Top 10 Tips for Building a BI Dashboard

1. Before building your dashboard, spend a healthy amount of time determining what business information should be displayed on the primary display.

The best time to work through these issues is before you start working on the technology. Sometimes it is good to get an outsider that isn’t embedded in all of the issues of the day to help you work through this conceptually.

2. Don’t try to put everything in the dashboard

The dashboard should be the top level view for that person. It doesn’t need to everything imaginable in it. It takes work to get to the real important issues but narrowing it down will pay off in the end.

3. Make sure you have access or can collect the data you want to measure.

This seems to be a no-brainer but you’d be surprised how many things that are important AREN’T being tracked or recorded. Sometimes you need to add systems to get to the data you need.

4. Know the dashboard audience and understand what information they need.

It’s easy to make assumptions about what people need to make decisions. Make sure the people involved have input and a voice in the process.

5. Weed through the data forest to find the right data to use and know how to best present it to the users.

Don’t just throw a bunch of graphs at the decision makers just because they are available. Make sure you keep the concept simple and amount of information manageable. They can always drill down to get more information if they need it. Don’t clutter the top level.

6. Know your data. It’s easy to make wrong assumptions about the data so make sure you know exactly how it is being collected.

Dig into the data and make sure your dashboard doesn’t misinterpret the data that is using for the graphs. Bad data can cause huge issues down the road so spend enough time on working through the data and the collection process up front.

7. Build different dashboard for different departments. And make your dashboard modular so you can do this easily.

Not everyone needs to see the same information. Not everyone is authorized to see the same information. Do yourself a favor and build in modularity and a security model from the beginning.

8. Make sure the people that handle the data are involved in the process. If you don’t expect a few “gotchas” down the road.

When you talk to data people, you will always find out all of the details that most people don’t want to hear. They live there and they know the potential pitfalls of the data. Keep them involved.

9. Create a report to perform checks and balances on core dashboard data to increase credibility. Also, allow for drill down through the graphs for users but also for integrity checks.

Build in dashboard data quality by building a companion report to perform some checks and balances to increase confidence.

10. Track the decisions and changes made as a result of dashboard analysis; attach quantitative and qualitative benefits that result from the dashboard.

We all want to get value from our business investments. Tracking the decisions made using a dashboard and the outcomes from the decision can help you improve the quality.

And always be thinking about new data collections and their value to the dashboard. As business evolves, evolve the dashboard. The last thing the decision makers need is a stagnant out of date dashboard.