Home » Knowledge Center » Insights » Thinking Through Medtech Budgets & Projects

Thinking Through Medtech Budgets & Projects

The budget decides what you can afford. It does not decide what you are building. That is a separate set of decisions, and most teams leave them for January.

A budget line is not a program. It is permission to run one. The gap between the two is a short list of decisions that nobody owns in December because the program has not officially started and that everybody argues about in February because it has.

I learned this the hard way, running programs that had money, a start date, and a team but still spent their first quarter deciding what they were. Nobody called it a delay. We called it ramp-up, and we built it into the schedule the next year too.

This is not unusual. Researchers writing in MIT Sloan Management Review studied how companies green-light R&D projects and found that approval decisions tend to rest on technical merit while leaving out market potential, business planning, and financing, and that most organizations “fail to capture sufficient information on their failures,” so the same gaps repeat.¹ A program can clear the budget process with real questions still open. In my experience, it usually does.

The decisions a budget does not make

Three questions have to be answered before a program can genuinely start, and none of them are staffing questions.

What is this product not going to do? The requirements that drive architecture are a short list: intended use, the clinical setting, the primary user, connectivity, and the boundary of what the device will never attempt. A moving target in December is a discussion. A moving target in March is a redesign.

Can it be built on what you already have? Either the current platform carries it, or this is a new platform wearing a product’s name. That one answer changes the budget you just approved, and it is cheaper to learn it before the kickoff than after.

Which decisions lock? Processor family, operating system, regulatory pathway, and whether the product meets the FDA’s definition of a cyber device. These cannot be unwound in the spring without unwinding everything built on them. The others can wait. Knowing which is which is most of the job.

Why the timing matters more in medtech

In an unregulated product, a late requirements change costs you rework. In a Class II or III device, it costs you rework plus everything design controls attach to it. Once design inputs are approved, a change is a design change. It reopens verification, it touches the risk file, and it forces a decision about whether a new submission is required. That machinery does not scale down for small changes. And the engineering itself runs on both sides of the clinical work, so the question is never which phase you are in. It is whether the thing you are changing was supposed to be settled.

The changes you do not catch are worse, because they ship. Then they become modifications to an approved device. A 2023 study in JAMA Network Open found that modifications approved through FDA supplements were associated with roughly a 30 percent increase in the risk of recall, including Class 1 recalls. In the same analysis, device design was the single largest root cause of recalls, accounting for nearly 30 percent of all recalls and more than half of Class 1 recalls.² An unresolved December question does not stay in December.

What December is actually for

This work does not need the full team. It needs a systems architect, someone who owns the clinical and user side, a regulatory voice, and whoever will lead the program, for a few weeks. It needs decisions written down, with the reversible ones marked as reversible so nobody relitigates them in April.

December is the one month that has room for this. The organization is already in planning mode. The hard part is that the people best placed to do it are usually finishing the current year, which is the collision I wrote about last month. One honest option is to let a partner who does early-phase requirements and architecture work for a living carry the December decisions, so your own team starts January with answers rather than a debate.

A test

Take your largest 2027 program and ask one question: If it started tomorrow, what would be the first thing the team would argue about?

That argument is your December work. It will happen either way. The only choice is whether it happens before the clock starts.

If you want to think that through with someone who has run that December, we are glad to talk at SuntraMedTech.com/contact.


Sources
1 Thorsten Grohsjean, Linus Dahlander, Ammon Salter, and Paola Criscuolo, “Better Ways to Green-Light New Projects,” MIT Sloan Management Review 63, no. 2 (Winter 2022): 33–38.
2 Jonathan R. Dubin et al., “Risk of Recall Associated With Modifications to High-risk Medical Devices Approved Through US Food and Drug Administration Supplements,” JAMA Network Open 6, no. 4 (2023): e237699.

Written by:

Bryan Gilpin

President

Bryan has spent his career building and growing great organizations to deliver technology that improves lives around the world

Bryan Gilpin, President of Sunrise Labs

Knowledge center

From the archives

Most recent posts