Categories
Entrepreneur Tips and Discussions

Why are Health startups often duds? Especially digital ones.

I have spent some time looking at startup ecosystems of various industries and one that stands out is that there are very few unicorn level health startups, and literally none that are global….

This is quite surprising considering that health is something everyone needs, and the size of the sector in the traditional services and bricks & mortar sense.

Curious to know what people think is the missing piece of the puzzle, especially from the digital side.



View Reddit by NjabzView Source

32 replies on “Why are Health startups often duds? Especially digital ones.”

I’m going to go ahead and put government/regulations/security obstacles on the board, but I don’t personally know much about this field, this is just a hunch.

Having failed a healthcare startup I would say:

**Financing** – it takes a lot of cash to build healthcare services from scratch, it’s much easier if you have corporate budgets.

**Regulation** – breaking into the market requires high levels of compliance, again, this takes time, cash and expertise

**Evidence** – another factor that slows you down, healthcare services generally require studies to show impact, the rapid iterations of startups are counter intuitive to this kind of process

Overall I think you will see many unicorn startups in healthcare, but they will take a long time to get there. If you want a good proxy, look at wellbeing instead. Wellbeing is less regulated, requires less evidence and this less money to grow.

I was recently interviewing with some health startups, and they all mentioned the biggest challenge with being broader nationally (at least in the US) is that every state has different regulations with healthcare, and particularly with digital/telehealth medicine. This makes expanding services costly and drawn out. Healthcare is also a pretty deep rooted institution in most places, and changing it is very difficult, especially when seeking out patients as the ultimate consumer.

The modern concept of a software startup is related, perhaps to an inseparable degree, to the “release early, release often” strategy Bill Gates invented in the early days of Microsoft. The idea is that you put out (what we would now call) an MVP, then watch what your customers do with it and be willing to (what we would now call) pivot. The software developer and the early customers go on a journey together, hoping to arrive, at the end of the journey, at a valuable and worthwhile product. When this all works out successfully, the startup gets to be a unicorn, and the customer gets…

Well, the customer kind of gets screwed, don’t they? Early customers have to deal with buggy and poorly-thought-out software, and also with constant tweaking and changing. If the software was just perfect for a day one user’s needs, it’s pretty likely their must-have features will get dropped as the startup pivots to seek better or larger opportunities.

To be an early stage startup customer, you have to be emotionally invested enough in the success of the startup, and the feeling of being part of something greater than yourself, that you’ve willing to overlook your own objectively bad experience as a customer. Your patronage of the business is as much because you want a future product to exist, or for the world to change in some way, as because you think it’s good value for money _now_.

Why doesn’t this work in the health field? Quite simply, because people aren’t willing or able to engage in risky speculations when their health is at stake. The iterative, “share the risk with the customer” model is just less applicable to fields where the customer faces serious consequences if everything doesn’t work out.

Most startups fail, and most people aren’t willing to risk failure when it comes to their health.

I’m a healthcare compliance attorney with deep understanding of regulations impacting certain parts of the healthcare industry, namely on the payer/insurer/care coordination side of things. I’ve been trying to think of how to leverage my expertise to start a business other than your typical solo law practice/consultancy.

To your question, I see big opportunities in creating innovative healthcare delivery models, such as new ways to deliver primary care services through exclusive provider networks. Also, there is much money to be made in the bill negotiation/provider reimbursement space at the moment where there is no price transparency in health costs.

I know this is a broad question, but how would one make money off of having niche expertise in regulations impacting a particular industry (in my case health)? My only thought so far writing e-books a eventually monetizing a blog/website. Any thoughts?

Really it boils down to a poor understanding of the legal and practical landscape of medicine. Any digital healthcare platform is dead in the water if it strays to close to any of the narrow red lines that define what healthcare providers can and cannot do (especially from a distance without an in-person physical examination, when talking about telehealth). Those narrow red lines are also different in every country. Healthcare is a heavily regulated space. Then you have to add in the fact that between those narrow lines you have to find the combination of need and opportunity (nobody is doing it or doing it well, but it would be useful) and that means having industry-specific knowledge. Healthcare is no exception to the “tech startups tend to be great at tech and crappy at understanding the industries they want to apply tech to” rule of thumb. If you want to tackle healthcare then you’d better find someone who really knows the healthcare space as their core competency on your founding team.

I’ve previously worked in healthcare ops analysis and technology, and two factors that haven’t already been mentioned by others in the thread are the entrenched players (e.g. medical records) and the length of the sales cycle (it takes years, and the costs associated are significant (anecdotally, one person in the process used to complain about the expectations set with decision-makers by large pharma and “drug reps” being able to easily throw around significant perks (grants, travel, etc.)).

