Iterating on the Due Diligence Process in Healthtech VC
Due diligence as a VC — what are the special considerations for health tech?
**************************************************************************
Note 1a: since first publishing this, Humane AI, despite not being a health tech company, would fall into this category: failed on each point, but used marketing/hype similarly to Theranos (i.e. matched closely to a legacy success, Apple). Humane AI was founded by former managers at Apple, neither of who were founders with exits; these managers co-opted the marketing strategy Apple used for decades to engineer a gut feeling, much as Elizabeth Holmes engineered it via the ‘CEO look’ Steve Jobs was known for.
Note 1b: Hype is different from a Scam. VCs are often paranoid about investing in scams, and assume hype is a signal for a scam. It isn't necessarily. Journalists also mix this up. Fair reporting means the public is better alerted to concerns that are ‘real’ vs not (e.g. misleading the public) vs those that are fairly normal in the world of business generally/tech specifically when it comes to marketing/branding where building excitement for the product is a normal goal.
Hype: lots of buzz and excitement that often doesn’t really track with the product per se over time. Objectively obvious with high #s of signups but high market/PR spend as well, and typically high CAC and high churn. Usually not sticky and doesnt move beyond early adopters (i.e. it ‘doesn’t cross the chasm’). These companies usually burn out after money runs out. Key point: Nothing fraudulent or intentionally misleading. Advertising and branding often ‘appears’ misleading when taken at face value (product X will lead to Y; long apparent waitlists etc) but disclaimers are there for a reason and have only become more extensive over the years. Clubhouse was a hype-y company. Humane AI (pin) may be in this category as well. Google Glass…etc (so not restricted to startups).
Scam: may or may not have hype but always has something fraudulent/intentionally misleading (inflated metrics or intentionally misrepresenting the contents or ingredients). Lots of examples in and outside of tech. Usually journalistic or legal investigations focus on the scams not the hype per se because hype in and of itself isn’t in the public’s interest — the public usually sorts through the hype over time anyway. Scams can only be proven through audits (financials), investigations (everything else). Where possible, VCs should audit financials before assuming one way or another that a startup is scam (vs hype). With A.I. this becomes easier to do. As with all research, data, and triangulating that data (i.e. not relying on self-report), matters.
When I think about the most successful companies in tech, there’s almost no hype at the beginning, and the hype follows after early adopters start telling their friends (for D2C; B2B would be different), so it’s like a slow build-up or a wave that peaks and usually stays there or decreases only slightly to a newly calibrated peak and steady state— the hype is organic in other words, and driven by the customer. In the U.S. alone, with 400m people, successful companies take a sizeable chunk of that market over the span of a year or so (exponential growth beyond early adopters). Facebook/Twitter/Apple/Amazon/Tesla even One Medical, and On Running are examples. For startups I’d include Oura and Maven. All have had slow buzz about an exceptional product that grows through this network effect, which legitimizes a high valuation (and eventual IPO). Some companies have hype at the very beginning and *do* end up having a product that matches this hype — Uber may be an example of this, and arguably Noom, and of course OpenAI/Anthropic/Perplexity, but this pattern is rare overall.
Note 2: Sam Altman’s ‘startup playbook’ is a core place to start, even for investors. However I’d place team ‘before’ idea — a bad team sinks a great idea, but a great team can iterate to find a great idea or pivot the business wisely with minimal internal friction. He also has some key thoughts re: competitors (distracting to hyperfocus on them), the importance of a great product etc.
Note 3: Financial services startup, Frank, founded by Charlie Javice, provides another case study where this framework may have helped prevent a sale to JP Morgan, who was [allegedly] defrauded due to fabricated user metrics (millions of users instead of under half a million users). This was an example of a ‘scam’ with no hype per se.
