In response to 'Why we report data'

The thoughtful Dominic Hurst has recently posted about Why we report data. He's provided some excellent tips to meet the challenge of presenting data or user research that is understood and actionable:

  • Widen the ownership of the dashboard
  • Try just focus on the metrics that matter
  • Strip the dashboard of analytical jargon
  • Provide actionable insights

Beta dashboard used at Liverpool John Moores University
Beta dashboard used at Liverpool John Moores University(LMU)

I'd add to this the challenge of getting users to engage with a dashboard or report in the first place. You might share a regular dashboard to managers or stakeholders or have a screen in a lobby or team area. But how do you encourage people to read it, discuss it and act on it, rather than go 'dashboard blind'?

Dominic, in fact, provides the answer: Provide actionable insights. For me, that means:

  • KPIs, not metrics
  • Annotate

Avinash Kaushik defines a metric as a number (a count or ratio) and KPI as a a metric that helps you understand how you are doing against your objectives. So for me, the LMU dashboard is preferable to the Co-op example he cites, since the latter shows no KPIs or benchmarks.

Finally, I think we tend to forget that the choice of what and how to display data is an editorial choice and I believe we should make this explicit by annotating dashboards to explain trends, unexpected or special events etc. The effort to annotate is one of the ways analysts adds value and provides their audience with the actionable 'nugget'.