The most revealing number in the recent history of RevHealth is not seven minutes, though seven minutes makes the better press release. It is negative eight percent. In 2023, after years of expansion and a private-equity recapitalization, the Morristown healthcare agency's revenue slipped from $52 million to $48 million. The line went down. This is useful because corporate reinventions are easier to understand when they begin with an inconvenience.
RevHealth had been founded in 2006 by Bruce Epstein, Bruce Medd and Brian Wheeler - two Bruces and a Brian, a lineup with the tidy rhythm of a neighborhood law firm. They built a science-focused agency for pharmaceutical, biotech and medical-device companies. The work lives in the difficult stretch between a molecule and a person: explain the evidence to a physician, make the value legible to a payer, prepare a market for a new category, support a patient, and do all of it without wandering outside the guardrails of medical accuracy and regulatory review.
By 2022, RevHealth had reached $52 million in annual revenue. That July, WindRose Health Investors recapitalized the company. The amount was not disclosed. The founders remained shareholders and initially stayed in charge. Then came the awkward middle: what new CEO Ben Beckley later described, with some economy, as figuring out the ship once private equity had the wheel. By 2024, the founders were no longer involved day to day, though they remained on the board. Beckley arrived to rebuild.
The agency is really a translation machine
Advertising is the convenient label. Translation is closer to the job. RevHealth now organizes its offering around four connected practices: medical communications, market access, corporate communications and brand marketing. The customer is usually a life-sciences company moving from evidence to launch and then through a therapy's commercial life. Public work has touched Daiichi Sankyo; older trade coverage named Novartis and Merck. Other releases have referenced US WorldMeds and Recordati Rare Diseases.
Each practice translates for a different listener. Medical communications turns a dense body of science into education that clinicians can trust. Access teams explain clinical and economic value to payers and plan the support that helps a patient actually receive a therapy. Corporate and advocacy work assembles stakeholders around a shared cause. Brand marketing gives the whole argument a shape people can remember. Large agency networks can offer all four. RevHealth's pitch is that an independent shop can join them without the handoffs, turf and billing seams of a conglomerate.
One therapy, four translations
The case studies make the integration less abstract. For an emerging oncology biotech, RevHealth aligned payer, trade and patient-support planning around a treatment for a narrowly defined breast-cancer population. Its digital work recorded click-through rates 422 percent above control baselines. For a rare-disease company coping with shifting FDA label language, the agency prepared multiple content paths, centralized references and coordinated reviewers. A peer-to-peer broadcast went live three days after approval.
“Process should accelerate great thinking, not slow it down.”One of RevHealth's published operating principles
There is also the work that looks less like advertising. RevHealth helped seven national organizations in the chronic graft-versus-host disease community agree on a shared mission, identity, website and awareness day. Patients and caregivers gained one consistent resource. The operating model was designed to move toward independent governance. It is a helpful clue to where the agency fits: not simply making media, but arranging the coalition that makes a message credible.
Seven minutes is a provocation, not the product
In July 2024, RevHealth announced AgencyOS, built with PixieDust Labs on a tailored version of its multi-agent AI system. The headline claim was irresistible: banner ads that took two weeks could be produced in seven minutes. The system was trained around a client's requirements, brand and products. Its jobs included first briefs, timelines, draft manuscripts and references, claims libraries, staffing maps, email and banner prototypes, search and social analytics, and medical summaries.
That list matters more than the science-fiction name. These are the waiting rooms of agency life. A strategist waits for a brief. A writer waits for references. A project manager rebuilds a timeline. A reviewer finds an old claim in a new deck. Automation does not need to invent the campaign to change the economics. It only needs to return those hours.
The company's current language is subtly different from its 2024 launch. RevHealth now says it does not lock itself into one model, chooses the best available tools and does not treat AI as a product to sell. More than 90 percent of its people have adopted AI, it says, and over half use it daily. Kickoff time has fallen by half; editorial and quality-assurance processes have compressed by more than 40 percent.
This looks like a change of mind, or at least a maturing one. The named platform was a useful door into the conversation. The durable idea is tool-agnostic workflow design. The thing clients buy is still judgment under constraint. In pharmaceuticals, a fluent error is not a charming first draft. Medical accuracy, fair balance and regulatory review remain human responsibilities. RevHealth uses software to clear judgment's calendar, not to impersonate judgment.
What another team can copy
The portable lesson is not to buy the same platform. It is to begin with friction small enough to measure. Pick the brief that consumes a morning, the reference check that delays every round, or the kickoff that happens twice. Establish the old time. Put a human owner on medical or legal accuracy. Then automate the draft, retrieval or assembly around that owner. Seven minutes is impressive only because two weeks came first.
A practical copybook
- Choose a repeated bottleneck, not a futuristic demo.
- Record the baseline in hours, revision rounds and errors.
- Keep one accountable human at every regulated decision.
- Connect functions before execution, while the story can still change.
- Stay tool-agnostic so the workflow survives the model cycle.
It will not travel cleanly everywhere. The model depends on repeated workflows, usable source material, reviewers who agree on the rules and enough volume to make redesign worthwhile. A tiny team producing one unusual campaign may spend more time configuring the machine than using it. A client with fragmented data, unclear claims or no empowered reviewer will merely automate confusion. Integration also fails when medical, access and marketing leaders attend the same meeting but protect separate answers.
RevHealth still competes in a market crowded with giant networks, specialist medical shops and well-funded independents. Its independence is also qualified by private-equity ownership. The agency's advantage will not come from the word “independent” or from possessing an AI logo. It will come from whether those four translations - evidence, value, story and access - genuinely become one argument faster than competitors can assemble theirs.
The revenue dip makes the story believable. A company missed the easy upward line, changed leaders and reconsidered how the work moved through the building. It found that the boring parts were not beside the business. They were the business's hidden timetable. Making them faster gave people more room for the part no model can safely own: deciding what the science should mean to another human being.