My experience, doctors are the cheapest, most demanding, worst customers you could have for almost any product. I am not in healthcare per se, but I do work often with them when we happen to sell our products to family practices, clinics, etc.. I hate it. They will haggle about $30 on a 13k bill, take forever to pay, complain about everything, want freebies.. just the worst all around. I’ve talked with other business owners about it before and this is not unique to my situation. Why? I have no idea. It makes zero sense to me.

Health is global, but certain countries have different needs.

A country with universal free healthcare has other needs than … Simbabwe.

A country with lot of space like Australia has more need for telemedicine than Luxembourg.

It’s hard to scale something that’s inherently difficult.

This is a serious challenge. I’ve worked in healthcare innovation for many years, and there are fundamental challenges that make innovation so slow in healthcare:

1. Healthcare organizations (health systems, payers, policy makers) aren’t built for risk and uncertainty, and working with startups is inherently risky.
2. It’s not very easy to integrate new technologies into these organizations. they are typically very old and/or proprietary systems.
3. There is very little standardization across this ecosystem: providers (nurses and doctors) and systems vary greatly in how they document and bill, and the systems to capture and normalize this data is almost non-existent
4. Providers are resistant to change. Mostly, because of already high workloads, but also because many are old-fashioned and not well trained in technology

I could list a bunch more reasons. Unlike almost any other industry, healthcare really is an interdependent ecosystem where are the players (health systems, payers, policy makers and of course patients) have to work together. That kind of collaboration is difficult without major changes.

The burn rate and how long startup investment dollars last until (if?) the drug passes enough trials to warrant further investment – at least for drug development.

Compliance, capex, and investment dollar crunches kill service providers.

And yet, there are certain areas that are pretty awesome. Stand-alone ERs in the US are an example. Elder care facility roll-ups.

People who are good at caring for others very often don’t have the time or patience to learn a new system/tool/gizmo. Add to that the huge barrier to entry of the FDA (or any other health regulating body), and trying to compete with the existing big names in the industry. It just becomes a never-ending fight to just stay afloat. I worked for an established digital healthcare company, and it felt like that there. I can only imagine what it feels like for smaller firms.

There are a few success stories, and often they arise from breaking into a niche market or partnering with a bigger company.

I’ve counseled a few health duds before, and from my (admittedly limited) experience, it mostly boils down to:

1. **the delivery sucks**: “health in an app” rarely works because a lot of people like a good dose of human contact in their health. This is something a lot of wide-eyed, naïve entrepreneurs ignore starting out (I call it “The Church of our Lady of the App”).

2. **the market isn’t there**: your definition of “health” may not be widely shared. I’ve actually seen a really well-meaning lady try to launch a health startup around the concept of movie-nights-for-mental-health, which at first glance might seem like a cool idea, but her definition of mental health was pretty much “things I suffer from specifically”.

3. **lack of resilience**: a lot of well-meaning entrepreneurs give up when their vision is challenged for any reason, and in Health this happens perhaps earlier, more often, and more strongly than in, say, Social. This mostly stems from points 1 and 2 and is most definitely not Health-specific, but I do feel like it’s harder to work through in Health than other industries.

As for regulation, as many have raised in the comments: it does happen, but not as much as you might think, especially not in “digital Health” specifically. I’ve actually seen very few startups die straight up due to regulation; usually, if you’re doing Human Health and you have a significant regulatory burden, by the time you get to that part you’ve already gone through a lot of validation (see points 1 and 2 above), and if you make it past that, there are specialist investors and consultants that can pilot you through the regulation.

If you fail in points 1 and 2 it’s easy to blame it on “regulation” though, so YMMV.

In addition to everything that has been said, you don’t see unicorns also because the few successful ones get gobbled up by the giants in the industry

One thing I haven’t seen mentioned are insane sales cycles: 18 months is not unheard of. You’re selling to large institutions with inertia, insane politicking who are already way over their budget. Everything has to go through tons of internal reviews: IT, legal, privacy, purchasing. And quite often the status quo is there because of some insider connection, think of how likely are you even to find out much less avoid or change the mind of such a veto holder.

I share your observation. You have received many good suggestions of causes, several which I agree with.

I think therefore as a matter of understanding the hurdles even better one should look at those few startup companies that did make it in some sense.

My rough hypothesis is that to really reach the healthcare market, one needs to be big with lots of regulatory affairs staff, lawyers, and people who engage well with the political process. Maybe for good reasons? Regardless our political views, the healthcare market is bad for startups. Some startups have contributed though by targeting technological gaps and needs of these large players.