Note 4: Paul Graham’s essay ‘How to Raise Money’ is as key for founders as it is for investors. Especially this point (which links to my point 3 and 4): “Fundraising is a chore for most founders, but some find it more interesting than working on their startup…The danger of fundraising is particularly acute for people who are good at it. It’s always fun to work on something you’re good at. If you’re one of these people, beware. Fundraising is not what will make your company successful. Listening to users complain about bugs in your software is what will make you successful. And the big danger of getting addicted to fundraising is not merely that you’ll spend too long on it or raise too much money. It’s that you’ll start to think of yourself as being already successful, and lose your taste for the schleps you need to undertake to actually be successful. Startups can be destroyed by this.
When I see a startup with young founders that is fabulously successful at fundraising, I mentally decrease my estimate of the probability that they’ll succeed. The press may be writing about them as if they’d been anointed as the next Google, but I’m thinking “this is going to end badly.”
Hope the notes above help answer some of Qs I’ve received. I’ll update these as new insights come in.
Ok onto the original framework!
***************************************************************************
During my book tour so far, I’ve had several questions about the role of tech/startups and healthcare (and some of those early questions led to this post). The book ends with the promising aspect of startups, Big Tech and legacy companies. Over the last few weeks, though, I’ve had more questions related to the venture capital/investment side of this, which isn't covered in my book at all. I’ve had so many questions in fact, that if I write another book, one chapter may be devoted to health VC and how to iterate on the standard due diligence process to avoid the classic Emperor with No Clothes issue (VCs probably know if a bet is a good one, but groupthink/FOMO/hype can cloud this wisdom…leading to reliance on subject matter experts during the due diligence process). Recent SEC regulations make it crucial for venture firms to conduct prudent due diligence.
As I began dabbling in VC in 2022, I learned that my background in clinical medicine AND journalism were powerful reminders that triangulation and verification are key (‘trust, but verify, ideally with data and third parties’) aspects of due diligence.
Why? Because we need not look far (eg. FTX and Theranos, and to a lesser degree WeWork) to see that hype and selling a good story can go a long way. Especially when it comes to health, the story can be a tear-jerker (we saw this with Theranos/Elizabeth Holmes) because face it: health, illness, death, and dying are emotional.
Here’s the rub: a big raise, even for a startup that fails fast, with all the ingredients for failure, still makes money for the founders (‘golden parachute’) — founders win either way *unless* they’re convicted of fraud, which is a tiny proportion of them (eg Bankman-Fried, Holmes). As such, it’s imperative to iterate on how VCs conduct due diligence and take a rational approach with a very tiny splash of something else (which I’ll get to: the last two points of the framework).
To help avoid being swayed, while doing due diligence for a solo healthtech-focused GP fund in 2023, I began compiling some ideas into a framework that goes beyond the traditional “product-market fit, traction, team” triad. My approach is more akin to how a doctor might make a diagnosis — except the diagnosis is binary: success or failure (i.e. a good bet or a bad one).
This is the skeleton framework (I expand on this below):
1.Talent/size of founding team
2.Product IP vs. Market spend and any relevant early user feedback (be wary of no real IP and neutral or negative UX; IP/Market Spend correlation)
3.Hype (be wary of)
4.Rapid sequential raises/overblown valuations (be wary/ask why)
5.Culture red flags (retention as only one aspect)
6.Absence of key roles/absence of internal auditing (ability to trust the data ie internal validity of the reported data provided to investors)
7.Gut Feeling
8.Time
The goal is to help steer back to the ‘rational’ and avoid the pitfalls that faced FTX, Theranos (both on the extreme end), and others on the less extreme end (e.g WeWork, but also more recently Forward Health).
I used this framework for every single health tech startup that pitched me in 2023. Most didn’t pass the first two to three points. Two passed all 6 points but not the 7th/8th. Only one of them passed all 8, for me (then it was up to the GP).
This framework may not always lead to an immediate ‘yes’ or a ‘no’ to invest, but it does help nudge triangulation (seeking more data, verifying that data). It can lead as much to ‘finding a great investment’ to ‘avoiding a disastrous/legally fraught investment.’ I still use this framework when a VC colleague asks for my input on an investment.
To ground the framework in real examples, I’ve pointed to common patterns observed across multiple tech/healthtech companies and past high-profile failures such as Theranos and FTX and, to a lesser degree, Forward Health, Babylon Health, and WeWork.
