Selecting the correct KPIs and metrics to place on your dashboard is arguably the most important phase of a enterprise dashboard project. All too often, in its rush to step along the software development life cycle as quickly and painlessly as possible, the IT department and project sponsors don’t give enough time for the business subject matter experts to think about the best metrics to create dashboard charts, tables and alerts for.
This point came to mind as some Dashboard Spies and I watched a video by Avinash Kaushik, a big figure in the world of web analytics. If you don’t know this very talented man, please check out his blog on web analytics called Occam’s Razor. Lately he’s been talking about the idea of Web Analytics 2.0, but that is another topic (listen to his podcast on Traditional Web Analytics is Dead). Oh, heck, here’s a recent screengrab to whet your interest (and I might as well reveal the punchline – Web 2.0 leads you to understanding by examing Clickstream, Multiple Outcomes Analysis, Experimentation and Testing, Voice of the Customer, Competitive Intelligence and Insights. Here is the diagram:
Back to our dashboarding topic, here is the screenshot and link to the video about Customer Centric Web Decision Making, in which he posits the controversial idea that traditional web analytics is dead. He shows a dashboard-like slide with what he calls 8 x 6 x gazillion metrics:
The question and challenge is: In a world of countless KPIs and metrics, how do you measure the right things and achieve real insight?
That is the topic of his presentation (of course he applies his thoughts to the topic of web analytics, but it is insightful to watch regardless of the type of data you wish to analyze. Here is the video itself:
In this video, Avinash mentions the value of dashboards.
He has a great comment (note that I edited it below the screenshot for clarity – also, he was obviously not speaking to dashboard spies like us as he starts with the disclaimer “I don’t think you really care about this…):
I don’t think you really care about this, but dashboards are great – but do focus on the critical few. The only thing I’m going to take away from dashboarding things is this: Especially as you go higher up the chain of command, you don’t want them to think too much. If you have dashboards that have massive amounts of numbers on them, and it means that she [the CEO] has to think, that is not a good thing. Because if she starts thinking, she’s going to ask a lot of questions, and that’s okay, but you’ll consume a lot of time and you’ll have to give her all the background. It’ll take 90 days. So, as much as possible, present “pre-chewed” data so that people don’t have to think too much.
And here is the bottom line: Don’t over saturate your dashboards – limit it to 8 metrics per dashboard.
By the way, Avinash has a book out named Web Analytics: An Hour A Day that is definitely worth reading. Also, he has decided to donate the profits to charity!
Tags: Web Analytics Dashboard, Avinash Dashboard, Metrics, KPI