From Zero To Knowledge, Through Listening and Business Intelligence

How many of us have been through situations where the developer behind a project builds a product that suits their vision, rather than the needs of the end users of that project? That’s exactly what we walked into when hired to build out the next set of Business Intelligence dashboards for a healthcare company.

The previously hired company built a suite of dashboards across Finance, Operations, and Medical departments. In the 6 months prior to our group taking over, of the 15 dashboards created by this company, only 2 dashboards had been referenced, by a total of 3 users. Not quite the bang for the 6-figure buck, along with annual maintenance fees in the tens of thousands of dollars. Comments surrounding these dashboards focused around immediately knowing they would never be used and how upset leadership was because the company did not create any dashboards that were tied to the end users’ needs. To cap off these original dashboards, the information in a handful of them was incorrect, rendering it dangerous if they had been used.

Business Intelligence (BI) projects are more complicated than they may seem. Often times the work is described as creating pretty pictures, not recognizing the collaboration needed from all of the different players. This lends itself to Business Intelligence Developers, Architects, and Analysts to run with minimal scope and their own vision to what is needed.

I’ve found a successful BI project follows a different roadmap:

Step 1: Listen. If you can’t hear what the needs are, you are only creating to satisfy yourself, instead of others. From Quality to Operations, Finance to Sales, and so on, meetings between the project owners (aka the leaders driving what information is needed) and the creators should be open discussions about the team’s needs.

Step 2: Expand the project owner’s mindset surrounding the use of data. Often times, end users suggest they only need a few things, usually things that they already receive. Our ability to help these users understand that there is so much more that we can provide is key to turning a report into an interactive analytics showcase. Do you want to know how many assessments are out of compliance, or do you really want to know who is causing the assessment to be out of compliance, why the assessment is out of compliance, and whether there’s been improvement over the past 3/6/12 months regarding compliance?

Step 3: Let all users be heard. Interactive analytics dashboards are not to be built and left for dead. After going live, allow a 30-day enhancement window, where the full set of users have access to it and the ability to provide feedback. Most changes will be small, what we would consider agile-type updates…suggestions such as using different colors to make a chart easier to follow, providing pop-ups to show the next-level analysis, and drill-down capabilities to show deeper user detail. Once this window closes, the dashboards do not become out of mind. As time goes on, users will recognize that more and more information can be made available to help their decision-making. And the BI team should be there to help in any way possible.

Back to our situation of building the next set of Business Intelligence dashboards. Following these three steps above, our 15 dashboards had much different scalability than the original 15 built by the previous company. Ours were used daily, weekly, and monthly by over 40 users. The knowledge gained from these dashboards included 50% compliance improvement on filling out assessments, expanded sharing of overtime payroll information, operational efficiencies in turnaround times, and better depth of knowledge regarding employee turnover. It even led to our team running a training class to allow super-users to become creators of dashboards, rapidly expanding the company’s data footprint.

Know that your data has many important stories to tell. If you are not sure which one to start with, we are here to listen.