The pharmaceutical industry is the same. The companies actually selling medicine globally are very few in number, very large, and mostly quite old. However, there is a vibrant startup ecosystem, because there is a profitable business model investors and entrepreneurs recognize: develop innovative drugs and analytics, bring to early stages of clinical trial, and then be acquired by a big player, all efforts funded by risk capital.

I think the acquisition of Flatiron Health by Roche is an interesting case where a startup in digital health had an exit along those lines. It is interesting that both Apple and Amazon are looking to get a foot in on the healthcare market. If some of the bigger players develop a business model into which creative technology companies can insert novel technology, then this may turn into how pharma ecosystem looks today. I don’t think we’re there yet. However, with healthcare costs becoming the largest expense, and growing fast, for many governments, the people who figure out the business model will unlock a lot of value. But it requires both governments and big companies to do some heavy lifting before startups can reliably become major contributors to the ecosystem.

Digital healthcare still needs to go through diagnostic testing and trials just as if it were a drug in order to make health claims. The levels of proof (safety and efficacy) are also very different in every country.

Aka this is expensive! And you don’t know if you can make that return

Also no matter how digital it still requires some device that will also need to be validated. E.g. I know of a potential clinical study where the client (biotech company) was required to have 4,000 subjects’ data recorded by an Apple Watch to show that their designed program was effective in increasing compliance post heart attack. And furthermore following their protocol will decrease chances of another heart attack.

Even something as advanced as the Apple Watch has the tools and capabilities to support this but if they were to make claims that it will serve as a tool to reduce/elevate/maintain or stop/start some health function it would increase the risk of litigation.

Which is expensive!

Top of the head issues/risks:
– apps crash
– deviCes crash
– patient compliance
– who is going to pay for the apple watches
– devices break
– who will monitor data ? Real time? Who is going to pay for that
– what will be the emergency protocol for a life threatening event ?
– are there better/less risky options? Eg stronger medicines, transferring risk to family/caretakers, etc.

Finally I think there will be more “digital” treatments in healthcare in the future but at this time you are right, there are missing pieces to the puzzle barring financials, two of which are device access and risk of use for designer and subject.

Innovative technologies alone are usually enough to disrupt other industries, but I guess when it comes to healthcare it’s a lot more challenging to build up the type of synergy and momentum required with so many regulations and red tapes to cross. Meaningful changes must be supported and even initiated right from the start by the regulators to really have a chance.

It’s a huge risk to the existing revenue stream which is enormous so probably no one wants to commit. I mean people are already having a tough time trying to understand how the insurance companies negotiate drug prices with their suppliers. It’s a massive system worth $3 trillion dollars per year with too many complex connections. Who wants to assume such responsibility on the regulator’s side?

“unicorn level health startup” suggests a few rounds of funding, with growing capital each round corresponding to growth in the product or TAM. but in addition to the hosts of reasons people have left (capital-intensive, regulatory policies, enormous culture change), it’s important to recognize that healthcare is a long game…

i don’t think that the bootstrapped greenfield startup would be able to scale fast enough to _effectively_ serve the users for each round of funding _and_ demonstrate growth in the market (18 month runway for each round… there’s always something you sacrifice). and “the market” are patients. and patients are people!

Had a loss in digital health.

The answer is that it costs so much money to go to market and nobody wants to invest in a prerevenue startup.

And the bigger players in health move so god damn slow that you’ll be out of money by the time
they can act.

Even if a hospital you’re working with is really interested in your product, a big issue is red tape — there’s a ton of it. Hospitals tend to move slowly, much more slowly than a blooming startup can often afford.

My favourite is *Connecting for Health* – digitizing patients records in UK.Total cost £12b. It dwarfs the top failed IT projects by an order of magnitude.[https://en.wikipedia.org/wiki/List_of_failed_and_overbudget_custom_software_projects](https://en.wikipedia.org/wiki/List_of_failed_and_overbudget_custom_software_projects)

Can you imagine a startup the burns twelve billions pounds and fails to deliver??[https://www.theguardian.com/society/2013/sep/18/nhs-records-system-10bn](https://www.theguardian.com/society/2013/sep/18/nhs-records-system-10bn)

​

It is the biggest failed IT project in the world ever.

Nobody went to jail. Nobody claim any responsibility for the failure.

Why it failed?A single biggest reason why it failed is the fact the fact that health care in UK is provided by government. No real knowledge what technology is feasible, how much it would cost, what patients actually need and which of their needs can actually be satisfied given how much is collected in taxes. Government taking on such a project basically means there is no upper limit on how much money can be burnt. And as you can see clearly the limit only kept moving up.

