Analytics software is helping National Grid improve how it collects late payments, a crucial exercise for the utility provider whose customers owe it millions of dollars for gas and electricity. The software helped National Grid escape "Excel hell," in which analysts manually analyzed data and built ineffective PowerPoint decks to share their findings with business leaders.
The presentations fell flat because they didn’t allow people to look at the data from different angles or perform ad hoc analysis, says Rory Abbazio, National Grid's director of IT business enablement. Abbazio knew the organization needed a more dynamic tool so he turned to Tableau, a provider of self-service visualization software, for help.
Rory Abbazio, National Grid's director of IT business enablement.
National Grid’s project is one of many instances in which companies leverage analytics to glean insights they wouldn't otherwise find in their troves of business data. It’s a booming market. Gartner estimates the market for business intelligence and analytics software will hit $17 billion through 2016 and grow to $20 billion by 2018.
Analytics helps collections agents target deadbeats
When Abbazio joined National Grid earlier this year to revitalize the company's field service management operations he made an alarming discovery. IT was dumping data extracted from Business Objects, MicroStrategy and other BI tools into spreadsheets, which collections agents took with them on their routes so they knew who owed money and how much, as well as the age of the debt.
Yet National Grid had no formal strategy or calculus to recoup the tens of millions of dollars owed by some of its 3.5 million New York, Rhode Island, and Massachusetts customers. "I realized that we had a huge data problem," Abbazio says.
That’s when Abbazio turned to Tableau, which he had implemented while working in IT at Boston Scientific, where he helped sales staff perform ad-hoc queries. Using Tableau, which renders information in bar graphs and other charts, the company sliced and diced customer data by credit score, median income, geographical data and other attributes.
It then built a scoring model to rate customers on their ability to pay. For example, it learned from Tableau that customers who were anywhere from zero to 30 days late and who owed $100 were most likely to pay. The model showed the collections team which customers to prioritize.
Abbazio then funneled the Tableau insights into Salesforce.com's field service software, enabling him to "geocluster" customers to optimize the routes agents should take to collect money or shut off power for those who the data suggested had a low probability of paying.
Sign up for CIO Asia eNewsletters.