Evidence-Driven Decision-Making
Banasiewicz’s evidence-driven approach combines operational data, research, and other evidence to inform management decisions. It emphasizes computation and simulation as ways to examine current conditions and test possible actions.
THE IDEA
Using Evidence
Banasiewicz traces evidence-based practice (EBP) to medicine in the early 1990s and its later adoption in education, health care, social policy, policing, and management. He argues that synthesizing academic research, while valuable, gives managers insufficient guidance on using the operational data their organisations collect. His discussion of learning paradigms in Chapter 1 distinguishes observation, theory, computation, and simulation.
The proposed evidence-driven approach places more emphasis on computation and simulation. Sales records, production measures, customer feedback, and external data can help managers examine current conditions. Simulation can then explore how a proposed action might perform under different assumptions.
This approach combines context-specific operational evidence with findings from research and relevant experience. A general finding may help explain a pattern without answering the immediate question facing a particular organisation. Conversely, recent internal data may describe current conditions without explaining their causes or showing what will happen after an intervention.
Useful evidence must fit the decision. Managers need to examine data quality, timing, relevance, and the assumptions behind an analysis or simulation. Collecting more data does not by itself resolve uncertainty. The aim is to make better-supported choices and update them as conditions change.
When to use it
Adjusting marketing spend based on daily sales data and customer feedback analytics.
Using simulation to test the impact of supply chain disruptions before they occur.
Combining internal performance metrics with external market research to refine product features.
Integrating real-time operational data with historical research to set competitive pricing.
FROM THEORY TO EVERYDAY LIFE
In practice
Retail Pricing Adjustment
A retail chain combines point-of-sale data with a simulation to examine pricing options under changing demand. Research on consumer behavior helps inform the model, while current sales data helps check how well it fits the chain’s conditions.
Supply Chain Optimization
A manufacturer combines production records with external market data to simulate possible supply disruptions. The team uses the results to compare inventory options and checks the model as new information arrives.
A MOMENT TO REFLECT
Put it to work.
How can your organization systematically incorporate real-time operational data into strategic decisions?
What types of evidence are you currently overlooking that could reduce decision uncertainty?
In what ways can simulation and computational learning complement the traditional evidence you already use?
CONTINUE THE CONVERSATION
Source texts
- Evidence-Based Decision-Making How to Leverage Available Data and Avoid Cognitive Biases (Andrew D. Banasiewicz)