What is needed to control warranty costs and prevent fraud?
Our previous blog “The Dark Side of Warranty Management “ highlighted the scale of warranty fraud. Depending on the source, warranty fraud is estimated to be 3 to 15 percent of the total warranty cost. In individual companies, the amount of warranty fraud can be much higher. Defrauding warranty providers can also be the primary activity of some individuals and companies as illustrated by the following examples:
A Massachusetts man defrauded Cisco Systems of more than US$15.4 million. He used a loophole in Cisco’s SMARTnet program that provides customers with technical support, including advance hardware replacement.
A cluster of service agents was set up in order to generate claims. Up to 97 percent of their claims were fraudulent, while the volume of real repairs was negligible. The total annual billing of this service agent cluster was in millions of dollars.
In addition to fraud, warranty costs may be inflated e.g. due to poorly drafted contracts, employee sloppiness or competence issues.
In one case, a technician changed an expensive display module in 70 percent of the repairs, while this would have been needed in only 1 percent of the cases.
Warranty control framework for fraud avoidance
What can then be done to detect and prevent fraud, and to avoid other excess costs? A number of methods are available. Traditionally, claim validation has been the main control method. Analytics software providers offer analytics as the primary method. Some companies rely on pre-approvals and audits. However, none of these methods is sufficient alone.
Contracts and rules form the basis for cost control. Without them the controls do not work. Issues found, for example in claim validation, cannot be properly acted on, as there is no clear contractual basis for claim rejection or legal actions.
Claim validation alone is not enough as it is impossible to build bulletproof validation. It is always possible to add a fictional service activity or a part or to generate a non-existing claim, where every field meets the rules as such.
Analytics can help to indicate potential issues, but it doesn’t provide transaction level resolution. There needs to be a resolution for each claim – approved, rejected, approved with modifications.
Consequently, a combination of methods is needed. We call this combination the warranty control framework. Having a proper warranty control framework not only improves fraud detection and prevention, but also provides, transparency and cost control over warranty management as a whole. The framework is composed of the following four key elements:
- Contracts define the interaction between the warranty provider and the customer, the service agent and other parties in the warranty chain and the rights and duties of each of them. Contractual rules are the basis for customer claim entitlement and service claim validation.
- Transaction controls include processes and methods for handling customer warranty claims and warranty service claims, from the customer issue through servicing and financial settlement of the claim. The main focus is on customer warranty entitlement, service agent claim validation, and material returns control for defective items.
- Analytics assist in identifying claims with high likelihood of being unjustified, whether through fraud, abuse, or sloppy processes, and to identify service agents and/or customers with high propensity to commit fraud.
- Service network management involves setting up a network structure that minimizes fraud opportunities, incentivizes service agents to right behavior, and provides predefined and clear procedures on actions taken if fraud is detected.
Figure: Warranty Control Framework
How to set up a warranty control framework?
With one global manufacturer, we went through the following cycle:
- First we built regional (and later global) rules and policies. These rules for example defined acceptable service agent claims. Compensation for warranty service claims was also standardized. This allowed different service agents and countries to be compared with each other. Differences in claim volumes and cost levels allowed areas requiring further investigation to be identified.
- We continued with rule-based validation, simple analytics and service agent auditing based on the analytics findings.
- Over time we created more sophisticated analytics and found more dubious cases.
- Continuous improvement was needed to ensure operational effectiveness and keeping up with the new methods invented by the service network.
Having the key elements in place does not automatically guarantee that fraud is under control.
As discussed in our previous blog, on the high level, all processes and policies – validation rules, claim process, statistical data analysis, customer entitlement – may be in a good shape, but the devil is in the details. When you dig a bit deeper, you notice big holes here and there.
In addition to having all required processes, tools and policies in place, competent personnel, for example warranty administrators and auditors are also needed. We have over and over again seen situations, where some persons see no issues in claims that are obviously fraudulent for others.
Book: Warranty Fraud Management
Our book introduces dozens of contractual rules, and around 100 entitlement checks, validation rules, and analytics methods that are recommended to detect and avoid different types of warranty fraud.
Blog: The Dark Side of Warranty Management: https://www.linkedin.com/pulse/dark-side-warranty-management-matti-kurvinen
Author : Ilkka Töyrylä