Patient Support Operating Models: The Router Nobody Built

Eight pharmaceutical manufacturers have each built a set of affordability programmes for a flagship medicine. Not one of them has built the thing that tells a patient which programme is hers. That function does exist in every case, and in every case it is a telephone number.

Eight manufacturers have built the routes and none has built the router

I spent part of this week doing something a brand team could do in an afternoon and almost nobody does. I opened the patient support pages for eight flagship branded medicines, each from a different manufacturer, across immunology, oncology, cardiometabolic disease and obesity, and I counted the doors.

A door is a distinct route a patient can take to afford her medicine. A copay card is one. A bridge programme that covers her while a prior authorization is contested is another. A patient assistance programme for the uninsured is a third, an independent foundation a fourth, a cash pay price a fifth. Each has its own name, its own enrollment path and its own eligibility condition, and the conditions are frequently mutually exclusive.

Every brand in the set published several. The range ran from three to seven routes built by the manufacturer, rising to as many as nine once the government programmes the page refers her to are counted. The counting rule matters and I will come back to it, because a stricter definition lowers the numbers considerably. What a stricter definition does not touch is the finding underneath them.

The audit covered eight brands and found the same shape in each

Five of the eight offer an insurance type selector, which asks a patient to declare whether she is commercially insured, on Medicare, on Medicaid or uninsured, and then shows her a filtered list. That sorts the doors. It does not open one. One manufacturer runs a genuine eligibility engine, and it sits two domains from the brand page, tests a single programme, and returns a refusal to a commercially insured patient without routing her anywhere else.

Discoverability is not the problem, which closes off the easy explanation. Every list sat one click from the brand's main patient page, two at most. The patient can find the doors without difficulty. What she cannot find out is which one she is standing in front of.

A list of conditions is not the same as an answer

Consider what the page actually asks of her. One brand's copay card requires commercial insurance, including exchange, federal employee and state employee plans, plus residence in one of the fifty states or four named territories and a prescription for an approved indication. It then excludes anything paid in whole or in part by Medicaid, Medicare, the VA, the Department of Defense, TRICARE or any state pharmaceutical assistance programme. The same manufacturer's assistance programme applies instead if she has no insurance, or has Medicare Part D, or is having difficulty paying, and is assessed case by case against a household income threshold the page does not publish.

Those two paragraphs are both accurate, both necessary and both written by lawyers who were right to write them that way. Read by a person who has just been prescribed something and does not know what an exchange plan is, they are a quiz with no answer key. She is being asked to perform an eligibility determination on herself, using categories drawn from benefit design, in order to receive help she has already been told she may qualify for.

In the commercial organizations I have worked inside, this is not experienced as a design failure, because no single person ever sees it. The copay card team owns a card that works. The assistance programme team owns a programme that works. The foundation relationship is managed by someone else again. Each component is well built and the compound is nobody's.

The routing function exists and it runs on a telephone

Here is the part that turns a website observation into an operating model observation. Every brand in the set closes with a phone number, and in at least two cases the page says in plain words what that number is for. Programme eligibility is assessed by speaking to a case manager. One oncology page lists a single number for checking whether insurance covers the medicine, learning about copay options, finding out about free medicine, and speaking to a representative, all four.

So the routing function is not missing from the operating model. It has been built, it is staffed, and it works. It simply runs as a call centre rather than as a system, which tends to have three consequences a brand team can feel in its numbers without ever tracing them back here.

It is sized to call volume rather than to prescription volume, so its capacity is set by how many people ring rather than by how many people were prescribed. It operates during business hours, which are the hours a working patient is also at work. It also produces almost no structured data, because what a case manager resolves in four minutes of conversation tends to be recorded, at best, as an enrollment in whichever programme the call ended in.

The patient who cannot tell which door is hers appears in no denominator

That last consequence is the commercially expensive one and it deserves to be stated carefully. A manufacturer can see, with precision, how many patients enrolled in each programme. What it cannot see is the patient who opened the page, read five sets of conditions, could not determine which applied to her, and closed the tab.

