Driving insights from Data
One things we hear from clients almost daily is “I have loads of data but its not being used, what I need is insights!”
The quandary of business today is that technology is delivering data in increasing amounts, faster than its helping us get the insights from it.
One area that is starting to make a difference is the idea of software powered “dashboards”. A dashboard is pre- programmed to bring the data together in a way that is easy to digest, so that patterns can be seen and issues identified.
Let me be clear, a good dashboard is more than just a web portal. The developers should have already thought about the key applications and what users most often need to look at, and have brought this to the fore. It should have graphics that land the key points visually and impactfully. It should have exception reporting to identify the trends or changes that are most important to the business. An intelligent dashboard can be user and application specific. Thinking about category management data, if you are a Tesco oriented manager you should be able to view a “Tesco view” of the data automatically(for example).
We have made one major change to our reporting that is in line with this thinking. Our data has always been organised by category and by retailer on our dashboard but now we are introducing a new “application orientation” but application area. For example, in a category review with Sainsbury a user might want to understand what our data is saying about Assortment, or Pricing, or Innovation. We have re-cut the entire data set by these kinds of topics creating dashboards that put all the data relevant to this issue on one page. We are calling them “Storyboards” – because each one creates a story on a page.
Of course the holy grail of dashboard engineering is to put multiple data sources into one tool. rather than going to the dashboard for panel data, the dashboard for sales data and the dashboard for Shopper Intelligence, you see it all in one place. Perfectly doable. but needs to be developed from the user backwards not the data outwards.