How To Launch And Learn With Big Ideas And Small Resources
Properly integrating and utilizing health-related information generated by sensors and genomic technologies presents a tremendous opportunity and I had an idea of how to approach the problem.
By Irene Gabashvili (Founder, Aurametrix)
Once upon a time, the path from an idea to a product was possible only through teams of specialized experts spending months or years on prototyping or writing elaborate business plans. Then came the “think small” era, encouraging entrepreneurs to quickly launch on simple ideas and later figure out what to do with them on the go. The business climate is now changing again and entrepreneurs are encouraged to target more ambitious ideas. Should they?
I always liked big ideas. The kind like “it will be done in 10 years” so let’s play in the sandbox. Examples of my projects included artificial intelligence tools for geneticists in the early 90s and Next Generation Sequencing (NGS) in the first lustrum of the 2000s – although genomics had become less exciting for me after having realized that, for most people, nurture has a far greater impact than nature.
So I became interested in another area that was no less ambitious: how to measure the relative effect and consequences of nurture. Metabolomics – the study of small molecules that tell a story of what’s happening in the human body at every given moment – could be a solution. Analysis of expired air from the body seemed to be the least invasive and most convenient approach to identifying these small molecules. Far from being science fiction, breath diagnostics are starting to become deployed in a variety of research studies and boutique services.
I began collecting information and experimenting on my own. Breath sensors targeted for the consumer market have already been developed that detect hunger, bad breath and alcohol.
Could I create a prototype leveraging the data I had collected in my own independent research? I kept thinking about what I could achieve in a relatively short timeframe in the absence of a huge initial capital investment.
Then it occurred to me: No matter how advanced sensor technologies are, measurements of nurture will never be complete without observations based on human senses – all six of them. Interpretation of these observations would need to be supported by facts and relations underlying “expert reasoning”. Properly integrating and utilizing health-related information generated by sensors and genomic technologies presents a tremendous opportunity and I had an idea of how to approach the problem. In 2009, I started discussing it with numerous experts – medical doctors, food scientists, nutritionists, software engineers and business gurus. They agreed my approach could work, so I rolled up my sleeves and started implementing the solution.
My first prototype was a simple calendar-based app. It taught me about the importance of the user experience – even though at that stage I was the only user – and paved the way to the next version. Over time, the system evolved into something very different from food diaries, calorie counters and electronic health records. It was a way to analyze our little failures and victories, tie it to brain chemistry and cell metabolism, and contribute to a common understanding of human biology.
The road to launch was full of challenges: finding the right people, making decisions when there were many unknowns, probing for users’ feedback, testing, tweaking and all the time finding new ways to do things. The Aurametrix system was “learning” and evolving, and it had to be taught slowly and patiently. I was learning too, mastering new programming languages and databases, expanding my domain expertise into new realms and blending it with rigorous development skills. The product was alpha-released in the fall of 2011. Then I iterated it twice before publicly launching it in April 2012.
Since then, the Aurametrix user community has been steadily growing, proving that big ideas can be done with few resources given enough determination. Every individual who is able to use Aurametrix to discover something new about him or herself is our success.
My focus has shifted on how to take Aurametrix to the next level. Should I apply its algorithms to big problems in other domains? How about narrowing down all the little apps that could be built on top of the engine?
The possibilities are endless, and there are only a few winning strategies.
But isn’t taking risks the best part of being an entrepreneur?
Editor’s note: Got a question for our guest blogger? Leave a message in the comments below.
About the guest blogger: Irene Gabashvili is Founder of Aurametrix. Her work experience includes research positions at Stanford School of Medicine, New York State Department of Health and Hewlett Packard Labs. She holds a PhD in biophysics with postdoctoral training in bioinformatics and experimental techniques from a wide variety of disciplines. Follow her on Twitter at @igabashvili.