I’ve listed these points *in order* for a reason (if the startup fails at point 1 there's no need to head to point 2 for instance — saves time and energy!). In general, I’ll make the decision ‘no’ once I reach the point number where the startup falls short, but for the sake of completeness (and to provide comprehensive feedback to the partners), I generally go through all 7–8 points for every company.
Also, you’ll see that the ‘market size’ isn't on the list. Market size for health, wellness, longevity, etc is huge, and growing. The market size for financial startups/crypto is similarly huge, and growing. It doesn't matter how huge it is — what matters more is whether the startup has a good chance of gaining market share of a huge market. This one point doesn’t matter for most startups in healthtech — the market is big, and we get it!
Note: the list isn’t perfectly written or formatted because it was not originally an article or a blog post — it’s simply a set of notes I’ve used for awhile now that may lead to a mindful pause, so to speak, and iterates on the most common and imperfect framework for due diligence in venture capital. I’ve fleshed out the points with examples for clarity.
1.How talented (and large) is the founding team?
Most VCs do background checks/due diligence on the CEO but not the founding team. It's best to look at each founder closely, and ideally interview them with a subject matter expert/peer in the industry (if the VC doesn’t have the expertise) because most CEOs liaise closely with their cofounders on all decisions
For the CEO especially — has there been a notable exit? If not, do they have special skills/talents/experience, ie the top 5%? (eg the Musks/Jobs/Bezos of the world). Being great at raising money is a good thing (but note Paul Graham’s note above) but even more important is being bullish for A player talent (seeking them, hiring them, giving them the resources they need to deliver results, and retaining them).
Have the other founders ‘earned’ their spot as founders? Are they in the top 10–15% of what they do? And/or have they had notable exits? What about managerial/leadership training and experience (this question is less important but a bonus if the first two are met)?
How many other founders are there? (a large founding team can slow down/impact key decision-making)
- size matters: in rare cases (eg Anthropic) a large founding team works because they’re all A players from a successful competitor. In most cases 2–3 founders are optimal (too many decision makers slows down progress, leads to infighting and other issues)
- ***this is an underrated question because B or C player founders don’t scale great companies primarily due to talent issues. Only A player founders are able to both recognize and retain A player talent (due to ego, perceived territorial elements etc — this links to culture and a deeper dive on this is here). Musk, Jobs, and Bezos are all A player repeat founders that retained A player talent, scaled and succeeded.
2.Product: IP (Theranos example)
(2a) IP matters as much as product-market fit
What “is” the product? Walk through it, ask ‘dumb’ questions, imagine the user/customer ‘journey.’
Does the product ‘depend’ on a third party (eg for processing, for facilities)? This is a higher risk bet (eg Theranos used a 3rd party analyzer for labwork), but sometimes a necessary intermediate step (e.g. Mac using intel chips).
Savvy design or UI is less important vs real IP (esp proprietary AI algorithms). This was also an issue with Theranos — design elements like a flashy website, futuristic offices, to even how the founder dressed show the influence of design.
Be very direct: What ‘is’ the IP here? What makes this *truly* different than your competitors?
The failure of Babylon Health for instance was one of overselling ‘proprietary AI.’ In 2025, specifically for any company claiming ‘proprietary AI,’ in healthtech, we can compare it with the next best user alternative, which is ChatGPT. ChatGPT today can best most physicians in diagnosis (recent research shows this), and is very good at incorporating clinical guidelines and lifestyle medicine advice. The updated models will only be better at both diagnosis and preventative care/guidance.
Rephrasing the question specifically for AI-heavy IP: is it 10–100x better than ChatGPT?
(2b) The users ARE the marketing & PR team (at first, through Series B, and sometimes to C at least)
To rephrase Kevin Hale (YCs) brilliant point, from way back in 2014 (!) about high market spend: “marketing is the tax you pay because you haven’t made your product remarkable…word of mouth growth is how a lot of the great companies grow.” Your marketing and PR teams, in other words, should be tiny: social media management is a good spot to invest in for instance, but not ads or billboards. Hiring one publicist, perhaps on a fractional basis, to hire inbound media requests (especially crisis PR items) is helpful, but save the agency hire for Series B/C.
