LillyDirect: What Lilly Actually Changed, and What You Can Copy
Lilly did not build a better customer experience. It removed two of the three people from its market, and that is a different kind of move entirely.
LillyDirect gets written about as a digital platform, which is the least interesting thing about it. Strip the website away and what is left is a pricing and channel decision that changed who Lilly has to satisfy in order to get paid. That distinction matters to you, because most companies are about to copy the wrong half of it.
What did Lilly actually change?
The route, not the experience.
When the platform launched in January 2024, David Ricks framed it around burden. The press release said a complex US healthcare system "adds to the burdens patients face when managing a chronic disease." Speaking to Axios the same week, he was blunter about the model: "We're used to buying consumer goods directly from manufacturers all the time on online websites."
Underneath the language, three things changed at once. A self pay price, so coverage stops being the gate. A telehealth route to a prescriber, so access to a physician stops being the gate. Direct fulfillment, so the pharmacy counter stops being the gate.
Notice what is missing from that list, and compare it to your own plan. No survey. No workshop. No experience function. Three structural changes to how the product reaches a patient, made by commercial leaders.
Did it work?
On Lilly's own account, yes, and at a scale nobody in this industry predicted two years ago.
At the J.P. Morgan Healthcare Conference in January 2026, Ricks told the room that "basically a million people a month go online and buy GLP-1 medicine directly from a drug manufacturer." He also said the vials sold through the channel had become roughly the second best selling obesity product after Zepbound in an autoinjector.
Read that carefully. The million a month figure describes the whole direct channel across manufacturers, not Lilly alone, and both numbers are the CEO's own claims rather than audited disclosures. Treat them as directionally real and precisely unverified.
Even discounted, the signal holds. A route that did not exist in 2023 is now moving product at a volume that competes with the company's own flagship presentation. Ask yourself what share of your brand's volume moves through a channel you did not have three years ago.
Which part of the problem did it solve?
One of three, and naming which one is what makes the case usable.
We model the distance between your science and your patient as three sequential paths. The Path to Prescribe , where a patient is identified, diagnosed, referred and a clinician decides. The Path to Fulfill , where that decision survives coverage, prior authorization, cost and dispensing. The Path to Adhere , where the patient stays on therapy long enough to get the benefit your trial demonstrated.
LillyDirect is a Path to Fulfill solution, and an elegant one. It takes the stage where most earned value is lost in the US and shortens it from weeks to days. It does very little for the other two, which is not a criticism of the design. It is a reminder to check which path you are actually funding.
Why did it work, in terms you can use?
Because it dissolved three of the four conditions that make consumer models fail in pharma, rather than managing around them.
Your market has three people in it. A physician chooses, a payer covers, a patient uses. Lilly did not solve that. It deleted the payer from the transaction by setting a cash price, and routed around the access problem with telehealth. One buyer, one decision, one path.
Your feedback loop is partly closed by law and partly invisible by structure. A direct channel hands Lilly first party data it never had. Who ordered, who refilled, who stopped, and when. That is the loop closing, and it is the most undervalued asset in the whole program.
Your value event is sustained therapy months after the transaction your system counts. In a refill channel you can see persistence directly rather than inferring it from claims data that arrives late and incomplete.
The fourth one is the hinge. Regulation governs claims, not burden. Nothing in the rules prevented Lilly from doing any of this, because none of it is a claim about efficacy or safety. It is an attack on friction, and friction was never the regulated part.
What will most companies copy, and why is that wrong?
They will build the website. The website is the least transferable part.
The portal is downstream of a pricing decision. LillyDirect works because a meaningful number of patients will pay cash for a GLP-1, which is true of GLP-1 medicines and almost nothing else in your portfolio. In oncology, in rare disease, in most specialty categories, the payer cannot be deleted because the therapy costs more than any patient will put on a card.
