Credit Decision Engine Software - A complete implementation guide

This article will help you to answer some of the fundamental questions about credit decision engine software and its implementation tasks before sinking your teeth into the most critical part of your bank's new digital transformation.

• What is credit decision software?

• How does credit decision engines work?

• The benefits of using credit decision software.

• Where can you buy one or rent one? 

• Whom do you need to hire to run a credit decision software?

• “How long does it cost and take for a full integration?

What is credit decision software?

Credit decision software is a computer program that can be a standalone system that runs on servers and personal computers. It can also be a web based application that can be used with other cloud based enterprise systems connected through APIs or Application Programming Interface.

Credit decision software comes in two varieties. One of which is a command line interface and requires heavy computer programming knowledge and computer programming skills. This types of programs are often designed or even made for computer scientists or software engineers. They can provide a quick implementation but the content that goes into the decision engine often opaque to other people in the organization. Which makes it hard to detect issues or debug.

The other type has an extensive user interface that can help not only the programmer, but a vast array of risk management professionals and product management people to easily understand what is programmed into the decision engine and how to make changes. 

Organizations today often require multiple parties to be involved in operating a credit decision engine software, so we believe that having an user interface can go a long way when it comes to spreading the understanding of what goes into a decision to the rest of the organization. 

How does credit decision software work?

Credit decision engine’s main job is to take in data that’s provided to the software from a variety of different sources and run pre-programmed business rules to determine a logical or mathematical result. 

The decision engine must have a set of logic machines that can run simple if-then-else statements. For example “if the customer’s FICO score is greater than 600 and the customer’s gross income is greater than $3,500 per month, then present product off B.”

These statements above automatically touched a few data points which the decision engine must take in order to run through that statement. First, it must have the understanding of a FICO score. 

A FICO score is a credit risk rating credit score invented by Fair, Isaac and Company to measure the person’s credit worthiness. The credit decision software must be able to find and ingest this piece of information. 

Most of the major credit bureaus in the US and around the developed world have contracts with Fair, Isaac and Company to calculate each individual they are monitoring in their system using Fair, Isaac and Company’s statistical algorithm.

So how does the credit risk software get this information from these credit bureaus that may have this score that it needs to run credit risk rules? Sometimes, it is fed into the credit decision manually by a risk management staff. These risk management staff might take the person’s information, enter it int an interface provided by the credit bureau.

The credit bureau then returns a credit report with this score at the top of the credit report. The risk management staff then put that score into the credit risk decision engine. You can imagine that this task might take hours or days and make it a very bad customer experience.

Today’s credit decision engine software has APIs or a handshake over the internet using a universal language to connect to other third party enterprises. The credit score engine or software should connect to credit scoring bureaus and retrieve this score and other attributes automatically from the credit bureaus within a fraction of a second.

The benefit of using credit decision engine software

One of the benefits of using credit decision software is automating all of the manual decisioning processes a bank and credit union encounters daily.

This technology also lessens the chance of error, reduces the amount over time and increase’s customer satisfaction. Banks and credit unions can focus on branding and marketing to push more applicants through a digital onboarding experience powered by a decision engine software.

A decision engine won’t replace a fraud analyst or a risk analyst completely. Sometimes, decision engine software can’t make conclusive decisions because it doesn't have complete information or the information that’s supplied by the applicant is erroneous. 

One of the other benefits of a decision engine is that it has to have a built in decision engine workflow to triage these out of band situations. The decision engine software should have a queue which the fraud or risk analyst can further investigate these situations by asking for additional documentation or decline the application altogether or have the applicant reapply.

While the application is under consideration, the decision engine software should put this application in suspension mode and clearly communicate to the end customer that their application is under review and set the appropriate expectations on how long it will take to adjudicate their application in its current status.

There are many more benefits of using credit decision software and compliance is one of them. Credit decision software must keep track of all the decisions, consents, and promissory notes signed by the applicant.

It should also save all of the decision trees, scores, attributes and decision review paths as part of the application. If there’s ever any issues with this application, the decision engine software should provide all of these datasets or information used to produce the approval or denial results to the bank’s compliance department and their regulators.

Where can I buy credit decision software?

At this point, you might be thinking to yourself that a decision engine forward might cost an arm and a leg. It certainly was back in the days when computing cost was high and infrastructure was not available to support the amount of automation demanded by today’s consumers.

There are many cloud based solutions with intuitive interfaces and third party integrations as well as workflow management all in one and sometimes free of charge as a larger enterprise package.