The marketing and PR team should be the company’s first 100/1000/10000 customers! Using a sourdough analogy: that’s the product ‘starter.’ A solid marketing & PR team brought in later (eg Series B/C) simply amplifies this so it simply grows and sustains itself over time.
How do users really feel about the product? If it’s a great product, you’ll probably hear about it organically, in an enthusiastic way, often. I would also include here that VCs should check forums like Reddit for ‘user experience’ feedback regularly. Redditors are highly engaged and due to the anonymity, are more likely to share honest feedback (good and bad).
The ‘feature’ of a product is its UX, and, indeed, the most reliable way to scale is not through formal marketing but simply through word of mouth growth (i.e. ‘the product is *so amazing* that I’m telling everyone to get it’). A proxy for good word of mouth growth is online organic praise/discussion of the product (Reddit, X, etc). The Reddit principle applies less often to seed stage companies that have few users, but it’s notable, and a good sign, if seed companies have highly engaged happy users that share often online. One company I pushed forward at seed stage in 2023 had a highly engaged and dynamic Reddit community praising a health tech product and its impact on their lives. This was a helpful data point.
To rephrase Kevin Hale again: the biggest companies today (Apple, Google, Twitter/X, Tesla, Amazon) began as startups that didn’t spend funding on marketing early on. Ad-spend was allocated wisely (e.g. Apple’s Think Different campaigns) to simply amplify organic growth to reach the edges of the market only once the product was 100x better than the competitor’s: IP should be inversely proportional to market spend (high market spend early on often signals a bad product that isn’t growing; low market spend early on often signals a good product growing organically)
Great IP 1/∝ Market spend
Why Theranos raised money: they were vague about their IP and sold a story (via marketing and PR) very well to distract from not having IP, all which leads to the next point…hype.
3.FTX example (hype)
FTX disproportionately relied on hype/celebrity endorsements to ‘appeal to authority’. Celebrity endorsements can be powerful marketing tools, because celebrities are aspirational. Yet, without product quality and operational excellence, they risk being distractions. Did most of the celebrities chosen have expertise or experience in crypto? No. FTX employed celebrity influencer marketing (there’s the marketing bit again) to disguise not having a great product.
When it comes to healthtech: the authority should NOT be celebrities (or at least not ‘just’ celebrities), but rather the scientific community/physicians (conventional MDs), health policy experts, epidemiologists, etc who are respected for their clinical, scientific and research acumen as well as their ability to be a ‘visionary’/pioneering innovator (where is healthcare headed to ‘next’?). Note: both elements are needed (many leading health experts, especially in academic settings or legacy institutions have antiquated approaches to healthcare). Look to the founding team here, and the scientific/medical advisory board for clues.
Often I’d see small companies point to 2–3 celebrities as ‘endorsers’ but it’s not terribly difficult to obtain these. Celebrities are friends with other celebrities, so word of mouth about an ‘exciting investment opportunity’ can spread. An investment can be as tiny as $500, with a disproportionately large equity stake or advisory role if the celebrity agrees to be featured in a pitch deck for instance (and this is common/I’ve seen this in decks).
It’s increasingly more common for a founder to be a celebrity themselves (e.g. lifestyle/fitness companies founded by athletes; several celeb-backed makeup and alcohol brands) — again: access. This was a key part of FTX — excessive celebrity endorsements, which helped market the company, but the celebrities themselves were not clear how FTX ‘worked’ (which then became their defense).
This can also include access to influential individuals either within regulatory agencies, or who have power over those agencies. This was clear with Theranos, and why Holmes’ board was stacked with influential individuals with a track record in government and policy who could potentially have sway with government agencies (simply because: ‘we like doing business with our friends’ and ‘we trust who our friends endorse’).
When does hype work? When the IP is reliable, with a trusted market. Think: Apple’s launch events. These work purely because the anticipation is consistent year after year (or increasingly quarter after quarter), and users are accustomed to the product living up to the hype (i.e. the IP is reliable).
