Cleveland Police has partnered with Experian to develop a new data management system that has helped the force save £250,000 per year by turning its chaotic collection of records into a single citizen view for each individual's file.
The police force turned to Experian after the effects of 25 percent cuts in staff and the challenges of recording accurate information at crime scenes left it with a messy set of data. The result was a collection of over 1.8 million person profiles for a population of around 600,000.
The vast quantity of duplicate and incomplete data forced officers to search through numerous records to find full information on an individual, and what they found often lacked a full view of their criminal records and their threat to the community.
"The risk of that in relation to policing is that we are going to fail to have a holistic view of individuals within our community, and that goes for offenders as well as victims of crime," Maria Hopper, data protection officer at Cleveland Police, tells Computerworld UK. "We're not only keeping track of those who are offending, but we also risk losing sight of vulnerable individuals also."
Recent high profile cases show the dangers of inadequate police data. They include the Soham murders, where police took two weeks to uncover previous sexual allegations against school caretaker Ian Huntley, and the Fiona Perkins investigation, where repeat calls to the police force were missed because they weren't connected to the individual's record.
Cleveland Police has experienced similar issues when the common names were difficult to find across hundreds of thousands of records. The tried to use third-party data matching tools to reduce the number of duplicates, but missing information in the records made them difficult to match.
"It matched what was already in the system, but we knew what was in the system was rubbish, and we didn't have a solution to stop the rubbish going in," says Hopper.
They searched the market for a supplier that could develop a solution, and the only company that said it could provide what they needed was Experian.
Experian has been heavily investing in data science to create new business from its vast quantity of data and extensive history of analytics. In the US, for example, the credit reference agency has transaction records on 200 million American citizens going back 16 years, which it claims to be 99.9 percent accurate.
"Experian had come in to deliver an awareness session to Cleveland Police in relation to their data quality arm, and some of the products and suggestions that they put forward made me believe that they could assist us," says Hopper. "They also had really good quality data, something that we obviously did not. So I reached out to Experian and said, this is what I'm looking for, could you deliver something, and they believed they could deliver a solution to all three elements that we wanted.”
Experian used its verified reference data to clean Cleveland Police’s existing data and then cut out any duplicate information in the reference agency’s data management platform.
It replaced the duplicates with a single holistic view of each individual and their interactions, which helps the force to understand any threat they pose and respond more quickly and appropriately. They named this database of person records the "Golden Nominal".
Over 100,000 records that couldn't be cleaned because they lacked quality information were given date of births that allowed them to be automatically merged.
The new system helps the call handlers who create 90 percent of new records by creating a call management system that linked through an API directly into the Golden Nominal. They can use this system to search through old records or to create new ones, speeding up response times and improving public safety by providing a clearer view of criminal information.
Cleveland Police claims that the system has annually reduced the growth of new duplicate records by 100,000 per year, save £250,000 annually, cut 10 years of man hours, and reduced daily security clearance times from 1.5 hours to 15 minutes.
"We have amazing people working for us and they are the lifeblood of policing, but data is what makes them tick, and if you haven't got a healthy state for both data and your people, then you're not going to achieve what your objectives are," says Hopper. "So police forces must wake up to understand how important data is within the organisation."