When starting gets easy, share is no longer won at the counter. It is kept on the Tuesdays, one refill at a time, inside the system that can tell you whether the patient came back.
A franchise loses its base one un-refilled prescription at a time. When a pill makes starting easy, share stops being won and starts being kept.
A franchise does not lose its base in a boardroom. It loses it the way most things are lost: quietly, one prescription that never gets refilled, on an ordinary Tuesday, weeks before anyone runs the report that would have caught it.
The moat was the needle
For years the thing protecting an injectable franchise was not the brand. It was the injection. A patient who started had cleared a small wall, and clearing it again somewhere else took effort.
Then the wall came down.
In December 2025 the FDA cleared an oral form of semaglutide for obesity. In April 2026 it approved orforglipron, the first GLP-1 you can take as a small-molecule pill, any time of day, no food or water rules, with a type 2 diabetes filing close behind.
A patient who can start on a pill can start one somewhere else just as easily.
When the cost of starting falls, the base stops being something the injection holds for you. It becomes something you earn back every month, or watch walk out the door to the pill that made the next start easiest.
The base you count is not the base you have
A brand counts its base in prescriptions written. The market counts it in patients still filling a year later. Those are different numbers, and the gap between them is where share actually moves. Start with how much of the field is still open: fewer than one in ten people who could benefit from a GLP-1 are on one. Most of the contest is not over patients anyone already holds. It is over the people no one has activated yet, and the recent starts already slipping out.
The twelfth month, not the first
The hard part of a cardiometabolic franchise was never the first prescription. It is the twelfth month. Across real-world cohorts, a large share of patients who start a GLP-1 have stopped within a year, and the steepest drop comes in the first three months, well before a clinician sees them again. The nausea shows up on a Tuesday. The next appointment is eight weeks out. The refill lapses somewhere in between, and no one is in the room when it does.
Each lapse is a patient whose weight or A1C tends to drift back, often within months, and whose next conversation about therapy may not include the brand that started them. Holding the base is the same work as serving the patient well. The reason to keep someone on a therapy is that the therapy was working.
The answer lives where a brand cannot go
To do anything about that Tuesday, you would have to know who started last month, who is about to miss a refill, who was prescribed and never filled. That information sits inside the health system and the plan, behind clinical and privacy walls. A manufacturer cannot see it, and should not.
A lapse is often an access event, not a failure of will. And one line decides what can even be said: what anyone may say to a patient turns on whether that patient already holds a prescription — branded to those who do, unbranded and guideline-based to everyone else. That is a legal boundary, not a style choice. The patient who never started and the one quietly leaving are reachable only from inside the data, by a partner the system has authorized to look.
You only know the base held if you can read it
Opens and clicks describe a campaign, not a patient. The only honest measure of whether a base held is in the claims and the clinical record: starts, fills, refills, therapy continued. Through the contested window the work divides cleanly — persistence support for the patients already on treatment, and unbranded activation of the still-untreated, each staying in its lane.
When a new entrant lands, the first patients to go are the ones who were never fully activated.
Reach them first, inside the system that sees them, and the base holds. When the category settles, the franchise that kept its eligible patients on appropriate therapy is the one that did not have to buy them back at a premium.
Activation that runs inside the system, not around it
Essence designs guideline-based activation for the eligible population and coordinates it through a national partner network of health systems and health plans. Every campaign is authorized by the system or plan that holds the data before it runs. The patient-level data never moves to the manufacturer, and Essence never holds or views it.
Messages to patients who already hold a prescription can be brand-specific. Every other message — to the eligible patient who has not started — stays unbranded and guideline-based, which is not a preference but a legal line. The work is measured on real-world claims and clinical data, not opens and clicks, so the result reads as starts, fills, and continued therapy.
It runs as a single coordination point across the network — hundreds of health systems and plans reaching more than 100 million patients — as a production deployment rather than a test, with governance built into every step.
Sources: AJMC and Novo Nordisk/Eli Lilly approval reporting (2025–2026); semaglutide discontinuation cohort (n=40,262, EASD 2025); Rodriguez PJ et al., 12-month GLP-1 discontinuation (n=125,474, 2025). Essence model and scale: company positioning materials.
See What Patient and Physician Activation Looks Like for Your Brand
See how Essence holds the base through the oral GLP-1 window — activating eligible and lapsing patients inside the system that sees them, and proving starts, fills and persistence in real-world data.