Fact-based decision making works by turning a disagreement into a question that evidence can answer. Instead of asking who's right, you ask what you'd need to see to change your mind, then go and look.
The idea grew up in quality management, where it now goes by the name evidence-based decision making. The ISO 9000 standards behind ISO 9001 listed a "factual approach to decision making" as a core principle until their 2015 revision renamed it, and they ask that data analysis be balanced with practical experience. Fact-based decision making and evidence-based decision making name the same practice: settling a choice on what can be checked.
A fact here is something two people could check and agree on: last month's sales, a delivery time, a test result. Data is the raw material, and a fact is the part of it that answers the question in front of you. An opinion, however experienced, is a prediction about what the facts would show.
The commonest failure is quiet. A number arrives that agrees with what you already thought and feels like proof, which is how confirmation bias slips in dressed as evidence. The defence is to name, before you look, the result that would prove you wrong.
Personality quizzes often ask whether you base your choices on objective facts rather than emotional impressions. Answering yes says how you see yourself. The practice shows up on the calls where your feelings run strongest, which is exactly where checking feels least necessary.
For a month in 2001, two of us argued over a price that a one-week test could have settled. I was general manager of a chain of motorway service-area shops and certain we could charge more for the Nokia charger that dominated our sales, because customers bought it when they needed it there and then.
"You want to put the price… up?" Jim was just as certain a price rise would cost us sales, so I suggested we "do some research and revisit the decision at the next meeting", and the research became a month of me trying to out-argue him. A simple check of our average weekly sales would have given us a baseline, and one week at the new price would have answered the question for both of us.
When Jim finally agreed, we sold the same number of chargers and revenue rose by 50%: neither of us had decided on the facts, and I'd only happened to be right.