Dashboards Don’t Drive Decisions - Gethyn Ellis

Dashboards Don’t Drive Decisions (and that’s the real analytics problem)I have a new conference talk that made it debut at Tenerife Winter Sessions last weekend and is scheduled for another outing at the Birmingham Power BI Meetup Group #PBIBrum at the beginning of March. The talk is called “Data-Driven Storytelling with Power BI” and this is the most provocative statement in the presentation, and it’s intentional:Dashboards don’t drive decisions.If they did, most organisations would be operating perfectly by now.

They’re not short of dashboards. They’re not short of data. They’re not even short of beautifully designed Power BI reports. Yet the same conversations keep happening. The same issues resurface month after month. And the same decisions get deferred, diluted, or quietly ignored.

If that sounds familiar, it’s not because your data is wrong. It’s because your dashboards stop at information.

The familiar meeting that goes nowhere

You’ve probably seen this play out.

A dashboard goes up on the screen. People nod.
Someone asks a sensible question.
Someone else offers an interpretation.
A third person disagrees.

And then the meeting ends.

No decision.
No clear action.
No change in direction.

The dashboard did its job, technically speaking. It showed the data. The numbers were accurate. The visuals were clear. But nothing happened next.

This is the uncomfortable truth: showing information is not the same as enabling a decision.

Why dashboards feel useful but change very little

Most dashboards are built with good intentions. They’re designed to be comprehensive, neutral, and flexible. They show performance from multiple angles. They let the audience explore and “draw their own conclusions”. That sounds sensible.

In reality, it’s where things break down.

Humans don’t make decisions from charts. They make decisions from understanding. And understanding doesn’t come from being presented with options and hoping meaning will emerge. It comes from context, narrative, and relevance to a specific choice that needs to be made.

When a dashboard presents ten charts of equal importance, it silently asks the audience to do the hardest work themselves:

What matters most?
What should I focus on?
What does this mean for the decision I’m responsible for?

Different people answer those questions differently. That’s why dashboards often create debate rather than direction.

“Interesting” is not a success metric

One of the most dangerous words in analytics is interesting.

If someone looks at a dashboard and says, “That’s interesting,” what they usually mean is:

I can see patterns
I’m not sure what they imply
I don’t yet know what to do

An interesting dashboard might get attention. It might even get praise. But if it leaves the audience thinking “Hmm…” instead of “Here’s what we need to do”, it has failed its most important job.

A successful analytics product doesn’t just inform. It reduces uncertainty at the moment a decision needs to be made.

The missing ingredient: decision intent

The real issue isn’t visualisation. Its intent.

Most dashboards are built by starting with the data:

What tables do we have?
What measures can we calculate?
What breakdowns might be useful?

Decision-driven analytics starts somewhere else entirely:

What decision is currently blocked?
Who owns that decision?
What would change if we had clarity?

When you start with the decision, everything else sharpens:

Fewer metrics matter.
Visuals become explanatory, not exploratory.
The report develops a point of view.

This doesn’t mean manipulating the data or hiding nuance. It means accepting responsibility for guiding the audience, rather than outsourcing interpretation to them.

Why better dashboards aren’t the answer

When organisations realise their dashboards aren’t driving action, the usual response is to build more of them. Or rebuild them. Or redesign them.

But without a shift in thinking, you just end up with nicer dashboards that still don’t lead anywhere.

The constraint is not Power BI. It’s not DAX. It’s not modelling. The constraint is that most teams have never been taught how to design analytics around decisions, rather than around data structures.

From dashboards to decision tools

This is exactly the gap the Data Accelerator is designed to address.

The Accelerator isn’t about teaching people how to build more reports. It’s about changing how analytics work gets framed in the first place:

starting with real business decisions
designing reports with a clear beginning, middle, and end
reducing cognitive load rather than adding visual complexity
turning Power BI outputs into tools that actually influence behaviour

When teams adopt this approach, something interesting happens.

Meetings change. Conversations move faster. Decisions become clearer.

Not because the data is different, but because the analytics finally lead somewhere.

A simple test for your last dashboard

If you want to pressure-test your own work, try this:

What decision was it designed to support?
What action should follow from it?
Would two different stakeholders reach the same conclusion?

If those answers aren’t obvious, the dashboard isn’t finished yet.