Web-based healthtech startups have invention risk. Surprise!

As readers of this blog know, I am interested in the potential of novel social networking and other internet-age techniques to transform our creaky healthcare system. A particularly intriguing class of startups is using these social techniques to change the behavior of patients in ways that lead to better health. For example, helping people to eat less, or avoid pre-diabetes, or manage their medication better, or avoid hospital readmission by taking better care of their heart disease.

The exciting thing about these web-based startups is that they use Lean startup methodology, have low capital requirements, and are staffed with young, energetic, change-the-world types.

And, putting on my investor hat, they have market risk perhaps, but no invention risk, because all this internet stuff is engineering right?

So, I was quite surprised recently, when I had reason to think more carefully about this class of companies, to realize that I think they do indeed have a pretty substantial invention risk. And interestingly, most of the ones I have looked at don’t realize they have invention risk, and are not managing their businesses as if they have invention risk. That is a worry.

Behavior change companies

There are some common themes among the startups in this class of behavior change companies.

  1. They often use social networking techniques, and virtual group social interactions to motivate change.
  2. Some of them use clever sensors that automatically track data that is then presented in motivating ways, requiring little or no effort on the part of the patient.
  3. Some use gamification (ie they try and make participating fun).
  4. Some use clinically-validated protocols. Some try behavior change approaches based on the work of various gurus.

Invention risk varies. Depends on customer segment.

Take as a simple example a product that has the potential to improve the health of a patient (eg helps them lose weight, or helps them manage their medication). And assume that the team developing the product has leveraged to the full all the latest lean techniques and has an elegant product.

My favorite question is “What is the definition of “it works” for this product? And does it indeed “work”, and what is the proof point that it works?”

The right answer to this depends on the target customer.

Patient as customer

If the patient is the customer, which by my definition means the patient pays (not the insurer or the employer), then the definition of “it works” might be that the product does indeed improve a patient’s health some measurable amount when the patient uses the product. And there would be a whole discussion about how much better the result needs to be compared to doing nothing to count as “working” or as “providing value”.

Now what if the first generation product does not “work” according to that criteria? What if it is supposed to improve medication adherence but on average it does not lead to medication adherence that is any better than whatever the patient is doing today?

Then this product has invention risk. Because the team has to “invent” a way that this product will indeed help patients to take their drugs on time (better than they do today). And that is most definitely not an engineering problem. And it may well not even be possible. And who knows how much time and money it will take even if it proves possible. Classic invention risk.

Employer or insurer as customer

Many of these startups make the case that most patients do not pay for themselves, and thus it makes more sense to target either the insurance companies that cover these patients or the employers that pay the premiums for the insurance.

In either case the value proposition is that the employer or insurer pays a bit extra for the new “app” for each patient in their large pool of covered individuals, and in return the total costs of covering the pool drop as a direct result of the health improvements described above.

So what is the definition of “it works” in this case? Well, the product certainly needs to improve patient health and thus at a minimum perform as well as it would for the case where the patient were the customer.

But for this customer group, with the value proposition I outlined above, there is a higher hurdle for “it works”. Now the product has also to lead over time to lower health-costs for the individuals that use it.

What if it does not? Well then the answer might be to “invent” a way that it does reduce costs (back to invention risk), or perhaps pivot to a different customer segment or use case.

Aspirin not vitamin

Unfortunately, when the insurer/employer is the customer there is an additional challenge. Because in order for these products to get real mindshare (be an aspirin not a vitamin – to use the investor cliche), the product needs not only to reduce costs for those patients who use it, but also to get used by a significant portion of the relevant patient pool. Otherwise it will not have a meaningful impact on the total costs of the insurer/employer, which is what it will likely take to be exciting to this customer class.

So now, the definition of “it works” has just got more challenging. In addition to the performance requirements above (makes patients who use it healthier, and reduces costs of managing those patients), now the product needs to be used by lots of patients. In other words, when offered to a large pool of appropriate patients, patient adoption needs to be high.

Now this is a very different type of requirement. Somehow the company needs to “invent” a way to get widespread adoption of its app. This has been very challenging in the past, and definitely fits my definition of “invention risk”.


To me there are some important lessons here.

  1. The target for “it works” depends a lot on the target customer segment.
  2. The level of invention risk could range from small to huge, depending on the target segment, and on the definition of “it works” for that customer type.
  3. In many cases it seems very likely these behaviour-change companies will be facing invention risk. This in itself is not a bad thing. In fact, if successful, this would likely allow them to create strong barriers to competitors.
  4. BUT, companies facing invention risk need to manage their product development quite differently to those with no invention risk.
  5. In particular, the approaches to managing invention risk are quite different than the practices adopted by most web-based startups (which don’t have invention risk). So, one needs to select role model companies carefully!


  1. […] to hardcore science projects, one of the most counter-intuitive things I have come to realize (here) is that many of the new, web-based healthtech startups that aim to change patient behavior, and […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: