Big bets require experimentation. But experimenting in large teams and larger alliances can be especially hard. The more diverse the group of partners, the greater the complication and the louder the welter of competing voices. Each group brings its own protocols, standard operating procedures, and approaches, which can sap the capacity to be innovative and experimental.
During a crisis, one of the hardest calls to make is when to embrace conventional wisdom, and when to try something new.
Experimentation can save lives.
Big alliances are essential to big bets. To innovate with and within this sort of alliance, you can:
- Scrutinize the existing protocols. Every field has standard operating procedures. Ask hard questions of them — and don’t settle until each is answered.
- Create an outside team of advisors. When you’re in the throes of a big effort, particularly during a crisis, you can have a hard time seeing things from any perspective. A call with friends and former colleagues with relevant expertise can keep you fresh and allow you to float and refine new ideas.
- Build a dashboard that supports experimenting. A clear method of measuring results in real-time. If you can measure what is working and not working, then new ideas can be tested and scaled.
- Start new partnerships with small conversations. Find out what can work and what can’t, testing the firmness of the ground one small step at a time.
- Fast data is better than no data, or even superior to slower data. If we’d been perfectionists about what data we used in Ebola, the epidemic would have been much worse. The best measurements you receive might come from a young kid riding a motorbike around town — it’s better than nothing.
- Don’t be afraid to fail. You can’t upend an existing protocol without a viable replacement. You can’t develop a viable option without considering some bad ones.
From the book:
During the 2014 Ebola epidemic in West Africa, I learned the importance — and risk — of testing new innovative solutions on the fly. Necessity is essential to invention, but there are lots of ways to mishandle a crisis. Remember to remain objective about existing protocols, seek an outside team to avoid groupthink, work with fast data (even if it isn’t perfect), and don’t fear failure.