The Pulse
The Fifth Circuit closed the shortage-era door on compounded GLP-1s
On August 27 the Fifth Circuit upheld the FDA's removal of semaglutide and tirzepatide from the drug shortage list, in two unpublished opinions.
That was the legal basis compounders and telehealth platforms used to mass-produce copies of Wegovy and Zepbound. What remains is patient-specific 503A compounding for an identified patient with a documented clinical need.
One detail from the tirzepatide opinion matters for operators: the FDA discounted access complaints collected through a Hims & Hers website because they could not show when, where, or why a patient failed to get the drug. The court agreed.
If your program routes most patients to the same compounded product through the same funnel, that is the pattern law firms are now flagging. Ask counsel what "patient-specific" looks like in your flows, not only in your pharmacy contract.
Zepbound goes back on CVS Caremark's formularies October 1
CVS Caremark is adding Zepbound back to its commercial template formularies on October 1 as a co-preferred option next to Wegovy, for plan sponsors that choose to cover weight management drugs.
Caremark dropped Zepbound in July 2025. Some patients who wanted tirzepatide anyway ended up paying cash through a telehealth brand.
From October 1, a share of them can get the same molecule for a copay through their own doctor. That is the deep dive.
Sword confirmed the Headspace deal at a reported tenth of its last valuation
Sword Health confirmed on September 16 that it will acquire Headspace, targeting a Q4 close. No price was disclosed; Fierce, Axios and Bloomberg have reported $200M to $300M in cash, against the $3B valuation Headspace carried after its 2021 merger with Ginger.
The announcement leads with 20,000 employer clients, Cigna and Kaiser integrations, and a 15,000-provider network. The consumer app barely features.
Contracted, renewing revenue is what got bought. A consumer subscription gets valued on its retention curve, and a buyer who cannot see one assumes the worst.
The Deep Dive
On October 1, some of your cash-pay tirzepatide patients get a cheaper way to stay on treatment. It does not run through you.
Caremark manages pharmacy benefits for roughly 88 million people. Plan sponsors that cover obesity drugs on its template formularies will have Zepbound and Wegovy at the same tier from October 1.
For a patient with that coverage, branded tirzepatide becomes a copay at the retail pharmacy, with a manufacturer savings card on top. You are charging that same patient hundreds of dollars a month for the compounded or self-pay version.
That gap does not need a competitor's ad to find it. The patient's own pharmacist will.
1. You cannot see the cohort, because intake never asked.
Most cash-pay funnels skip the insurance question on purpose. It adds a step, it lowers completion, and until now the answer changed nothing.
From October 1 it decides who is at risk. A cash-pay tirzepatide patient with Caremark commercial coverage and an employer that covers obesity drugs is one pharmacy conversation away from leaving.
So the first move is data, not copy. Ask your active book two things, whether they have prescription coverage through work and which company manages it, and store the answers as profile properties rather than survey responses nobody opens again.
2. The patient is not leaving treatment. They are leaving your pharmacy.
That distinction decides the response.
A patient who can get the same molecule for a copay will take it. Fighting that with a discount means matching a copay against a cash price, and no margin survives that.
What can survive is the relationship. Three routes, in order of how much revenue they keep:
Route the script yourself. If your clinical model allows a prescription to a retail pharmacy under the patient's benefit, offer it before they ask. You keep the provider relationship and the membership fee, and lose the pharmacy margin you were about to lose anyway.
Sell the care layer as a care layer. Provider check-ins, dose management, side-effect support, community. Priced as a service, not as a medication bundle.
If you can do neither, make the exit clean and tag it. Cancel reason: switched to insurance coverage. That tag is your January winback list.
3. January gives them back, if you kept the door open.
Plan sponsors reset benefits on January 1. Coverage gained in October can vanish at renewal, and employers dropping GLP-1 coverage has been a live trend all year.
A patient tagged "left for coverage" in October who loses it in January is the warmest lead you will have that month. Their provider relationship is with you, their dose history is with you, and January is when this category peaks anyway.
Build that winback now, while the reason is fresh, rather than reconstructing it in December from cancellation notes.
The takeaway: treat October 1 as a scheduled churn event. Find the exposed cohort this week, decide which of the three routes you can actually offer, and tag every coverage-driven exit so the January winback has a list to send to.
Quick Takes
End the abandonment flow with a question, not a last offer
The people still in your questionnaire abandonment flow after the last-call send have said no to every offer in it. One more discount is the weakest move left.
Send one email with no offer that only asks why they stopped: lettered options, plus "reply with the real reason if it is not on the list."
Two returns. The answers fix the front end, because a recurring side-effect worry belongs in the quiz and the ads, not in a better email. And a flow that ends by generating replies from the segment about to go quiet is protecting the sending reputation its length depends on.
Your retention data is probably not in your ESP
Seen this more than once in reviews: the "retention report" is open and click rates by cohort. That is engagement, being read as retention.
The table that answers "who is still paying in month four" lives in the billing system or the client's own database. An agency that never asked for access is reporting on the wrong table.
One Thing to Try
Pull your active GLP-1 patients paying cash for tirzepatide.
Send them one plain-text email from the provider this week, before October 1: "Some insurance plans start covering Zepbound on October 1. Reply with the name on your pharmacy benefit card and we'll tell you what it means for your treatment with us."
Store every reply as two profile properties: coverage yes or no, and the benefit manager's name. Anyone who confirms Caremark goes into an October 1 segment, and you decide now which route they get.
Under an hour. And you become the one who told them, instead of the pharmacist.
If retention is your biggest revenue leak, that’s what we fix. growthtrigger.xyz
Reply to this with your biggest retention challenge. I read and respond to every single one.