When you decide to buy a decision engine software, we courage to ask the following questions:

• How often do new features and functionality are added to the decision engine?

• How long is the product road map and deployment schedule?

• How many third party data services are already integrated with the decision engine?

• Does it have a workflow orchestration system that my staff can use to review applications when they are in a review status

• Does the decision engine software track all of the decision points, third party data used to make these decisions and the version of the decision path used in each application?

• Does the decision engine have an user interface that my risk and product department can use to edit rules and build new rules?

• Does the decision engine software have a robust set of APIs for it to communicate with the rest of my enterprise system?

• And does the decision engine run on a subscription cost or a one time cost?

• Finally, does the decision engine software come with any maintenance and support? 

The cost of these credit decision engine software itself should be minimum. Sometimes the decision engine software is free to use as part of another enterprise system purchase. But a stand alone or a cloud based decision engine software should cost no more than $1,000 a month minimum subscription fee plus a minuscule transaction fee, say 10 to 15 cents per transaction.

Of course every credit decision engine or software comes with a variety of features and functionalities. These costs and contracts also depending on the length of the relationship and maturity of the projects as well. 

Who do I need to hire to run credit decision software?

So, now that you have a fancy decision engine software purchased and ready to go, what to do next? If you are a bank or a credit union, you may have a project team working on a larger project and folks within that team might be the point of contact to work with the decision engine software companies engineers.

Here is a list of critical personnels needed to ensure a successful decision engine software implementation or integration project.

• Project Manager: This person is responsible for overseeing all of the requirements needed for a successful integration. From defining all of the fields and attributes needed on the online digital in take form, to all third party data bureaus needed to retrieve all of the attributes to make a complete data package to run all of the rules, at least to the best of the team’s knowledge at that time.

• Risk Manager: A risk manager’s responsibility is to produce all of the business rules necessary to satisfy any KYC or Know Your Customer policies. Another role for the risk manager is to build or at least produce all of the rules requirements for credit risk management. We mentioned a rule example earlier in our article. The risk manager may also influence which identity bureau or credit bureau the bank needs to make these critical identity and credit decisions.

• Product Manager: A product manager’s role is becoming more important over the years with the advent of mobile phone and fast internet as well as design cues and techniques influenced by big techs such Apple, Meta etc. One of the main responsibilities of a product manager is customer experience. How many steps is the customer willing to take to complete the application or when should we let the customer know that their application is looking good and on its way to be approved.

• Pricing Analyst: This person’s role is to research all of the regulatory requirements on how fees, interest rates, duration of a credit product or a credit line should be set for the bank’s product and services. For example, the pricing analyst needs to set the minimum and maximum loan amount on a personal installment loan. The fee and interest or APR (Annual Percentage Rate) this personal loan should carry. Lastly, the pricing analyst should also set the duration of the personal loan right into the credit decision engine software. Most of the time, the pricing analyst should work with the risk manager to make sure what the risk management policies’ intent is fulfilled with the product management settings. That is to say, both of these roles should work together on risk-based pricing principles.

• Integration Engineers: Internal engineering resources are needed sometimes if you are integrating the credit decision software into a wider set of enterprise systems. These engineers should have a good understanding of the APIs and Webhooks both internally from various enterprise systems and externally of the APIs provided by the credit decision software company. These two teams must work hand in hand to pass information and decision results back and forth between them to build a customer experience that’s designed by the product manager.

• UI/UX Designers: Sometimes the bank and credit unions might have their own UI/UX (User Interface and User Experience) designers. It is critical to bring these designers into the conversation along with the product manager. These designers must be made aware of the capabilities of the credit decision software and especially the input/intake points the credit decision software is required to make decisions at every step of the way. To avoid the situation where certain fields are missing from the UI/UX design or unnecessary fields collected from the design but can be supplied by the third party, it is a great idea to include all of the designers into the conversation and review the end-user product flow as soon as possible.

If you are a bank or fintech startup, you may have to heavily leverage the founders and the founding team to play all of these rules. You may even have to lean on the people behind the credit decision engine company to help you make some of these critical decisions. Perhaps that’s an item of consideration when you sign up with a credit decision engine software company and see if the company is also willing to help you along the way if they have the business and injury knowledge. 

How long does it take to integrate with a credit decision engine software?

Now that you have the troops assembled and the credit decision engine software picked out. The question is how long will it take to get my first application through? Here are some things to consider that could be a time sink before getting started.

