Value Validation Project · Creative Strategy
A working brief for the messaging shift from compounded semaglutide to branded, FDA-approved GLP-1 medications.
In March 2026, hims & hers announced a strategic shift in its US weight loss business: a partnership with Novo Nordisk to offer branded, FDA-approved GLP-1s — including the Wegovy pill and pen — and a move away from marketing compounded semaglutide. The company is guiding to 19–28% revenue growth for FY2026 and expects to add 100,000+ new weight-loss subscribers per month within weeks of the branded launch.
That's a real number to grow into, on a real timeline. It's also a messaging problem, not just a supply-chain one: a meaningful share of the brand's weight-loss creative over the past two years has been built around access and affordability — the case for compounded medication over the branded original. That argument is being retired at the exact moment the branded product needs to convert at volume.
This isn't a rebuild. Most of what's working should stay exactly as it is — the shift is narrower than it looks.
Preserve
Close
Built to the same diagnosis-before-hypothesis structure I'd use on day one — filled in for this specific gap.
"Same app. Same 3 minutes. Now it's the actual Wegovy."
Reframes the switch as zero added friction — protects the low-effort value prop while making the upgrade concrete and specific.
"We used to explain why compounded was fine. Now we don't have to."
Names the shift directly instead of pretending it didn't happen — brand-voice-consistent honesty builds more trust than silently swapping the offer.
"FDA-approved. Novo Nordisk-made. Still doesn't feel like a pharmacy."
Carries the two new trust signals (FDA, Novo Nordisk) inside the existing irreverent voice rather than switching into clinical register to deliver them.
Paste directly into Nano Banana 2 or Midjourney. One strong reference style beats five weak ones — generate 2-3 variants per prompt, then pick against whatever's already the top-performing visual register in the library, not against personal preference.
For "Same app. Same 3 minutes."
Editorial product photograph, a smartphone showing a simple telehealth app open on a kitchen counter beside a glass of water and a single medication box, morning window light, warm neutral tones, shallow depth of field, no clinical props, no lab coats, lifestyle-documentary feel, shot on 35mm film, natural color grade. --ar 4:5
For "We used to explain why compounded was fine."
Candid portrait, a person in their thirties laughing mid-sentence, looking slightly off camera, seated in soft window light in a plain unstyled apartment, gentle film grain, warm color grade, direct unposed energy, no medical setting or props visible, feels like a screenshot from a testimonial video, not a stock photo. --ar 4:5
For "FDA-approved. Novo Nordisk-made."
Still life, an official medication pen packaging placed casually beside everyday objects — keys, sunglasses, a half-finished coffee — on a plain wood table, soft daylight from one side, matter-of-fact composition, treated like a normal object, not a pharmacy or lab display, editorial rather than clinical. --ar 4:5
I don't have account access, so this is the read I'd run the day performance data exists — not a claim about current numbers.
Pull the last 60 days of weight-loss creative and tag everything still carrying a compounded/affordability argument — quantify the actual size of the retirement, not just the direction.
Ship the three hooks above as a small test against the current best-performing weight-loss creative — same spend tier, same audience, isolate the reframe as the only variable.
Read cold vs. warm audience split separately from day one — the diagnostic above assumes they'll need different proof, and that should be confirmed or killed fast, not assumed for a full quarter.
Everything above is built from public filings, the public ad library, and open reqs — no internal data. That's the point: this is what I can already do with what's visible. Give me the dashboard and I'll tell you which of these hypotheses survive contact.