Overview
This series examines newly founded, independent German start-ups using big data analytics (BDA) in management decisions. The study follows 3,395 ventures and 12,261 firm-year observations: firms founded in 2010–2015 were observed through 2018, with exits and insolvencies tracked through 2023. It associates BDA adoption with higher costs and sales volatility, lower survival, and higher insolvency probability; among surviving adopters, it finds higher sales, stronger employee growth, and a greater likelihood of receiving venture capital. The series explains this trade-off through complementary skills, infrastructure, management routines, runway, security, and integration. Its conclusion is not that analytics guarantees productivity or growth, but that BDA can amplify an organisation’s existing capacity and expose gaps before returns arrive.
Learning outcomes
- BDA adoption is associated with higher costs, greater sales volatility, lower seven-year survival, and higher insolvency risk among the German start-ups studied.
- Among surviving adopters, sales and employee growth are higher, but the evidence supports scaling more clearly than higher labour productivity.
- The clearest upper-tail amplifier pattern appears in employee growth, not a proven larger sales effect for high-performing firms.
- Tool access is not capability: value depends on skills, infrastructure, decision ownership, runway, security, and feedback.
Keywords
Preview every chapter · Unlock full text, audio, and video from chapter 4
PlansFAQ
Does BDA improve German start-ups’ performance?
Not uniformly: in the study, BDA adoption is associated with higher early costs, sales volatility, lower survival, and higher insolvency, while surviving adopters show higher sales and employee growth.
Why can BDA increase both risk and growth?
Because BDA requires complementary skills, infrastructure, management routines, runway, and decision integration before returns are certain; those capabilities can magnify costs and volatility early and scaling capacity later.
Does higher sales among surviving adopters mean higher productivity?
No. The sales association becomes statistically insignificant when current employees are controlled, so the evidence supports organisational scaling more clearly than higher sales per worker.
Is adopting an AI or analytics tool enough to build capability?
No. The OECD’s 2026 selected SME sample shows broad AI access can coexist with mostly isolated, shallow use; integration requires ownership, training, security, and feedback.