Afterwards, the conclusion was obvious, the way to avoid such disaster in the future is to make the money follows the services, no delivery, no payment; when there is competition most importantly. Competition is a single incentive that provides higher quality for lower price. Once you remove competition from the equation you end up with £12b bill.

With private equities and fraction of that money burnt the leader of such a failed project would be thrown out of the company and risk charges for mismanagement. But not with public health care, here there are no thresholds other than public outrage that only kicks in when mass media started covering this.

I was questioning myself about this recently. Why there is no startup that, improves the health care system we already have?

I live in Brasil, so must of my thoughts is focused on our system.

A startup in the health care would be focused on NOT to substitute the actual system. But to improve the system and, give the companies an opportunity of better way to dealing with the data they already have and, the new data that comes with the improvements.

**What would be the startup:**

**Users:** we would have three types of user. The population, hospitals (private and public) and institutions. These users use for free. We do that because we need them to be active on the platform.

**Services:** in the beginning, we would provide better way for people get medical appointments, virtual appointments for non-critical, social assistent, online prescription, etc.

**Revenue:** this is where get complicated. I was thinking some options like

Users buy a gift, wich would be a necklace or bracelet, with a QR code for the health profile (that can only access by the user and doctors, this information can be viewed by the user).

We could use the users data to offer discounts on drugs and other related heath products.

Could use the data to sell information to insurance companies and advertisement companies.

Those two last one’s would be chosen by users, the platform comes with the default option of discounts on pharmacy, but they can change that.

Again, this is very complicated. Even we start small, getting a state and try. Seeing how things work with the aspects of dealing with engagement of the users, pharmacy, insurance companies, state government and the platform.

It’s very complicated but not impossible.

I co-founded a HealthTech startup. There are a lot of reasons. Regulation is the most obvious. If you need to be HIPAA compliant, you need to be from day one. You likely won’t see one at the global level because healthcare is highly fragmented across countries.

In America, most people don’t view healthcare as something they purchase/pay for. You purchase plumbing services, you purchase legal services, you purchase car service. With healthcare, you HAVE insurance. Insurance is supposed to pay for all of those things you need for healthcare. Anything that insurance doesn’t pay for is a sharp stick you need to deal with. This pretty much eliminates the entire B2C market in Healthcare.

Most of the B2C healthcare opportunities focus on the upper-middle class and above. Typically, people can associate their health with some sort of valuable outcome (better focus, more stamina, better training, more energy, more work time, etc). This leaves a very, very small percentage of the population available to even consider paying for your product. The obvious exception to this is Oscar – but I’d argue that’s because health insurance isn’t really a B2C market.

Most healthcare plays end up as B2B because that’s how the money flows. This ends up being a bit of a chicken and egg type of situation. To be profitable, you likely need to build a revenue stream from a large employer, a healthcare system, or a health insurer. The challenge here is these types of places all want data that your product or service is actually going to benefit them. If you can’t fit your product into an existing procedure or approval path, like a surgery or FDA approval process, or run a clinical trial to demonstrate the value, you’re pretty much dead in the water.

This puts healthcare startups in a place where they need to find an alternative revenue stream to prove their product/service before transitioning to their actual target revenue stream. This is a huge challenge.

—–

There is a lot of money in support services for healthcare, but those companies typically get umbrella’d under a larger tag and simply serve healthcare as one of their industries. This means a lot of companies that benefit the healthcare industry aren’t tracked specific to healthcare.

I think it is fundamentally hard to construct a value proposition. Because of the insurance model, the party that gets a benefit is often not the party that pays the bill. You need a compelling value proposition for patients, providers, and payers all at the same time.

Combine that with a bunch of regulations and a super long sales cycle and you have a recipe for a tough market.

The industry stat is that it takes on average 17 years for the adoption of new technologies in the healthcare industry. That means that if there’s a new method or device to do something better there’s a solid chance that by the time it’s getting adopted by doctors it’s no longer the best. The underlying psychology of physicians is skeptical and risk-averse. A lot of physicians are so burnt out they don’t have the time to learn about new tech.

That’s just part of the battle. You can get doctors to use your new tech, but it’s even more difficult managing the reimbursement for your new tech. Getting major health insurance companies and Medicare/Medicaid to cover your tech is a grind. Then comes the costly and time-consuming clinical trials for the FDA to consider your tech safe and effective.

All of this combines to years and millions of dollars being required before you even get any sort of revenue. It’s counterintuitive to the traditional culture of move fast and break things.

Great question. I think … Health industry sector is too small. The total size of health industry is 1/10 the size of AC/DC power convertor size. The pond is too small. There is no big fish.

Leave a Reply

Your email address will not be published. Required fields are marked *