By now, we all know that we are supposed to be making data-driven decisions. Yet, most leaders still struggle to use data effectively. I think that one of the main culprits is that we’ve become data-intensive rather than data-driven.

I recently worked with a leader to streamline his customer satisfaction reports. His report had eight columns of metrics and thirty rows of data – one for each department in his business. That’s 240 discrete pieces of information on one report. No one can processes and keep track of that much information at once. The report was too overwhelming. It was typical of most reports that I see.

I often see reports fall into three traps:

1) They mix outcome data with diagnostic data.

2) They are focused on information rather than action

3) They try to have the most amount of information in order to answer ANY question that the user might have

The result – reports with hundreds or even thousands of data points that are almost impossible to sift through for insights.

I worked with this particular leader to correct these three problems.

Separating outcomes and diagnostic data
Take a look at your dashboard the next time you are in your car. These days, there are only a few indicator lights: oil pressure, engine temperature, and check engine. The lights don’t tell you what’s wrong. Their point is to alert you to take action. When you take the car into the mechanic, he or she plugs a diagnostic tool into a port under your steering wheel. That tool pulls the detailed diagnostic codes needed to determine the problem.

By separating the outcome/action data from the diagnostic data, car manufacturers have greatly clarified and simplified the driver’s decision-making.

The same can be true with business reports. You don’t need diagnostic and outcome information on the same report. It adds too much detail. At best it distracts you from making a decision. At worst, it obscures the decision as you focus your attention in the wrong places of the report

The leader with whom I was working needed to know upon which departments to focus his attention. That was his main decision. Once he knew where to focus, we could use the detailed diagnostic data and KPIs to figure out the specific problem.

To make his initial decision (where and how to focus) there were only two pieces of data that he needed: whether the department was hitting its target and whether the performance increasing or decreasing. So, right from the start, we were able to eliminate six columns (180 pieces of data) from the report.

Focusing on decisions
Reports shouldn’t be designed to simply inform, they should be designed to drive actions. The closer your report aligns with your decisions, the more effective it will be. Often the data on a report is two, three or four steps removed from the decision. To move your report closer to the decision, ask yourself what criteria you use to determine action (and what those subsequent actions would be).

Joe chose one of three simple actions in response to his department’s performance:

• Take major and immediate action
• No immediate action needed but needs to monitor performance
• No action required at all

These actions might sound overly simple. But, they represented the first set of decisions he made in determining how and where to act. All of the extra data on his reported, clouded his ability to triage and make this simple set of decisions.

His decision rules and actions are summarized in the following chart:

We now had the data aligned with his decision process.

Focusing data
Trying to have one report that can answer any possible question reduces the report’s effectiveness at answering any specific questions. There’s too much data to absorb. The best reports have least amount of information needed to take action.

Our final step was to redesign his reports to align his data and decisions. His new report simply placed each department into the appropriate box on his decision model.

Now, instead of having a report that forces him to wade through 240 pieces of data to determine an action he has a report where the actions pop right off the page. Of course, this report doesn’t provide the specific action but it does tell him where and how to focus. Notice that although this report is based on metrics and data, it doesn’t show them. It doesn’t need to at this point. His question is where and how to focus. This report answers that question simply and effectively.

Because the report is organized around his decision criteria (as opposed to the lower level information), he can also quickly assess the overall health of his organization:

• Only one item in green – This raises a flag right away.
• The majority of items are declining in performance – Could be a sign of overall organizational health or of some significant issue
• “Immediate and major action” has the most departments – We probably need to make some major organizations changes

In addition, he can more easily find patterns that might help drive his actions:

• Do the departments in the lower left have anything in common? (e.g., same manager, using similar programs, etc.)
• Do the departments in the upper right (if there was more than one) have anything in common? Can we leverage anything from those departments?
• Are the departments that are showing improvement ones that we been working on fixing?

Of course, all of those insights and questions could have come from the original report. It would have just taken a lot longer and the relationships wouldn’t have been as visible.

There are a lot of bad reports out there. We’ve become accustomed to reading and accepting them. But we don’t have to.

Many leaders tell me that they don’t have enough time to think about their data, their actions or about the “right” reports that they need. That is probably because they are spending too much time pouring through unnecessary data and reports that don’t help them make decisions.

The target/trend report is just one type of decision-based report. You can create your own reports that are much more specific to the types of decisions you make. The key is to work backward from the decision and action. When you do, you will have decision-driven data that will be much more effective for supporting data-driven decisions.