• Identity Bureaus: Banks and credit unions must have an identity bureau contracted if they want to have a complete digital experience built for their clients. These identity verification bureaus are easy to deal with. They don’t require extensive research of your business before getting you a contract to use their identity database. There are so many to choose from. We encourage you to go through our Connect Marketplace and click on Identity Verification Partners. Expect the contract to take a month of two to complete. Most of the time, these identity bureau partners can open up their sandbox APIs and get you started before the contract is signed to hasten the speed of execution. They want to get paid too.

• Credit Bureaus: Credit Bureaus and their onboarding process can be extremely difficult and we encourage you to start that conversation as soon as you get up in the morning. They may require business licenses, lending and banking licenses, on-site visits and vetting your permissible purpose (e.g. why do you need to pull credit reports on consumers?). These contracts can take months and we encourage you to get started right away. These credit bureaus are also shy about opening up their sandbox before signature is acquired on a contract. We have a few of them listed on our Connect Marketplace as well. Once you are on board with the credit bureau, you can install the security certificate as well as your member IDs right inside of the credit decision engine software’s third party integration portal. 

• Bank Transaction Bureaus: Some of the banks and credit unions completely skip the credit bureau check and use banking data or even their own banking data as data points for underwriting. There are a few offerings out there and most of them are all consumer consent based where the consumers have to log into their bank account and transaction history is then downloaded and examined for underwriting. These vendors and partners are easier to deal with and the cost of these banking transaction pulls has fallen to near zero because of new banking regulations. So beware of the cost of these contracts before you sign. These vendors have sandbox sign up right on their website and you can sign up and start programming right away. If your credit decision software already comes with a third party integration module, then, you can just sign up and drop your user name, password and certificates and activate the vendor right inside of your credit decision engine. Keep in mind that the bank transaction vendors have an interface for the consumers to login, if you are designing the digital experience yourself then you need to pass the token to the decision engine and the decision engine software will use the token to retrieve all of the banking transaction data to execute further credit underwriting rules.

• Payroll Data Provider: There is a new breed of data providers specifically focused on payroll data. Similar to the bank transaction data, these payroll data brokers let the users log into their payroll system and retrieve payroll records or split payroll payout and route it to another account number to repay goods, services and loans. These providers are easy to work with and there are a few of them in our Connect Marketplace. You can sign up and get into a contract with them fairly easily without a lot of back and forth. Again, similar to the bank transaction data providers, the bank may have to manage that login interface and have tokens sent to the software decision engine to do the next steps.

• Loan Management Systems and Bank Cores: If you are a bank, the chances are that you already have a bank core. Whether it’s Jack Henry, Mambu, Thought Machine, FIS or Fiserv, you may have already faced difficulties interfacing with your own core. There are companies such as PortX that simiplies that process and can connect a number of digital onboarding systems and credit decision software companies right through their API layers and right into your core. Integration into the bank core systems is not a trivial process but PortX of the world will make that process a little easier. When it comes to Loan Management Systems, there are some tools that offer no API integration which makes loan onboarding difficult and delayed. Then there are a few more modern systems that has a full range of API integrations end points that can interact with digital onboarding systems and create a must more flexible customer onboarding experience 

Implementation pitfalls

All of these third parties' contracts take time to negotiate and sign and it’s completely out of control of the tech and product team. These contract negotiations will almost always impact the timeline of a decision engine software integration project.

These projects can take up to six months to complete even if all the contracts are signed and team members in place, plus all of the identity and credit management rules are all thought through. 

What we have observed is that 20-30% of the homework is done up front and it becomes a discovery process once the team starts to work together. Between chasing contracts, finalizing the rule sets and which product and data set is needed as well as workflow design, the project can run over time and over budget.

However, we encourage you to use this simple guide we put together and start asking simple questions before rushing to sign a contract with a credit decision software company. 

We will update this credit decision software implementation guide frequently to help our community to make more accurate decisions both from a project implementation standpoint and from a credit risk management perspective

Team at LendAPI

About LendAPI:

LendAPI is a full digital onboarding system for banks and lenders alike. LendAPI offers a complete DIY product builder, user interface driven decision engine that can power any application process. Banks and lenders can design their consent, offer agreements, email templates. The entire customer experience is managed through a back office workflow orchestration platform that communicates with brokers, branches, back office personnel and end customers.

LendAPI Workflow Orchestration:

LendAPI Rules and Decision Engine: