Data-driven decision making works by agreeing what you'll measure, watching it, and letting what it shows choose between your options. The aim is to replace the loudest opinion in the room with a number everyone can see.
Its weak point is the choice of measure. Most organisations track what is easy to count and what makes sense from the inside: jobs completed, tickets closed, targets hit. A customer judges you on something else, often the waiting nobody measures.
So start with the result you care about, and only then ask what data would show it moving. Turning that into a hypothesis you can test, with the expected number written down before you look, stops the data being read to suit the story you already prefer.
It overlaps with fact-based decision making but works at a different scale. Fact-based decision making settles a single disagreement with a checkable fact; data-driven decision making builds the habit of measuring before deciding, across many decisions and over time. The two often work together: a dashboard raises the question, and a single fact settles it.
Some people prefer to call it data-informed, to keep judgement in the room when the numbers can't see everything. The difference matters most when the data is thin, new or measuring the wrong thing.
Every measure on our dashboard was red, and when my boss asked what the core of the issue was, all I could offer was "I don't know." We were managing a client's repair jobs in progress, and the metrics inherited from previous management tracked what mattered to the business rather than what a customer waiting for a repair would notice.
So we mapped the same data onto the customer's journey, step by step from the job being booked to the repair being done. The core of the problem stood out like brake lights on a dark night: the time it took service providers to order parts, a long delay that none of the old measures showed.
When we focused on that delay, repair times came down and the client was happier.