Dashboards have become buzzwords in the analytics world and inside many a credit union business unit. They have been lauded as the magic pill for credit unions who want to use their data to better understand their members, drive value, and meet specific business goals. However, if you ask any analytics practitioner, they’ll tell you the same story over and over: they put in hours of time and effort to build a dashboard that exactly meets the specifications from the business only to have that work languish somewhere on the internal network.
And while no one technique can ensure a dashboard will deliver value, there are some key steps that can materially improve the success of your business dashboards.
Start Early, Ask Questions
- How well do you know your audience? While you may know the data, the end user knows the business. Before you meet with them, do some preparation work on possible data sources you might need, and identify questions you know you’ll need answered. When you meet, start with broad questions and work down to detail.
- What do you want this dashboard to do for you in a perfect world? Use your first meeting to get to know the big-picture needs of the team who will use the dashboard as well as the details of how they’ll use it. It can also spur conversation on content that is easy for you to include but they may not think to request.
- What are the next steps you want to take? Someone on the front line of a credit union will have different dashboard requirements than a board member reading a summary. Do they need to take an action you can build, like a link to another specific report? Or just understand big-picture trends? If they want to export data to Excel for additional analysis, find out what that analysis is you may save them valuable manual time.
- What detail-level data needs to be included? While the goal of the dashboard might be to understand call center volume, they may want to be able to see that information in more granular detail. By CSR name? By using a unique identifier of some kind?
- Who will use this besides you? This is an opportunity to make sure you don’t miss additional use cases that could make the dashboard easier to understand for many users.
Confirm Hands-On Specifications
- Were there any specific visuals you had in mind? Some people can be very particular they may even have another image or dashboard in mind when making the request. Consider what they already have access to and even bring examples of other dashboards with you; sketching visuals on a napkin or iPad can be a useful way to get on the same page with others before wireframing the project.
- What filters would you like on this? Should any data be excluded before it reaches the dashboard? (A colleague once spent days working up a mortgage dashboard and included HELOCs, which weren’t of immediate interest to those who would be reading it.) This is an easy way to save time and demonstrate skill with your audience.
- How fresh does this data need to be? Find out the time frame of the data they want shown by default and how often they want it refreshed. This gives you an opportunity to make sure your dashboard provides relevant insights while saving development time if less customization is necessary. In addition, it creates an opportunity to manage expectations if you are not able to provide near-real time information.
- What are your KPIs? The best dashboards take you through the most important information first, then allow users to drill into what they need to know. Identifying specific KPIs to highlight is important for visual impact and can reveal additional questions you need answered and help focus your data sourcing tasks.
Click here to learn more about how PenFed’s Mortgage Analytics team learned to iterate and build more actionable dashboards in the process.
PenFed’s mortgage team went through multiple versions of their core dashboards, working to incorporate best practices, highlighted here. Click to enlarge.
Manage Expectations & Build Relationships
- Start smaller than you think. The above covers a lot of ground but don’t be afraid to break someone’s dream dashboard or suite of dashboards down into pieces. What is most important and could be most useful to have right away? Identify this and work to secure something you can deliver sooner rather than later, then plan and communicate how it will be added to and when.
- Set expectations for iteration. Focusing on value-added low hanging fruit can jumpstart a business unit’s relationship with business intelligence or analytics. Ensure you have clear, feasible deadlines communicated on both sides. In your kickoff meetings, emphasize that back-and-forth dialog is a requirement and that you’re there to get them what they need.
- Meet before your final reveal. Checking in along the way is essential to avoiding a disappointing dashboard unveiling. Depending on your credit union’s preexisting data quality, you’ll want to make sure that your calculations or sources align with any the unit requesting your work is already using. If they see numbers that immediately differ from their own, it erodes trust and creates delays and more work. Similarly, you can use check-ins to ensure that the dashboard’s design is suited to the audience’s needs and preferences. Always be ready to iterate and build in time to have these necessary back-and-forths, while also being cognizant of the requestors time.
- Future-proof your work. When you hand off a dashboard, keep in mind any training the users may need, including how to access the dashboard, use filters, export, and request help or changes from you as the creator. Create a reminder for yourself to check in on how the dashboard is performing for the requestor or even schedule a meeting in the future to confirm how well it’s working. These conversations can help avoid abandoned dashboards and create opportunities for the team to make a given dashboard more useful.
In this webinar, Kelly Gage of Red Canoe Credit Union ($1.1B, Longview, WA) talks more about best practices for business intelligence requests.
As relationships between business units and their embedded analytics experts or any stand-alone analytics team develops, iteration will be required. Patience and open communication are key. Along the way, the report builders will get better at fielding questions, asking questions, and keeping projects moving. Dashboards can be powerful tools for any credit union, fed by anything from a set of key Excel documents to an advanced set of data warehouse sources. Regardless of where your credit union stands in their analytics journey, dashboards, when done right, allow the business to come together around one source of data-driven truth and see, explore, and ask more questions about data that can reveal new and improved ways to deliver value back to the member.