4.[for Series A, B, C specifically] Valuation & Growth.
When companies overemphasize their valuation and growth, without substantiating it through investor-dictated milestones, it can signal trouble. Forward Health failed late in the funding cycle (Series E) for instance, so it can occur at any time and it’s easy to miss.
For *any* startup raising very soon after a previous raise, ask founders:
Were there issues with releasing the previous tranches (were milestones not met for growth to allow for timely tranch funding release?)?
If the answer to this is vague, ask :
Were there external signals of poor growth — eg key partnerships dropped, sudden high marketing spend, referral incentives, quick partnership agreements with low impact on growth, media reports of product issues that impacted growth, users switching to competitors at a high rate, high apparent CAC/Customer Acquisition Cost?
Historically, as we saw with examples like WeWork, companies that over-rely on free memberships or heavy user incentives often face long-term retention issues. Other companies fail due to an issue with a key partner driving user growth, and so forth. Still, others fail due to a media report or in-depth investigation (e.g. Theranos).
Current investors will not typically openly disclose if tranches are being withheld due to growth issues/milestones not being met.
High valuations often mask internal dysfunction, especially in companies that haven’t met their core KPIs, but usually very few (e.g. current investors) are privy to this information. So, look for other signs. If previous lead investors are excited to share more publically, and vet the valuation, that can be a good sign. Sometimes founders themselves will share that the valuation and new raise was at the recommendation of an early lead investor — this is helpful information.
Then ask:
Is this a down round disguised as an up round? Why or why not? [note that current investors may support this as a down round label can be perilous for current investors]
Is this an extension round disguised as an up round? Why or why not? This matters more for media/PR spin. A Series extension can sometimes signal money mismanagement and is rarely a positive sign indicating growth.
Key question if there are signs of funding mismanagement: is anything in place to fix this/a contingency for the next round eg. hiring a CFO and ideally also a third-party audit?
5.Culture is a second product (link to my Medium analysis here)
Many startup failures are linked to having too few A players especially at the founder or early employee level — see also point 1.
As a VC, checking ‘culture’ is tough, & retention is only part of it (in a tough job market especially, it’s not a valid marker). Glassdoor reviews aren’t helpful (usually employees are warned they’d be sued/identities can be disclosed; usually only the most disgruntled employees dare post etc). Similar to Google Reviews, an absence of bad reviews means very little, and myriad good reviews, in close proximity to one another, also signal very little. Negative reviews, if spread apart over time from multiple users, are a helpful data point, but not on their own useful without further information and triangulation.
Reddit is very helpful for product feedback from users (see point 2), and rarely helpful for employee feedback (but often more helpful than Glassdoor).
The only real way to assess culture, which is linked to performance, is to have anonymous discussions with employees (and if possible, former employees, who are usually more honest). Due diligence helps with this, but not every VC takes the time to search through LinkedIn to ask these questions (most rely on their VC network or word of mouth). Similar to doing a reference check on an employee, I’ve always called former employees as part of due diligence on a startup: most will be honest.
If the individual in question has a track record of success/is an A player and/or already swiped up by a competitor it’s an even more trustworthy source.
One great Q to ask current employees, borrowed from Adam Grant: is: “What happens here that doesn't happen elsewhere” (and simply wait for the response).
One great Q to ask former employees, in a way that avoids placing them in a position to disparage their former company or risk divulging IP: “Would you personally invest in this company, knowing what you know now?” and “Let’s imagine 3–5 years into the future and this company has failed, what was the most likely reason?” [this question is especially helpful because it sheds light on product weaknesses/IP weaknesses, culture, and ethics eg flagging early FTX/Theranos/Babylon Health types].
As it relates to culture and HR: Are key roles unnecessarily consolidated that shouldn’t be after a seed round? (eg HR with COO; CFO with CEO)? Why might the founders be incentivized not to have an HR lead?
