Build a Dashboard to Drive Decisions

We have a dashboard for that.

I remember asking for some information and was told there was a dashboard for that. What I got was basically a 9x9 matrix with the exact same chart repeated over and over again. Yikes. I had no idea what I was looking at.

We need a dashboard for that.

How many dashboards have you seen that have so much information, you have no idea what to do with it.

But before we start talking about KPIs, charts, colors, or filters, ask one simple question.

What decision will you make from this information?

Dashboards facilitate decisions.

A dashboard is a tool.

Like any tool, it's only valuable if it helps you accomplish something.

I've seen organizations and individuals invest months building beautiful dashboards and reports ... six months later, no one was opening them.

Not because the data was wrong. Because no one knew what to do with it.

Start with the decision, question, or action.

Sometimes we revel in getting information. It makes us feel secure. But if you're not doing anything with that information, it's not useful.

Here's what I mean: Instead of figuring out what information you need to build a sales dashboard, figure out what decisions you need to make to grow sales. 

• Are our customers satisfied?

• What sales channel needs attention?

• Is our new marketing campaign working?

When we begin with the decision, the dashboard becomes easier to design.

Instead of asking, "What should we measure?" we start asking, "What information helps us make this decision with confidence?"

Good dashboards tell a story.

Our brains are wired for stories. We naturally organize information into narratives because stories provide context. Without context, our brains dump information.

A good dashboard doesn't overwhelm you with data: It answers questions.

Think like a reporter.

What causes you to read an article? The headline.

The first rule you learn when you're a reporter is to start with the most important information first. Then the supporting details, then the context.

That's exactly how dashboards should work.

1. Who is the audience?
2. What decisions/actions/questions do we need to address?
3. What information do we need?
4. Do we have that information?
5. Do we trust that information?

Let's walk through an example.

Imagine your executive team has the same conversation every month.

"We're behind our sales target. Why?"

Someone suggests building a sales dashboard.

Before anyone starts pulling reports, ask a few questions.

Who is the audience?

The executive leadership team. Marketing, Sales, and Finance are all going to use this dashboard to understand what's driving growth.

What decision, action, or question should this dashboard support?

The question isn't, "How are sales?"

The question is: Why aren't we meeting our sales goals?

That's a much better starting point because it tells us what we're trying to understand.

What information is needed?

Now we can identify the information that supports that question.

• Are we generating enough leads?
• Are enough leads becoming opportunities?
• Are opportunities turning into customers?
• Are certain products, industries, or territories underperforming?
• Is our average deal size changing?

Notice what we didn't do.

We didn't ask for every sales metric available.

We only asked for the information that helps answer the question.

Do we have that information?

Now it's time to look at our systems.

Do we capture lead sources?

Do salespeople consistently record lost deal reasons?

Do we know which campaign generated the opportunity?

If the answer is no, we've found a bigger issue than the dashboard.

Do we trust that information?

Finally, ask whether the data is reliable.

Does Sales define an opportunity the same way Marketing does?

Are opportunities updated consistently?

Are people maintaining the CRM, or are key details living in spreadsheets?

Because if you don't trust the data, you won't trust the dashboard.

Making it make sense

Now that we know what question we're trying to answer, we can decide how to present it.

Start with the most important information. If someone only has thirty seconds to look at the dashboard, what do they absolutely need to know?

Then provide the information that helps explain why.

Finally, include the supporting details for people who need to dig deeper.

The order matters.

If your audience has to hunt through twenty charts before they understand what's happening, the dashboard is making the story harder to follow.

Just as important is choosing the right visualization.

Don't pick a chart because it looks interesting. Pick it because it makes the answer obvious.

If you're showing a trend over time, a line chart lets people see whether performance is improving or declining.

If you're comparing products, regions, or salespeople, a bar chart makes those differences easy to spot.

If you're trying to understand where work is getting stuck, a funnel quickly shows where things are dropping off.

The best dashboards don't win design awards. They reduce the effort required to understand what's happening and decide what to do next.

When someone opens the dashboard, they shouldn't be asking, "What am I looking at?"

They should be asking, "What should we do next?"