Decisioning engine

I. Introduction to decisioning engines

A decisioning engine is a software tool used by credit providers to assess an applicant’s creditworthiness, measure them against their product line, and ultimately determine which (if any) products they qualify for.

Modern creditors of all types (including FIs, card program managers, lenders, and other financing providers) rely on decisioning engines to deliver accurate predictions of borrower behavior, mitigating the risk of lending by restricting access from any applicants who are likely to default on their debt. Using fair and impartial underwriting and decisioning tools also helps credit providers avoid discrimination in compliance with the Equal Credit Opportunity Act (ECOA).

This article will explore how decisioning engines work, what benefits they offer to credit providers, and what challenges those providers may face in implementing them.

II. How Decisioning Engines Work

Decisioning sits within the larger process of loan origination. Coming after applications, data gathering, and underwriting, decisioning marks the final step of approval or rejecting an application, or providing a more nuanced answer that pairs a potential borrower with a range of acceptable credit products or financing terms.

How do decisioning engines operate?

{emphasize}

Each decisioning engine has its own methods for assessing underwriting data and delivering approvals. Most follow a similar general process:

1. Apply knockout rules. Before running any complex algorithms or pulling data from third-parties, decision engines can rule out some applicants based solely on the information they provided. If a borrower is underage, living outside the creditor’s operational area, or lacking income necessary for a specific product, the engine can immediately reject their application, in some referring them to a partner who is better suited to finance them.
2. Enrich applications with integrated data. Decision engines supplement the information on the application with integrated data from third parties, both for verifying the applicant’s identity and investigating their creditworthiness.
3. Calculate risk scores through predictive models. Given the applicant’s income and credit history, a predictive model can assess their likelihood of repaying the debt and the risk of delinquency or default.
4. Perform an affordability analysis. Before rendering a final decision, the engine might also compare the applicant’s likely behavior against the creditor’s current portfolio performance,  risk appetite, and available capital.
5. If necessary, conduct a manual review. While decision engines can be fully automated, some credit providers prefer to manually review applications as well, most commonly in industries with a low volume of high value accounts. 

{emphasize}

Decisioning for B2B credit products

Decisioning for business-to-business credit products poses its own challenges. For one, their significant size makes B2B credit accounts a lucrative target for bad actors. And since most anti-fraud tools focus on consumer identity theft, B2B credit providers are at greater risk of fraud with fewer means to fight it.

Typical consumer identity theft sees criminals borrowing on someone else’s name and credit, but when targeting B2B providers, bad actors are more likely to attempt synthetic identity fraud by fabricating a business profile to apply for credit. Alternatively, they may pose as recently dissolved businesses, knowing that most databases won’t have the most up-to-date information. Whether they’re shells for fraud, money laundering, or other illegal activity, these businesses can look legitimate to most underwriting and decisioning tools.

To mitigate these risks, B2B credit providers can enlist screening tools that specialize in business applicants, from the basics (like its age, locations, and EIN) to more detailed information (like its status and standing with state and federal agencies, or any liens on the company’s assets). These specialized underwriting and decisioning engines give providers an in-depth view of business identity risk, preventing potential exposure with shell corporations and non-registered entities.

III. Benefits and challenges of implementing decisioning engines

Most modern credit providers use decisioning engines or other algorithmic tools rather than judgement calls and manual calculations. Compared to manual underwriting, any decisioning engine brings major benefits of streamlining, efficiency, and compliance. 

But creditors must also weigh the benefits of individual decisioning tools against each other, comparing their strengths and weaknesses to find which is the best suited for their portfolio, clientele, and budget.

{emphasize}

Different decisioning engines often seek to distinguish themselves from competitors by excelling in particular areas:

• Data sources. A wider range of data sources can lead to more informed and accurate decisions.
• Configurability. Different credit providers target different segments of the market, and may need more options for accepting or rejecting applicants than a single credit score. Some engines allow more complex or nuanced approval settings.
• Compliance. Built-in tools for federal and state-level regulations ease the creditor’s own compliance burden.
• Platform integration. Decisioning doesn’t exist in a vacuum; engines that directly connect to a front-end application or automatically send approved accounts to an LMS help credit providers manage the entire lifecycle of the account.

{emphasize}

Challenges in implementing decision engines

While decision engines offer many advantages over manual decision making, implementing those solutions brings its own set of challenges.

When a decision engine is disconnected from applications, underwriting, or servicing, decisioning quickly becomes a disjointed, manual process. Similarly, predictive models may grow outdated and require regular updates. 

Most credit providers aiming to modernize their operation will address these issues by choosing a robust decisioning engine that can easily integrate into the rest of their tech stack, allowing them not only to streamline their decisioning, but leverage better data visibility to continually improve their processes.

Enhancing customer engagement through real-time decisioning

Advances in decisioning engines have fully transformed the customer experience and expectation in several credit industries, where rapid approvals are no longer a competitive edge, but instead a must-have for any credit provider.

Retail and point-of-sale credit providers, for example, rely on real-time decisioning to approve borrowers as they make purchases. Healthcare providers have also partnered with credit providers to provide fast, convenient decisioning that expands consumers payment options when pursuing treatment. In both cases, rapid decisioning has become an industry mainstay that significantly improves borrowers’ satisfaction.

IV. Integrating decisioning into your credit operations

Selecting and implementing a decisioning engine may pose challenges, but the benefits to a lending and credit operation are clear: efficiency, compliance, and customer satisfaction.

If your operation is looking to improve decisioning, reach out to us. LoanPro’s origination suite includes tools for automated underwriting and decisioning, whether through our native toolkit or through integrating your own preferred partners. We’d love to discuss your process and see if LoanPro is the right fit.