This leads to…
6.Note the absence of other key roles that provide oversight/internal audits [for Series A, B] — WeWork example
An experienced C-suite team (esp with experience in larger industries/corporations — Aetna, CVS etc) can do wonders for a healthtech startup that’s showing great signs of growth but needs stronger direction and leadership.
CFO is a role that can be helpful early, especially if it is a non-founder role (as they’re less likely to inflate numbers/fiduciary duty). Key question: Why might the founders be incentivized not to have a CFO?
What other oversights/audits are in place to guard against over-inflating metrics (growth eg conversion/purchases and financials)? Why might the founders be incentivized not to have oversights and audits?
[WeWork is a helpful example of a lack of oversights/audits. Frank, and the indictment of Charlie Javice, who fabricated user growth metrics to allegedly defraud JP Morgan, is another recent example of why third party audits are critical]
7.Trust your Gut Feeling
I wrote about gut feelings as they relate to medical diagnoses way back in 2018. As much as points 1 through 6 cover logic and rationality, gut feelings and intuition matter, but there’s a reason why this point is all the way at the end.
Even VCs with decades of experience and a track record of great bets, ignore their gut from time to time because the hype and FOMO is strong! :) Reports of FTX and Theranos, after the fact, point to this.
At the end of the day, it’s a gamble that can end up costing millions of *other people’s money* on top of, in extreme cases, fraud or misleading investors (eg FTX and Theranos) and legal costs.
The best advice I have here is to check in with yourself:
Does something ‘feel’ off? Does something ‘seem’ overhyped?
Does the founder ‘seem’ a bit unethical?
Does the feeling or pitch ‘remind you’ of a company that ended up being fraudulent, or simply failing miserably after a large valuation due to culture, product issues, lack of sustainable growth etc?
And so forth. Be aware that founders may try to ‘engineer a gut feeling’ by ‘matching’ closely with an established success via marketing and hype (e.g. copying e.g. how Holmes’ copied Steve Jobs’ ‘CEO look’ at Apple; how companies may copy marketing elements).
One, now publicly traded, company did this well: On Running. They appeared to use Apple’s marketing and brand strategy for their website and physical stores, but they could, because they could boast a phenomenonal founding team and IP (‘cloud technology’ coupled with beautiful design) — with a great product and team, marketing just amplified what was happening organically. As an organic user who would happily evangelize On Running whenever I could, I purchased stock the moment they went public in Sept 2020, and am glad I did :) I simply LOVE that company!
What’s the best way to get ‘better’ at trusting your gut? Test it! With public stocks, with people, with opportunities — is it calibrated well or do your biases interfere?
This brings me to the very last point…
8.Time
Take your time. Don’t jump at investing. Check points 1–6 diligently and rigorously. Score the company accordingly.
Then spend time on point 7 only, and simply sit with it (with time, point 8).
Give it a week, a month…whatever it takes. If it takes too long, it’s likely a no, or a ‘not now.’
Ignore the pressure you’ll undoubtedly get from founders — it’s not your money after all. It’s a bet, nothing more, nothing less. They will push for you to bet investors’ cash on them because they win regardless (see paragraph 4 of this post) — from experience this isn’t pleasant (and can end up with an unnecessary number of followups…in once case I heard from a startup an entire year later, persistently!) so be prepared.
Discuss the investment opportunity with experts, continue your due diligence calls (you can never do too many), and ask tough questions. While it's natural to confer with your VC colleagues about ‘whether they’ll invest,’ avoid groupthink — in fact try to engage your VC friends honestly on points 1–6 as well because they may have intel you don’t and vice versa (and you may help them avoid a bad investment or perhaps double down on a great one).
Lastly, if it remains a real toss-up: consider reducing risk in other ways (e.g. liquidation preferences to ensure the investment is paid back before funds go to founders) — if a founder pushes back on this, it’s notable. But really…why invest if it requires this much hedging?
From FTX’s rapid collapse to Theranos and WeWork, to even Babylon Health, the lessons remain clear: no amount of hype (especially be wary of tricksters) can replace the fundamentals of a strong IP-based product and promising (and triangulated/verified) growth metrics, to help make a decision to invest (or…not). Good luck!
