentic system builds a healthcare audience from a natural-language instruction, the compliance logic has to live inside the system rather than in the head of the operator who used to assemble the segment by hand. That’s an argument for the specialists — they have spent years encoding those rules — and it’s also a new category of risk. An agent that quietly constructs a non-compliant audience is harder to catch than an analyst who does.

What this actually is

Strip away the sector specifics and the pattern is familiar: a regulated industry’s advertising spend migrating to digital roughly a decade after everyone else’s, with the delay caused by compliance friction, and the migration finally triggered by regulatory pressure on the incumbent channel. Retail did this. Finance did this. Pharma is doing it now, at scale, with $26.2 billion in play.

The durable insight for anyone building in a regulated vertical is that the compliance burden is not purely a cost. It is also the reason a specialist can beat a far larger generalist. DeepIntent is not out-engineering the major ad platforms on raw capability. It is operating in a space those platforms find expensive and risky to enter, and it is being paid for that willingness.

The open question is the one nobody in the supply chain is incentivised to raise: pharmaceutical advertising moved online partly to escape scrutiny it had grown used to, and the scrutiny hasn’t followed yet. It usually does.

Reporting on industry spending projections and company announcements as covered on 22 July 2026. Growth and market-size figures are estimates from industry sources and interested parties. Not investment advice.