🎉 Happy 2025! By now, experimentation has proven its value across countless use cases. Yet, as Davis Treybig highlights, a significant gap persists between the leaders of experimentation and everyone else [1].
📘 In the January/February issue of Harvard Business Review, my co-authors—Iavor Bojinov, David Holtz, Ramesh Johari, Martin Tingley—and I delve into why so many companies struggle to unlock the full potential of experimentation, and what to do about it.
🔍 Our key insight? While companies are often ambitious about experimentation, they frequently stumble when it comes to turning that ambition into tangible results. Data scientists, eager to bring rigor to decision-making, often become the champions of experimentation. Their demand for rigor is vital for moving from intuition-based to data-driven decisions.
⚠️ However, as organizations embrace experimentation, this same demand for rigor can backfire. Data scientists risk becoming gatekeepers—fiercely guarding rigor but inadvertently becoming a bottleneck in experimentation throughput. By holding too tightly to the reins, data scientists limit the scale and impact of experimentation.
💡 What’s the solution? We propose a shift in mindset: instead of acting as gatekeepers, data scientists should empower others to experiment themselves, leveraging advances in modern experimentation platforms to ensure rigor while enabling broader participation. When we “teach others to fish,” everyone benefits:
1️⃣ More experiments, more innovation: We know that scaling the number of experiments accelerates innovation [2].
2️⃣ A clearer big picture: Evaluating experimentation at a team or program level helps identify where to focus additional resources.
3️⃣ High-leverage work: Freeing up data scientists allows them to tackle more strategic challenges [3].
For the details, check out the full article, whether online or in print
https://lnkd.in/gg-6F4VQ
[1] Davis Treybig, The Experimentation Gap: https://lnkd.in/gBtPaZHY
[2] Eduardo Azevedo et al., A/B Testing with Fat Tails: https://lnkd.in/gcUiUddP
[3] Eric Colson, Beyond Skills: Unlocking the Full Potential of Data Scientists: https://lnkd.in/gcJY_Nxn