Managing Expectations in Medical Device Development, Part 2
In my previous blog post , I explained how I co-founded a company with my University Professor and a hungry MBA graduate, managed to secure seed funding to develop a poof of principle prototype, then tried to jump straight into new product development without stepping back and looking at the big picture.
The business proposition was for a new application of an existing technology and had never been applied before to that particular patient group, so there were significant development unknowns. However, as technical director/CTO for the company, I realized the same issues kept popping up with our professional investors—expectations management and development timescales. This was entirely our own fault, as, when we were pitching to professional investors, our opening line was “We have a proof of principle prototype; all we have to do is miniaturize!” In hindsight, this was a mistake; to paraphrase Rumsfeld, we didn’t know what we didn’t know, which led to several issues.
Clinical Performance, Business Model, & Market Entry—Can You Have All Three?
As mentioned in part 1 of this blog, we did not see the value in performing a detailed review when beginning new product development, as we felt we were almost there and didn’t understand the value as, like all start-ups, we were severely cash limited. However, cash income was something the rest of management and the investors understood well, so multiple design optimizations were implemented in one iteration to reduce both consumable and device costs in order to “cover all angles” of the business model—make money on selling devices and make money on consumables.
To lower consumable cost, we developed a novel sensor manufacturing approach that could be highly automated and we were busy giving ourselves a “pat on the back” when the clinical study results came in. The results showed that the device was not commercially successful. After a few months of investigation, we discovered that the sensor was being used in a different way by the nurses than how we used them in the lab and that the low cost of the sensor and system was also directly affecting performance. Looking back, no other manufacturers who use the same underlying technology use a sensor manufacturing technique similar to what we developed—apparently, for good reason.
With the investors ready to pull the plug, the only option for management was to redevelop by diverting cash for other things, which ultimately hurt the company.
Things to do Differently
In retrospect, we should have spent less effort/tooling on cost reducing our disposable component and developed a market tester that used more expensive sensors (which we knew actually worked), allowing performance to be demonstrated in a real-life clinical setting. In part 1 of the blog, I mentioned how a formal product definition framework  may have identified some of these “to market” risks. This would have included a product pipeline strategy, showing what functionality/features and benefits each device generation would have from the outset, based on development risk and allowable timescales.
Secondly, during reports to management, both best and worst case timescales would usually be presented. Quite often, the best case timescale was the only figure that management actually heard, which then gets rolled into cash-flow projections and presented to the board. As the project inevitably overran, those conversations with investors became increasingly more difficult. A well-defined project gate structure  would also have allowed us to identify and manage risks earlier than we actually did. Clear stage gates would also have provided good communication anchors for management and investor discussions, as well as permitting staged investment.
This reminds me of the Star Trek: The Next Generation episode “Relics,” where Scottie (from the original Start Trek series) is passing on the benefit of his experience to a young whippersnapper named Geordie LaForge, chief engineer of the new Enterprise:
Geordi: I told the Captain I’d have this analysis done in an hour.
Scotty: How long will it really take?
Geordi: An hour!
Scotty: Oh, you didn’t tell him how long it would really take, did ya?
Geordi: Well, of course I did.
Scotty: Oh, laddie. You’ve got a lot to learn if you want people to think of you as a miracle worker.
With any luck, I will have a similar conversation one day, but hopefully without spending 75 years trapped in suspended animation, as Scottie did.