How often have you come across a dashboard that:
- has multiple pages
- doesn’t really tell the story you want it to tell
- remains largely unread by your intended audience, and/or
- is all of the above.
We’ve come across these many times before, and in fact we often get asked by clients to create them. Despite our best efforts, it’s often difficult finding a better alternative to use to demonstrate to them.
While the SiteCatalyst (Report and Analytics) dashboards are good at including charts and maybe a table and it’s pretty easy for anyone to pop a dashboard together, they can quickly become bloated with information, making it difficult to customise to what you specifically need. In many cases dashboards end up trying to cover too much – or in fact, pretty much everything, which turns into a sordid affair of multiple pages with pretty graphs, little context, and no recommendations. Phew – rant over.
Inspired by simplicity
I was recently inspired by Stephen Few’s work and his book on visualising data (Now you see it), and some of his examples particularly around dashboards and executive reports.
In fact, I was so inspired that I took up some of his principles and cracked open Excel and ReportBuilder and went to town on two different clients’ data sets, who had different types of conversions and interests.
Using Excel and ReportBuilder makes it easy to focus on performance, key metrics and trends, without all of the usual clutter and as it can be delivered on a single page, it’s easy to spot the things you need to perhaps go away and run some more analysis on.
Plus, it’s easy to refresh, easy to add in formulas for calculated metrics and perfect for adding things like trended standard deviations, anomalies, linear regression trends, etc.
And with Excel’s visualisations, you can display the data in a nice and concise consumable way.
A visually digestible summary
My 5 top tips for an executive summary would be that it should:
- include the metrics that matter to the business, digital and marketing teams
- highlight changes in metrics that should be investigated further
- be consumable “at-a-glance” – meaning that you shouldn’t have to decipher everything
- have comparisons so that you can quickly understand performance
- include a commentary – it’s far to easy for those not close to the data to misinterpret it.