She enrolled in nothing, so she appears in no programme's denominator. She called nobody, so she appears in no call log. She does appear, eventually, as an abandoned prescription with no attributable cause, which is the category where this loss currently sits and where it is invariably read as a price problem.

Price is sometimes the honest answer. Often it may not be, because the patient in question was eligible for assistance that would have resolved the price entirely. What stopped her may not have been cost at all.

It was an orientation barrier wearing a cost barrier's clothes, and the two have completely different owners and completely different fixes. One of them is a pricing decision taken at executive level. The other is a routing problem on a page the company already controls.

Four of the six Consumer-Grade conditions fail on a page the company controls

The Consumer-Grade standard is a closed set of six conditions, and this page appears to fail four of them. That is unusual for an asset a company owns outright, with no payer, pharmacy or prescriber standing in the way.

Recognition fails, because the system does not know who she is even when she is signed in elsewhere on the same property. Coherence fails, because the routes are presented as a list rather than as a path, and several of them lead to different domains. Effort fails, because the sorting work has been transferred to the person with the least information and the most at stake. Resolution fails, because nothing on the page closes the question, and the only thing that does close it is a telephone call she has to initiate.

Anticipation and Visibility survive. The company has anticipated that she will need help paying, which is why all these programmes exist, and it is visible to her that they do. The failure is specific and it sits between knowing she will need help and knowing which help is hers.

What a brand team can count on its own site tomorrow morning

The useful thing about this finding is that verifying it requires no data, no vendor and no budget. It takes one person, one browser and about forty minutes.

Count the distinct affordability routes a patient can reach from the brand's main patient page. Count how many separate domains she can be sent across while doing it. Then try to determine, as her, which route applies, using only what the pages say. Finally, find the mechanism that answers that question for her, and look at whether it is a system or a staffed line.

If the answer is a staffed line, that is not a failure of the people on it. In my experience those teams are among the most capable in any commercial organization and they resolve what reaches them. The question is what fraction of the people who needed them ever rang, and nothing in the current operating model can answer it.

The measure that would is the one this practice is built on. A Realization Rate anchored at the prescription rather than at the enrollment counts everybody who was prescribed, including the patient who never entered any programme, which is exactly the population the programme level numbers are structurally unable to see. The orientation gap is a Path to Fulfill problem and it is measurable the moment the denominator moves.

Methodology, what I could not establish, and the objection I would make

The audit was conducted on 10 October 2026 by opening each manufacturer's own patient support pages and recording the exact wording of every programme name and eligibility condition. Brand sites are rebuilt frequently and offer terms reset annually, so every figure is a point in time observation. Three of the eight brands were then re-verified independently against the live pages before publication, and the orientation finding held in all three, with two of the three stating on the page that eligibility is determined by telephone.

The counting rule is the weakest part and it has to travel with the numbers. I counted a route as distinct where the manufacturer presents it as a separate thing a patient does, with its own name, enrollment path or eligibility condition. Collapse that to money moving mechanisms only and the counts fall to roughly three to five per brand.

The orientation finding is unaffected by that collapse, which is why it leads here and the counts support rather than the reverse. Government programmes a manufacturer merely refers a patient to are reported separately and are not credited to the manufacturer, since it did not build them, though from the patient's chair they are doors on the same page.

Two sites could not be reached during the audit, which makes one brand's count an undercount rather than an estimate. No company in this set was selected for being worse than its peers, and the set deliberately excludes manufacturers where I have a prior working relationship.

The strongest objection a reader can make is that auditing a public website measures the brochure rather than the system. In oncology particularly, the practice's financial navigator absorbs most of this work, and a patient in that setting may never face the page at all.

The objection is correct as far as it goes. It may also describe the finding rather than refute it. A function discharged by a person the manufacturer does not employ, does not train, does not fund and cannot observe is a reasonable definition of a seam that nobody owns. The navigator is doing the routing, and in my experience doing it well. She is also doing it instead of the operating model, and the manufacturer learns nothing from any of it.

What I cannot tell you is how large the loss is. That would need the denominator this piece argues does not currently exist, which is an uncomfortable place to end and an honest one.

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: Pharma CX, why it stalls and what replaces it, Nobody knows their Realization Rate, including us and How to choose a pharma CX partner