Adaptive Machine Learning: The future of ecommerce fraud management

Online fraud is seemingly an increasing risk for businesses and organizations. Read our report for insights about how fraud prevention decision-makers are looking towards automated fraud management and machine-learning (ML) solutions.

PayPal recently commissioned a research study “Adaptive ML: The Future of E-Commerce Fraud Management," conducted by Forrester Consulting to understand the role of fraud management in today’s global firms. The findings revealed that as the world continues on its digital transformation journey, it is critical for organizations to have a strong enterprise fraud management (EFM) and security strategy.

Most firms have invested in tools and third-party partners to bolster their fraud management program, but it may not be enough – 97% of global fraud prevention decision-makers at e-commerce companies experienced fraud in the past 24 months.* In order to keep up, and transform with the world around them, fraud prevention decision-makers are looking toward automated fraud-management, and machine learning (ML) solutions.

Key findings from the report

  • Firms use both rules-based and EFM solutions. Adaptive machine learning is the future, as 83% of respondents noted it’s pivotal to their companies’ e-commerce fraud strategy.*
  • On average, it takes over a week to resolve a fraud incident. Respondents indicated that their firms lack the tools and resources to reduce fraud.*
  • Firms are looking toward machine learning and automation. 81% percent of respondents indicated adaptive machine learning solves their companies’ top EFM challenges, and 64% plan to invest this year.*

Ecommerce fraud management is critical for business success

Customers demand a frictionless, secure experience, and companies know they’ll lose business if they fail to provide. Fraud prevention specialists surveyed stated that EFM was critical to improving customer experience, increasing revenue, expanding into new markets, and achieving their organizations’ digital transformation initiative.*

Leaders in fraud prevention rely on EFM solutions and vendors

Firms understand EFM is a business must and have invested in solutions and third-party vendors to support their EFM strategy. The most valuable components referenced in this survey are rules-based software solutions (74%), machine-learning software solutions (70%), and advanced auditing and reporting capabilities (64%).*

When it comes to vendors, respondents noted looking for someone to help with the heavy lifting. The most important considerations when selecting a vendor are the ease of implementation, the ability to provide advanced technology, and the overall partnership and industry expertise.*

Respondents whose organizations have partnered with a fraud management vendor experienced more accurate identifications of fraud, a better customer experience, and better fraud resources.*

Fraud incidents are common and result in negative business outcomes

Fraud management strategies aren’t foolproof. 97% of respondents have experienced fraud in the past 24 months. The most common types were chargeback, account takeover (ATO), and transaction fraud.*

Once fraud is identified, respondents said it can take weeks to resolve. This has resulted in negative customer experience, lost sales and revenue, wasted labor resolving issues, and an inaccurate view of the customer.*

Companies need to bolster their fraud resources

Firms are experiencing fraud incidents because their EFM resources are inadequate. Respondents indicated that their organizations’ biggest challenges include:

  • Their solution flagging too many false positives or negatives.
  • Lack of resources to resolve issues, including headcount and employee skill sets.
  • Current tools are difficult to use.

Without proficient tools and resources, firms will continue to experience – and struggle to deal with – fraud incidents.

Automated machine learning solutions will decrease ecommerce fraud management challenges and improve business outcomes

Digital transformation will solve most EFM challenges and reduce fraud attacks. The shift to automated ecommerce fraud prevention will create more accurate and efficient reporting and management.

81% indicated that adaptive machine learning would solve their companies’ top ecommerce fraud management challenges, and 64% plan to invest or increase investment in EFM this year.*

Respondents expected these investments to decrease operational and compliance costs, their organizations’ overall risk, and response time, all while increasing customer satisfaction.

Conclusion

The pandemic accelerated digital ecommerce and moved fraud management in retail from nice-to-have to absolutely essential.

Our study found that:

  • Fraud incidents are common and result in negative business outcomes. ATO, payment fraud, and identity theft are on the rise. Merchants must protect against fraud to preserve business growth and provide a positive and frictionless customer experience.
  • Merchants struggle to scale fraud teams to address increasing chargebacks. Firms need to invest wisely. Instead of hiring more investigative, reactive staff to struggle through disputes, firms need to invest in preventative automated systems and processes.
  • ML fraud management is now essential. Our survey shows that automated, adaptive ML will decrease time-consuming and resource-heavy fraud management, and improve business outcomes overall.

Download the full study to discover what fraud specialists think of adaptive machine learning and how it can transform fraud management.

The Future of E-Commerce Fraud Management (PDF)

The Future of E-Commerce Fraud Management (PDF)

Was this content helpful?

Related content

Sign up to stay informed

Share your email to receive the latest enterprise updates, top stories, and industry reports.

*Required fields

We use cookies to improve your experience on our site. May we use marketing cookies to show you personalized ads? Manage all cookies