Dashboarding Requirements: In with the Old or In with the New?
When you start a dashboarding project do you start with what you have or what you want? The way you first approach your end user will have ripple effects throughout your designing process.
My recommendation is to start with a clean slate whenever you can. Not all stakeholders will allow or be comfortable with that approach, but the results are more likely to leave everyone happier and with a more complete product.
I've seen a number of engagements that start with the typical "send me what you have and we'll take it from there". While conventional wisdom would tell you to avoid reinventing the wheel, this approach ignores a few fundamental aspects of the dashboarding experience that should not be overlooked:
You're most likely implementing a new tool and new tools mean feature enhancements and restrictions. You might be putting yourself into a box by ignoring these changes. Your design will hinge on the flow and features that were leveraged in previous dashboards. For example, if you're used to putting together reports in Excel and move to a click driven platform like Tableau or Power BI, suddenly your creating static pieces of information that no longer need to be static.
People don't like change. By starting with what they've been using it acts as an anchor for any potential changes, which means they'll be much less likely to entertain changes. Inherently people are loss averse, taking away a graphic they've seen for a long time, whether it has any importance or not, you'll be setup for a battle you shouldn't have to fight.
The ETL process might be different, so you could be gaining or losing data. If it's a big change, you're more than likely losing some data at first, but this provides a great opportunity to ask the question "was that data necessary".
You're no longer asking what's important. You're just asking about what used to be important. At the core of this whole process should be a focus on using data for the right reasons and not just because it's available. By resetting you have an opportunity to take a different approach to the reporting/dashboarding process and make sure it's achieving results instead of just
Reports have a way of growing over the years and people will often forget why a metric or KPI was included. Sometimes they're just metrics that were supposed to be added for a short period and someone forgot to remove them or they might be metrics that have long been obsolete but the end user decided to keep it around "just in case".
Don't feed the beast! Hit the reset button and starting listening to the real business problems that are being solved instead of just creating the reports that are asked for.
If all else fails and you're forced to stick with what you have, take the opportunity to question whether aspects of the current reporting drive action or are just nice to look at. Sometimes getting rid of the noise is just as helpful as providing an answer.
How do you approach a new dashboarding engagement? What's worked for you and why? Contact me at any of the social media sites at the top of the page or write me from the Contact Page.