Data literacy programs: change decisions, not courses

September 3, 2026 · 5 min · Free, with full transcript

Does data literacy training change how a business decides anything? The position here is blunt: around 80 percent of programs teach people to pass a quiz, and Gartner surveys keep finding chief data officers who cannot name a business metric their literacy effort moved. Vocabulary like statistical significance is not the same as a category manager killing a promotion because the lift sits inside the margin of error. You come away with a working method: pick the recurring decisions, baseline them, and put the number in front of the decision maker at the moment of choice. Examples include a European retailer's delisting rule worth low single-digit millions a year, and a bank that halved model overrides by requiring written justification.

Listen to the podcast

4 min

Chapters

Key takeaways

  • Pick the three decisions your business repeats most often, write down how they are made today, and rebuild those instead of running a course.
  • Baseline each decision and measure margin, waste and override rate rather than course completions or attendance.
  • Put the relevant number physically in front of the person at the moment they decide, and let behavior change before understanding.
  • Require managers who override a model to defend the override in writing, which cut overrides by half in a month at one bank.
  • Read literacy platform case studies coldly: logins, courses finished and badges earned are not business metrics.
Read the full transcript

Host:Welcome to the Leaders Insights Podcast. Today's episode, data literacy programs that actually change behavior. Most data literacy programs are a waste of money, true or false?

Expert:True, and I'll be more precise. About 80% of them are training people to pass a quiz, not to make a different decision on a Tuesday afternoon.

Host:That's a big number. Where does it come from?

Expert:Gartner, the research firm, has been tracking this, and their surveys keep landing in the same place. A majority of chief data officers say their literacy efforts haven't moved a business metric they can name. When your own sponsors can't name the outcome, that's the tell.

Host:But people do learn the tools. They can build a dashboard now. Isn't that progress?

Expert:They learn vocabulary. They can say statistical significance, whether a result is real or just noise, at a dinner party. That's not the same as a category manager killing a promotion because the lift was inside the margin of error. Knowing and behaving are different muscles.

Host:So what do the good programs actually do differently?

Expert:They start from the decision, not the syllabus. At one European retailer I worked with, we didn't run a course. We picked 11 recurring decisions, pricing changes, store staffing, which products to delist, and rebuilt each one so the data was in the room when the decision was made.

Host:Give me a concrete example. What changed on the floor?

Expert:Delisting. Store managers used to drop products on gut feel. Nobody buys this. We put a simple rule in front of them. Units sold, margin, and whether the item pulled shoppers who bought nothing else. Within two quarters, they were delisting 30% fewer items, and revenue on that shelf category went up because they'd been cutting quiet earners.

Host:You didn't teach them anything though. You just gave them a rule.

Expert:Exactly, and that's the heresy. The best literacy intervention is sometimes to teach nobody and just change the default in front of them. Behavior first. Understanding follows. Most programs do it backwards and wonder why the classroom glow fades in three weeks.

Host:That sounds like you're admitting people can't be taught to think with data.

Expert:I'm admitting a course won't do it. Adults don't change how they decide because of a Tuesday workshop. They change when the new way is easier than the old way and someone they respect does it first. That's not pedagogy. That's plumbing and peer pressure.

Host:Peer pressure. So this is a manager problem, not a training problem.

Expert:It's mostly a manager problem. If a director overrules the numbers in every meeting to go with their gut, no amount of literacy training survives contact with that. I've seen a bank spend seven figures on an academy while the head of lending publicly ignored the risk model. Guess what everyone learned? They learned the numbers are decoration. Decoration you're contractually required to nod at. The fix wasn't more training. It was making that director defend, in writing, why he was overriding the model. Overrides dropped by half in a month. Turns out people trust their gut a lot less when they have to spell out the reason. Let me push on your ROI claim. Board-level

Host:return. Can you actually attribute a euro to this stuff?

Expert:Only if you set it up to be attributable, which almost nobody does. You pick the 11 decisions, you baseline them, you measure the outcome. Margin, waste, override rate. The retailer's delisting change alone was worth low single-digit millions a year. I can defend that number because we measured the decision, not attendance. And the vendors selling literacy platforms.

Host:The ones with the completion badges? They report engagement. Logins, courses finished,

Expert:badges earned. Their own case studies almost never show a business metric because their product can't touch the decision. Read those with a very cold eye. High completion, zero behavior change, is a common and expensive outcome.

Host:So if I run one of these programs, what do I do Monday morning?

Expert:Stop counting who finished the training. Pick the three decisions your business repeats most often. Write down how they're made today and put the relevant number physically in front of the person at the moment they decide. If behavior doesn't shift in a quarter,

Host:your problem was never literacy. Done. Want to know where you actually stand? Take the CDO self-assessment at MBA-training.com.

Read and go deeper