Reducing risk: how governments can use data analytics to prevent fraud
It is estimated that fraud costs governments and taxpayers around US$300m a day. At a time when governments are facing mounting pressure to provide better services to citizens, and with smaller budgets, the need to combat fraud and reduce losses is more urgent than ever – analytical techniques can help
Organisations across sectors entered 2023 more aware that the economic, political and social environment does not remain static, and that the conditions under which they operate may change relatively frequently. They have learned to pay attention to what happens, to be resilient, and to make decisions in near real time to prevent and head off risks that may negatively affect their business outcomes and the trust placed in them by customers and citizens.
One of those risks is fraud – a problem worsening as criminals become more sophisticated, better organised, and begin operating on a larger scale. Indeed, it is estimated that improper payments and other forms of fraud cost governments and taxpayers around US$300m a day.
At the same time, governments are facing contraction or no growth in official budgets and mounting pressure to provide services to citizens more efficiently, making the need to combat fraud and reduce losses to public coffers more urgent than ever.
Governments and other organisations are generating a huge volume of data and accelerating the use of technologies that work with it – more than ever before. This is good news and bad news: it gives governments more opportunities to fight criminals, but it also gives criminals more opportunities to find holes in data, security or processes. Governments therefore need to use a range of analytical techniques to combat fraud.
The challenges
First, let’s look at the data-related challenges when it comes to combating fraud – these include organising data, transforming data into insights, and integrating those insights into how people and processes work and make decisions. For most governments, data management remains the biggest challenge.
“Government institutions need to take an analytical approach to preventing fraud and see it as a comprehensive process. This should encompass generating and gathering data, applying analytics, putting information and insights into the hands of decision-makers, and integrating analytical decisions into automated systems. In short, applying the use of data for decision-making.”
Analytics needs good data. It doesn’t have to be perfect but good enough for the objectives it seeks to achieve. The reality is that many organisations continue to manage their data in spreadsheets and files distributed across systems. Even when data is digitised, it may have inconsistencies and be duplicated. And sometimes the most useful data will come in different formats and types, both structured in databases and unstructured as text.
Having electronic data and automated processes creates the necessary efficiencies. This will also enable governments to provide citizens with the services they need in a more timely manner and with less bureaucracy. To speed things up, certain processes need to be electronic and automatic, often with little or no human oversight of transactions and decisions.
Making changes
Successfully combating fraud requires actionable data, basic and advanced analytics, a versatile technology platform, and a responsive organisation. This involves making operational changes to the way people work, as well as cultural changes to the way they use information.
The human factor is critical – particularly people’s skills. Capabilities are required to organise and integrate huge amounts of data, select and experiment with the best algorithms to analyse the data, create models and integrate analytical results into people’s workflows, as well as to design and manage technical architecture.
As digital transformation progresses, artificial intelligence and machine learning also have a highly valuable role in recognising criminals and preventing instances of fraud before they happen.
Financial organisations, for example, are turning increasingly to AI and machine learning to combat fraud and protect customer assets while reducing false positives. This is critical when digital payments are made in seconds, leaving less time to monitor fraud in transactions happening simultaneously across multiple channels and countries.
To respond, financial institutions must adopt a risk mitigation model that includes response, detection, prevention, evaluation and management. Likewise, they need to pay more attention to the process of authenticating their customers, such as through 3D authentication and security protocols, unique passwords, biometric security measures and the use of tokens.
Successfully detecting and preventing risk and fraud requires banishing manual processes and spreadsheets and integrating analytical engines that use advanced analytics, strengthened with artificial intelligence and machine learning.
Governments that do this can speed up their response to risk alerts according to the severity of each case and provide researchers and analysts with the necessary insights to respond – in turn protecting citizens and the public purse, as well as reducing non-compliance.
Upcoming! Webinar: Fraud Fingerprints in Your Data – 30 May, 1pm EST (hosted by NHCAA)
Fraud, waste and abuse (FWA) is an ever-growing threat in healthcare and costs providers, payers and patients more as each day passes. Join SAS industry experts Tom Wriggins, John Maynard and Jason DiNovi and learn more about how to stem the growing tide of cases.
Topics include:
Healthcare fraud
Payment integrity
Programme integrity
FWA
Health equity
Behavioural analytics
AI in healthcare analytics
Register for the webinar here
Click here to find out more about how AI and data analytics can help prevent fraud, protect public services, maintain trust, and ensure taxpayers get value for money.
About the author
Héctor Cobo
Héctor Cobo is vice president at SAS, overseeing Mexico, Caribbean, and Central America. He led the start-up of SAS’ Americas Offshore office and has nearly three decades of experience in the business intelligence and analytics sector. Hector has participated in critical initiatives including credit risk, money laundering prevention, card fraud prevention, ATM fraud prevention, credit approval models (Scorecard), Data Warehouse, Balance Scorecard, campaign automation, customer analysis, customer churn prevention, consumer segmentation, operational risk, and debt models.
Héctor has a degree in computer systems engineering from the Technological University of Mexico with a master of business administration and operations from Texas McCombs School of Business.
About Partner Content
This content is brought to you by a Global Government Forum, Knowledge Partner.
