A version of this post will be published on SynBioBeta.
Let’s call it like it is. Being in startups and better yet being a founder has been the hot ticket for a while now. A lot has been written about how to start a startup, even in biotech, but I want to put a different spin on it.
Should you REALLY be starting that startup?
Stay with me here and don’t close your browser window thinking I am overly cautious and sound like your parents. I did found a startup (Sample6) and it continues to be an exceptional journey. BUT let’s be real: the experience is going to suck and at the same time is going to be the single most rewarding professional adventure you’ve ever undertaken.
So why am I saying this? Because as early career technologists like undergrads, grads, postdocs, faculty, or as employees at larger companies, we often have an incredibly poor view into the challenges of commercialization a.k.a. what it takes to build a business. It’s all worth it but you have to be absolutely convinced that the reasons you do it are worth crossing the valley of death to achieve it.
Let me constrain that a little more. This is not a post about consumer web, mobile, or even big data (substitute any software-based startup). This is about hard tech. Is it snobbish to think of the degree of difficulty of getting a startup in software as “soft” and in hardtech/hardware as “hard”? I don’t think so, rather it reflects the fundamental differences in the distribution of hurdles. The hurdle of getting to a proof-of-concept in IT is low, while the hurdles of scaling are high. The hurdles in hard tech are high early on, while they may be lower later on i.e. once you have approval in that field you have exclusivity for a good while.
But since this is a biotech blog and, more specifically, because the focus is synthetic biology, let me lay out what I want to highlight here. This post is about choices we as founders make and in particular in this case because we’re in biotech, it’s about the initial underlying technology. I say “initial” because in a shameless adaptation of whichever military general said it first: “No technology survives initial contact with the real world”. What it takes to build a startup successfully into an actual company is a technology that is exceptional.
Now you may think whatever it is you’re working on is the next big thing since sliced bread, but trust me: it most likely is not. Do not fall into the trap of trying to shoehorn your Ph.D. project into a startup. The worst that could happen is that you actually get funding and now have to embark on the long and painful road of trying to prove out a technology improvement that the market doesn’t care about that much.
So yes I’ve rambled on and told you a few things about what you shouldn’t do… so what SHOULD you do? What makes for a compelling product-market fit? That’s the question you have to answer. In this case for us biotechnologists it’s very often related to what the performance parameters of a technology are, but importantly thinking about it from a customer’s point of view sometimes yields more valuable realizations.
In some particular hierarchy, let’s call it Mike’s Ladder of Technology Awesomeness™, which we can debate (tweet at me @mkoeris or leave a comment), here they are:
Enable a behavior or workflow that wasn’t possible previously.
This one ranks the highest in my list because it’s applicable to not just therapeutics, but also diagnostics, medical devices, and industrial biotechnology. In particular, the story of Amgen’s anemia drug Epogen (EPO) comes to mind here. For those unfamiliar, their initial target market was the comparatively small segment of dialysis patients with severe anemia. By giving them EPO, patients experienced a life-changing (and rapid!) improvement in their condition — like being able to get out of bed and having energy to participate in normal daily activities as opposed to being in a state of constant and debilitating fatigue. If your technology does that for your target customer market, which by the way doesn’t have to be big, then that’s a really good start.
Better / faster (or whatever measure the market or your customer base cares about).
This one may really underlie or power the enabling behavior, but it may also just be straight up improvement in performance. Now let me qualify that statement. The performance improvement has to be BIG! I am talking 10X or more. Why you may ask? Because it’s hard to build technology and it’s hard to get customers to adopt it. At least be an order of magnitude better such that you won’t have that particular debate with your customers. You’ll have to work hard to convince me that your 13.2% improvement is meaningful. Biology doesn’t work very linearly; it’s fundamentally more logarithmic / exponential.
An example, this time from industrial biology and really also a shameless plug is Sample6. Our test for bacterial pathogens doesn’t need enrichment, so we’re short-circuiting the test protocol. That means we’re enabling our target customers (food manufacturers) to test their environments for the presence of bacterial pathogens within the production shift. Currently they have to wait one to several days for results to come in. That’s a meaningful change and valuable behavior shift.
Plenty of people think that by making something 10X, or better yet, 100X cheaper, makes for a good business. I would argue differently because you’re not able to capture the margin as well as you think. First your product has to be comparable or identical if you want to charge the same, and then you’re STILL going to have to give up margin to your customers to get them to switch. That’s a hard business to be in and requires capital, a big team, and a long time horizon.
Having said that there is a huge amount of opportunity for good product-market fit, but you need to be rigorous, too, about finding those opportunities. They don’t come to you — and they aren’t obvious at first. Iterate quickly and fail fast! The lean startup methodology that has been adopted widely in consumer web/mobile applications, and it’s worthwhile to start thinking about how that can be applied to biotechnology. My colleagues James Taylor, Euan Ramsey and I wrote about it earlier trying to frame the lean startup methodology in the context of biotechnology.
Last word of warning: too often a budding (bio)technologist is making choices based on defaults.
Think hard about the maximal value of your technology for each customer segment. Once you are convinced that you have found it, go hard after it!
Then email me; I’d love to both chew on it and see if you can convince me!