Last week I was speaking with my Head of Data Science.
We were deep in a discussion about Marketing Mix Modeling. Model structure. Elasticity curves. Lag effects. Calibration methods. All the mechanics that make MMM intellectually rigorous and commercially relevant.
It was a smart conversation.
But somewhere in the middle of it, we paused.
Because the model wasn’t the hard part.
The hard part was this:
Would anyone actually make a different decision because of it?
That’s when the real issue surfaced.
Most organizations don’t have a data strategy.
They have a comfort strategy.
The Illusion of Being Data-Driven
We say we’re data-driven. We say analytics drives growth. We say insight informs strategy.
And on paper, it looks true.
Dashboards are live. Models are refreshed. Reports are distributed. Performance reviews are scheduled.
But what actually changes?
Not the meeting. Not the slide. The behavior.
If nothing changes after the meeting, it wasn’t strategy.
It was insulation.
Why Comfort Wins
Real decisions are uncomfortable.
They require:
- Stopping something that once felt safe
- Reallocating budget from one team to another
- Admitting a previous investment didn’t pay off
- Taking personal ownership of downside risk
That creates tension.
And most operating systems are designed to reduce tension, not create it.
So instead of designing for decisions, organizations design for reassurance.
They ask for more analysis. They commission deeper diagnostics. They refine segmentation. They optimize attribution.
But they rarely force the harder questions:
- Who owns the decision?
- What behavior must change?
- What are we willing to stop doing?
- What changes tomorrow morning?
Without those questions, analytics becomes explanation.
Explanation feels productive.
But explanation without commitment is just a more sophisticated form of comfort.
The View From Inside the Analytics Team
If you’re a new or mid-level analytics leader, you feel this tension every day.
You inherit scopes that were sold before you arrived. You manage expectations you didn’t set. You are measured on quality, timeliness, and utilization.
You are not measured on whether better decisions actually get made.
You see recommendations stall because decision rights are unclear. You watch leaders nod in agreement and then revert to precedent. You feel the gap between insight and action.
And your performance review won’t ask:
- Did decision velocity improve?
- Did accountability become clearer?
- Did behavior shift?
It will ask:
- Did you deliver what was promised?
That’s not a data problem.
That’s an incentive problem.
Designing for Decisions Instead of Comfort
Growth does not follow information.
It follows commitment.
And commitment only happens when the environment demands it.
Designing for decisions means:
- Naming the specific decision your analysis supports
- Making the trade-off explicit
- Assigning clear ownership
- Linking consequences to action
- Measuring behavior, not just accuracy
It means redesigning meetings so they end with commitments, not conclusions.
It means shifting the scorecard from shipment to execution.
This is not about building better dashboards.
It’s about building decision systems.
Lead Anyway
It may be easier for me to say this.
I’ve had seniority. I’ve had air cover. I’ve had the authority to push back when comfort disguised itself as strategy.
Not everyone has that luxury.
But leadership doesn’t start with title. It starts with framing.
Even if you don’t control the incentive structure, you can:
- Name the decision.
- Ask what stops.
- Surface the risk.
- Force clarity.
Even if it makes the room uncomfortable.
Because comfort compounds.
And growth does not.
The Real Question
If your analytics function disappeared tomorrow, would decisions change?
If the honest answer is no, you don’t have a data strategy.
You have a comfort strategy.
And comfort never built competitive advantage.
If you’re ready to move from reporting systems to decision systems, download the Decisions by Design Handbook and start redesigning the environment where choices are made.
Because the future of commercial analytics isn’t better explanation.
It’s engineered commitment.
