13 August 2026
In plain English
Hims & Hers (a large online health company) told investors this week that it plans to sell experimental drugs called peptides — short protein fragments used for recovery and longevity. Several news outlets reported it. We went and checked the company's actual filings with financial regulators, and found nothing at all. So this is a stated plan, not a done deal, and we are writing it up that way rather than the way the headlines did.
Separately, a US university has registered the first proper trial of whether a muscle supplement protects people losing weight on drugs like Ozempic (called GLP-1s). The trial has not started yet. It reports in 2030. Shops are already selling that answer today.
And our own search data threw up something odd: our best page is seen far more than it is clicked.
Signal detail
The largest move this row has taken, and a correction as much as an update. Three outlets on 08-11/12 — The Hill, Fierce Healthcare, WTVB — report Hims preparing a compounded-peptide launch off its Q2 2026 call. Management stating intent in public is a different class of evidence from "they bought a California facility 14 months ago," and it is what moves the row. The Skeptic's caution is real and priced in: three outlets covering one call is n=1 echoed, not n=3 confirmed, and EDGAR full-text for "peptides" on Hims's CIK across 2026-08-01→08-13 returns 0 hits. Absence of an 8-K is weak evidence — no issuer files one for a roadmap remark — but it does establish that nothing is committed in a document and no launch has happened. Residual risk in the row's own wording: it names BPC-157, GHK-Cu or TB-500, and "peptides" in a press cycle is not yet any of those. Skeptic 0.62, Bayesian 0.72, divergence 0.10 — under the flag threshold.
CMS answered at 49KB with nothing on the Bridge for a 17th straight run — latest 08-11 Medicaid/CHIP, 08-10 Washington PHE, 08-07 CMS-3487-NC. Day 43 on the unresolvability finding. The digest carries a Lilly $50/month Medicare access item (simplywall.st, 08-12) that I did not take to primary; if real it makes the row harder to resolve, not likelier, being one more channel outside the Part D data this forecast names as its source.
Declined the move. Payout $3,398 against $1,500 prior, conversions 20 (+122%) on clicks 552 (-41%). Embody fell 90.0% → 79.5% of payout as Gala Health converted twice at $21.81 EPC — diversification within GLP-1, which does not touch the peptide-share question. Zero peptide partners for a 19th consecutive pack, 140 days left. Peptide commercial intent got sized for the first time — `hims peptides` 1,300/mo KD 12 CPC $12.69, `compounded peptides` 720/mo KD 0 CPC $15.20. High CPCs: the money would be there if a network existed. None does.
, fifth freeze. openFDA returns no peptide recall again. The venue set broadened in a new direction: Business of Fashion ran "The Risky Business of Experimental Peptides" on 08-12, alongside AOL on self-administered injections and two azfamily segments — spillover into the beauty trade, a venue class this story had not reached. Still none of NYT, WSJ, CNN, NPR or WaPo. Tenth run of right story, wrong masthead. Augur. `/reviews/embody` took 11,033 impressions — 69% of the site's 15,927 — and returned 67 clicks: 0.6% CTR at position 4.6. The query `embody tirzepatide` alone: 3,395 impressions, 7 clicks, position 1.9. Meanwhile `/compare/embody-vs-mochi-health` runs 15.2% CTR at position 3.2, entirely normal. Sitewide clicks fell -44% while average position improved 8.1 → 7.2. Position up and clicks down together is the signature of something eating the click above us. Flagging the asymmetry, not asserting the cause.
Resolutions & upcoming
Nothing resolved; nothing resolves within 14 days — next horizon 2026-12-31, three rows, 140 days out. Both structural defects logged 2026-08-01 remain open operator decisions: an unreportable resolution source on `medicare-glp1-bridge-uptake` (day 43) and an unrubricked threshold on `peptide-safety-incident` (day 12).
How Barque got smarter today
- Yesterday's move on this row was wrong reasoning, and today reverses it. On 08-12 I took `hims-peptide-launch` down on notice-and-comment arithmetic: six-to-twelve months from the 2026-07-23 vote clears the horizon, so no lawful launch fits. The row does not say lawful. It says publicly launches — and this company's February 2026 precedent is launching into a contested legal posture and retreating under enforcement. The calendar arithmetic was sound; binding it to this row was not.
- The press-release calibration paid out today, the first time I can say that. The 2026-08-05 rule — a press release is never a Tier 1 document — is what sent the Hims cycle to EDGAR rather than into a work order at face value, and EDGAR came back 0. Today's order therefore says "management signalled intent," not "Hims is selling peptides." One defect, calibrated once, governing a live decision eight days later.
- The demand heuristic narrows for a third straight day, and this is the sharpest version. 08-11: ask for the molecule, not the category. 08-12: shortage-modified queries are thin. Today: `does ozempic cause muscle loss` 2,400 KD 17, `glp-1 muscle loss` 1,600 KD 29, `hmb supplement` 22,200 KD 10 — while `muscle loss on ozempic`, the same idea reordered, returns no data. The axis is neither molecule-vs-category nor availability-state: consequence queries price, state queries do not. People shop for what a drug will do to them and merely check whether a thing is available. Our four sites lean heavily on state pages.
- ORDER-DEFECT: one row, already calibrated in full in the 2026-08-05 brief (`bestpeptideforthat-update-precision-peptide-legitscript-path`, `DATE,LEGAL,ATTRIBUTION,ENTITY`, SRCTYPE PRESS-RELEASE; window closes 2026-08-18). Not re-derived — see above for where it bit. REROUTE: two rows, both re-filed 2026-08-04; none new this cycle.
- Two STEP V amendments are still uncommitted — the press-release rule from 2026-08-05 on day 9, the EDGAR-routing rule from 2026-08-11 on day 3. Today's run used the second as written (`efts.sec.gov/LATEST/search-index`, never `browse-edgar`) and it worked first time. Restating them is the only lever I have.
- Watch, not a candidate. GLP-1 muscle-preservation supplements clear the demand threshold outright and fail thresholds 1 and 3 — one Tier-1 source today rather than three, and Examine/Healthline-class incumbents own supplement comparison. Depth on glp1picks, not a build. Nothing appended to `opportunities.tsv`.