So if you run a brand where coverage is mandatory, copying the channel gets you a handsome site that changes nothing. Your three person market is still three people. Your gate is still the prior authorization, and the published numbers on that gate are getting worse. A 2026 JAMA study summarized by Johns Hopkins found insurer rejections reached 40.7% of initial brand name attempts in 2024, up from 24.3% in 2018. Of those rejected scripts, 48.4% were never followed by a fill of that drug or anything in its class within 90 days.
The transferable lesson is not the portal. It is that Lilly identified the specific structural gate between its science and its patients, and then changed the structure rather than improving the experience around it. That question has a different answer for every brand you run.
What does it cost to own the whole path?
Scrutiny, and the detail is worth studying before you design your own.
In July 2025, after a nine month investigation, Senators Durbin, Welch, Warren and Sanders published a report on manufacturer telehealth platforms. Three figures from their findings on LillyDirect are the ones to note, not as a verdict on Lilly but as a description of what any company inherits when it owns the whole path.
Patients routed to a telehealth visit received a prescription 74% of the time. At one partner, more than 500 of 620 LillyDirect patients were prescribed a medication, against 39% of that partner's overall patients. LillyDirect users seeing a contracted physician were six times more likely to receive a Lilly product than another brand.
Durbin said the findings "shine a light on potential conflicts of interest and inappropriate prescribing." Lilly has maintained that it does not incentivize physicians to prescribe its own medicines.
Hold both facts at once, because both are true. The channel removed real burden for real patients. It also put the company in a position where every conversion rate becomes an exhibit, which is a structural consequence of integrating the path rather than evidence of intent.
If you build one of these, design the governance before the interface. Decide in advance how a prescriber is selected, what the patient sees before they meet one, and who audits the prescribing pattern quarterly. Those are answerable questions, and they are much harder to answer after a subpoena than before a launch.
What do you do if you cannot delete the payer?
You work the seams instead, and you measure the path.
Most brands will never have a cash option. For those, the gate is the interval between the prescription and the patient, and the work is unglamorous.
Find where patients stop moving. Put a name on each of those seams. Measure script to start conversion and time to therapy across your Path to Fulfill. Measure persistence and barrier resolution speed across your Path to Adhere. Then measure realized value against the value your brand had already earned, which is your Realization Rate.
Put those numbers on a commercial leader's scorecard rather than in a quarterly experience readout. We call the money sitting in that interval value leakage, the discipline that recovers it Customer Excellence, and the bar it is built to clear Consumer-Grade.
Lilly took the shortcut available to a company with a cash paying category. The rest of the industry has to do it the long way, and the long way is still a commercial system change rather than a CX program.
Key takeaways
- LillyDirect is a pricing and channel decision with a website attached, and it solves the Path to Fulfill rather than the whole distance to the patient.
- It worked by deleting the payer from the transaction, which is available to you only in categories where patients will pay cash.
- The direct channel closed a feedback loop pharma has never had, giving Lilly first party data on ordering, refill and discontinuation.
- Nothing in regulation prevented any of it, because none of it was a claim. Regulation governs claims, not burden.
- Owning the whole path from advertisement to prescription invites scrutiny of your conversion rates, so design the governance before the interface.
Questions to ask your leadership team
- For your lead brand, what is the single structural gate between the prescription and the patient, and who owns it?
- Would any meaningful share of your patients pay cash, and at what price? If none would, the direct channel is not your answer.
- What would you learn about persistence if you could see refills directly, and what are you inferring today instead?
- Which of your current constraints are regulatory claims rules, and which are operational burden you have simply never challenged?
- If you launched a direct channel, who in your organization would audit its prescribing patterns, and how often?
About the author
Wayne Simmons is the founder of The Customer Excellence AGENCY and the author of The Customer Excellence Enterprise (Wiley, 2024). He is founding faculty of the MS in Customer Experience Management at Michigan State University's Broad College of Business. He led global customer excellence in Pfizer's first Chief Marketing Organization and in Bayer's Customer Powerhouse. Related reading: Lilly and Novo solved one path, two are still open, Why consumer CX does not translate to pharma and How to measure customer experience in pharma







