In the following post, D’Amore-McKim School of Business Professor Koen Pauwels explains how lean startup methodology can alleviate potential big data complications.
In the 2016 Global Perspectives Barometer, about 800 Leaders of Tomorrow cited “innovation blindness” (the inability to recognize the need for a decision and staying passive in a quickly changing environment) as the most substantial risk for established companies in today’s fast-paced markets*. Indeed, legacy firms from Best Buy to IBM are struggling to adapt their business models to successfully innovate in the face of greater competition from both local and global startups. Instead of simply urging to “innovate like a startup,” the best advice recognizes both the uncertainty legacy firms face and the substantial resources they have over startups. Our forthcoming paper, “Combining Big Data and Lean Startup Methods for Business Model Evolution” in the Academy of Marketing Science Review, provides an integrated process for corporate innovation learning through combining the lean startup methodology with big data.
By themselves, the volume, variety, and velocity of big data may trigger confirmation bias, communication problems, and illusions of control. However, the lean startup methodology has the potential to alleviate these complications. Specifically, firms should evolve their business models through fast verification of managerial hypotheses, innovation accounting, and the build-measure-learn-loop cycle. Such advice is especially valid for environments with high levels of technological and demand uncertainty.
Following are some examples of big data learning challenges and our advice learned from lean startups:
Data challenge = Volume (Increasing amount of data)
More opportunities to confirm prior beliefs, while discarding disconfirming evidence. Use big data to come up with hypotheses to test through experimentation.
Data challenge = Variety (Different types of data)
Increased complexity of data and analyses make it difficult to communicate insights for decision making. Using incremental innovation accounting breaks down the complexity into simpler processes and metrics, which can be communicated more easily.
Date challenge = Velocity (Real time data)
Increased sensation of predictability due to immediate and rich performance information. Loop in build-measure-learn, which adds agility and ambidexterity and helps challenge data-driven illusions and mitigates unprecedented crises.
In general, we propose that firms should build on the lean startup methodology to help adapt their business models while at the same time leveraging the resource advantages that they have as legacy corporations. Have you seen any examples of how this was done?
* Neus, A., Buder, F. & Galdino, F. (2017). Are you too successful to digitalize? How to Fight Innovation Blindness. Marketing Intelligence Review, 9(1), 30-35.