Have you ever gotten home from the grocery store, unloaded all the bags and then looked around and said — what am I going to cook for dinner? That’s how we increasingly see the state of data in businesses today. The focus is primarily on obtaining and acquiring data: How do I get data, how do I structure it, where do I house it?
All of these are important questions and considerations, of course. But for businesses to become truly data-driven — as most are striving to do — they might be better off starting not on the supply side of the equation (grocery shopping) but with the question of demand: making dinner.
In other words, what’s the reason you want the data in the first place? Which business outcomes are you trying to drive? What decisions are you going to make? How are you going to activate the data?
Only then should you ask which data you really need. This could be in the form of personalization you need to generate, a capability you want to activate, a safety issue you want to mitigate, a performance metric you want to improve, a regulation you need to comply with, machinery you want to keep operating. What is the business outcome that — using data — you can optimize?
By starting with the demand question, it’s easier to hypothesize which data may be potentially relevant. From there, you can start to work upstream to determine how and where to get the relevant data, which is primarily obtained in two ways: captured in the ordinary course of doing business or acquired through an external channel, whether it’s a broker, an exchange or another partnership.
What we see every day and in every industry is that once the business outcome is defined, and the data asset is prepared and released, more opportunities become apparent — and in some cases, obvious — to leverage the data and turn insights into action.
This is changing our concept of the virtuous circle of data. When we used to talk about this, the conversation started with: Collect all the data you can, then use it to get insights, and the more insights you get, the more data you need. That’s changing in a way that shines the spotlight — once again — on business outcomes, not the supply of data.
Today, as data has become nearly unlimited in nature — and more techniques emerge to extract more meaning from it, the data’s potential keeps growing. So, instead of data leading to insights, which leads to a greater need for data, we contend that exceptionally well-defined outcomes require data, and once the data is in hand, even more and better outcomes can be realized. This is the new virtuous circle.
Let’s take our work with Network Rail, Britain’s principal rail infrastructure owner. The business had specific, measurable objectives: lower costs, enhanced safety and reduced travel disruptions by predicting and preventing maintenance issues, prioritizing work streams and minimizing the time rail workers spent on the tracks.
With the demand side established, it was time to move to the supply side. The rail operator didn’t have the data it needed on-hand to be able to make decisions and activate these decisions in service of the business outcome. However, it was clear that with current technology and leaning into the objectives, it was possible to collect data, in the ordinary course of their operations, by adding sensors to its more than 12,000 connected assets, such as track circuits, signal power supplies and switches.
Through data collection and AI-driven analytics, the business can now monitor train traffic, rail conditions and passenger patterns in real-time throughout the day. Ultimately, the desired outcome is to cut accidents and incidents to zero; real-time decision making (demand) based on a stream of real-time data (supply) and asset monitoring will enable that.
While not directly tied to revenue-generation, there is tremendous business value being generated by this embrace of the virtuous circle. Now that the business has more data assets than ever, it can start to brainstorm on what else can be done with them.
For instance, with the ability to see traffic and passenger patterns, it could start redoing train schedules to either cut costs or boost revenues, efficiently redeploy assets based on needs, and provide varying levels of service based on the opportunities that get unlocked.
Another example is a major construction retailer that had a simple desired business outcome: drive high-volume traffic (particularly contractors) into its stores vs. those of a competitor. The hypothesis was that by better understanding where people tended to go before and after a visit to its stores, it could offer targeted cross-promotion deals.
With the business goal set, it was time to research the options for data supply. There was no way to do this in the normal course of business; the retailer’s app captures location information only while it’s on, and for the vast majority of people, that’s only when they’re at home or in the store, not after or before they leave.
So, the retailer partnered with a mobile gaming company that leverages a phone’s real-time latitude and longitude data. It discovered the destination most often frequented by customers before and after a visit was a fast-food restaurant. That meant it could trigger campaigns that offer co-promotions between the restaurant and itself, specifically timed to encourage customers to return.
And once the virtuous circle begins spinning, the retailer could expand on these insights in other ways, such as offering fast-food delivery with customer orders, determine pathing within a store for optimal cross-selling and potentially lease out space within its own under-utilized parking or retail space.
In our minds, this is the new frontier for data modernization: mapping supply to demand rather than vice-versa and then — importantly — getting the right data to flow to where it’s needed to make decisions, when the decision needs to be made in service to the in-market activation of these decisions.
Over the last several years, we’ve been building these cathedrals to worship at the altar of data. Now it’s time to break the data free from these structures and let it start evangelizing its value by flowing to where it can actually make a difference. We need to stop focusing on where data is going to be stored and start working to ensure it can get to where it needs to go.
Businesses today are in a race not just to obtain more data but to figure out how to use it better, put it in the right place and then find even more ways to use it to fuel more innovations. This “activation” of data must be the new competency for data-driven businesses everywhere.
Bret Greenstein leads Cognizant’s Global Data Practice, focused on helping Chief Data Officers transform their businesses through Data Modernization.
Jason “Kodi” Kodish is Vice-President and Head of Guilds within Cognizant’s AI&A leadership team.
View editorial post