Data Strategy Guidelines for Lean Startups: Deriving Learnings Faster

Abstract

Lean Startup Methodology is an alternative to the conventional plan-and-execute business approach. New businesses following the lean startup methodology are inherently experimental and data driven. Emphasis is placed on deriving learnings to verify or reject value and growth hypotheses—this is, how value and growth is generated. This thesis claims current literature on lean startup methodology provides insufficient guidance on how to establish a data strategy, that is, the approach to collect, store, and use data to generate actionable insights and learnings.
This study is undertaken following design science research methodology. An IT artifact featuring three data strategy guidelines designed to facilitate learning in lean startups is developed, explicated, and evaluated. The artifact is designed using justificatory theories and concepts from authoritative authors on lean startup methodology, agile programming, and business intelligence. Three evaluation episodes are carried out to evaluate the artifact’s performance in terms of utility to the entrepreneurial community and knowledge contribution to the body of research on lean startup methodology. First, a summative evaluation is conducted by using a self-completed survey whereby volunteer entrepreneurs are asked to assess if the artifact increases the entrepreneurs’ knowledge on how to derive learnings. Second, a formative evaluation is performed by conducting a semi-structured interview with a business intelligence manager to identify areas of improvement. Last, a proof of concept is carried out to reveal its feasibility and whereby practical insights and pitfalls are documented.
Contribution to knowledge is derived from the novelty of the artifact and its attempt to bridge the research gap by offering a concrete approach to the challenge of collecting, storing, and using data in the context of lean startups.