Great entrepreneurs start with a problem worth solving.
And the really great ones can articulate it clearly and concisely.
Many successful startups begin when an entrepreneur notices a problem and falls in love with solving it. In the 4thly startup accelerator, the very first assignment I give to each entrepreneur is to submit a problem statement that answers this question: What problem does your startup solve, and why is it a problem worth solving?
Often entrepreneurs who are given this assignment write long, meandering paragraphs filled with jargon, buzzwords, and meaningless gobbledygook. So I loved when Alex Polovian, in our current cohort, wrote this crisp problem statement for his startup:
Drug discovery is slow and expensive. Pharma companies spend billions on the process, and often get deep into development before rejecting a drug candidate.
And really, all he really needed was the first sentence: Drug discovery is slow and expensive. Boom. There’s a problem worth solving, right there.
Alex’s company, ZebraML, uses artificial intelligence and machine learning (with neural network concepts) to look at behavior patterns in Zebrafish for use in pre-clinical drug trials for humans, making the process faster and more efficient.
“A problem well-stated is a problem half-solved”.
-Charles Kettering
Traditionally, of course, rats and mice have been used for preclinical drug trials. Zebrafish, it turns out, have some big advantages. For example, in the case of an in vivo toxicological evaluation of a pharma substance, results can be achieved in just a week, yielding much faster cycle times than with mammals. It turns out that the genome and physiological processes of Zebrafish have nearly an 80% overlap with humans, and the genome is very well-studied and well-understood, making Zebrafish an excellent species for this work.
Alex has developed technology that tracks the behavior patterns of Zebrafish and learns from observing these patterns. So, for example, if you pour caffeine into the fish tank, the ZebraML technology observes and remembers how the behavior pattern changes. The more substances you do this with, the more his software learns. Then during drug trials the technology observes behaviors of the fish and is able to quickly pattern-match across all the other behaviors it has seen, yielding rapid and meaningful conclusions.
The key benefit to the drug discovery world is being able to quickly confirm or reject hypotheses. One of the reasons that biopharma hasn’t been able to adopt Silicon Valley innovation methodology is that it’s really difficult to “fail fast”. Sometimes a company will have a 5–10 year investment in a new drug before they realize it’s not going to work that well. So they are faced with the decision to either cancel the project or push it forward even though it’s not that effective. Early hypothesis validation and rejection using platforms such as ZebraML allows pharma companies innovate more efficiently, and get the right drugs to the market faster.
Alex’s startup is one of the most exciting ones in our current cohort. ZebraML is applying Alex’s expertise in AI, machine learning, and statistical analysis to help bring new life-changing treatments to people worldwide.
But it all starts with identifying a problem worth solving, and being able to articulate it crisply and clearly.
This article was merged into my new book, The Launch Path, now available on Amazon.
