Phone Data Testing - Traps and Pitfalls to Avoid

Jason Horsley

To get more right party contacts (RPCs), collection agencies need to have the best data. The promise of “having the best data” is easy to say, but hard for most vendors to deliver. So how do you test your vendors to determine who really does have the best data? After reading this article you’ll know the best practices that you can implement to test the phones in the collections space.

Let’s assume you’ve already gone through a pre-qualifying process to select a few Data Providers to test. You evaluated their financial stability, regulatory compliance and critical business practices and have decided that if their products test out, they would be reliable and trustworthy choices. Now it’s time to evaluate each provider’s phone product to make a determination as to whether to engage with this provider and/or where this provider might fit into your overall strategy.

Here are three traps to avoid when setting up the vendor test:

1. Avoid the temptation to de-dup as you have in the past. Your goal is to determine the right-party-contact number for the subject, not to take deep dives into your provider’s databases. Traditional de-duping only serves to return you more bad numbers. Change your practice to flagging and keeping good numbers that happen to be a duplicate.

2. Avoid inconsistencies in how you hold each data provider accountable. For example, Provider A is not allowed to return strong lead phones, but Provider B can because it does not have the technology to distinguish between a subject and a strong lead. This discrepancy penalizes the “smarter” provider A and leaves you with a false comparison. Be sure to level set the testing parameters.

3. Avoid misinterpreting how your collection system is populating the returned data. This situation is particularly concerning when you have the de-dup function on, as many systems populate, based on order-of-receipt. A risk is that second and third providers could have good numbers blocked from being loaded to your system. Additionally, some systems can’t support phone meta data, like a phone classification for mom or roommate, which may be important to you. At a minimum, understand how your system ingests this data, so that you can account for it in your final evaluation.

When you’ve set up the test with integrity, you are establishing a mode of doing business that enables you to receive the best phone numbers, which impact the productivity of your employees and the results you produce for your clients.

Equally important to pre-test set-up are the steps that you attend to during the test.  Here are suggestions that you can implement during the testing period to ensure that that the test is pure.

1. The best dialing approach is to use a third-party dialing vendor to call ALL the numbers. Often agencies attempt to do this dialing on the collection floor. While this approach may be less expensive, it is not an accurate assessment. To comply with the FDCPA, agencies cease dialing subsequent numbers after initial contact is made. Very often another number for the same debtor is also a right party contact, but because that number is never called, it is never counted or credited, creating a missed opportunity situation for you to collect that debt.

2. The day of the week and the time of the call matter when conducting a comprehensive vendor phone test. For example, if Provider A phone calls are made in the afternoon and Provider B phone calls are made in the evening, then the advantage goes to Provider B every time, as it is a well-known fact that significantly better contact can be made in the evenings. When this truth is overlooked, the performance results lie to you and you’ll be at risk of making a potentially bad business decision due to faulty test design.

After the calling is complete, it’s time to dig into the results spreadsheet to conduct your analysis. There are four primary  pitfalls to watch out for during the analysis phase.

1. Do a complete evaluation even if one provider is offering a cheaper price. Buying on price with the attitude that the product is “good enough” can be a costly short cut. While the base cost of the data may be cheaper, that doesn’t mean it is a good value. The expense of time and resources to work inferior data is not worth the small data acquisition savings. Lead your organization to do the needed work to evaluate performance and price to make a value-based purchase.

2. Don’t choose quantity-of-phones over quality-of-phones. In most cases, more is not better, it’s actually worse. More usually means more wrong party contacts, something most of us are trying to avoid for compliance and efficiency reasons. Lead your organization to choose right-party-contact rates over just hit rates.

3. Don’t evaluate based on dollars collected instead of right-party-contact rate. Evaluating on dollars collected may seem like a good idea at first, but this approach makes the collector’s ability a variable in the equation. Having such an uncontrollable variable in the process, especially one that is difficult to account for, diminishes the value of the final analysis. Lead your organization to choose right-party-contact rates over dollars collected.

4. Don’t evaluate based on unique hits instead of unique right-party contacts. Big difference. Counts of unique phone numbers is a critical metric, but you have to make sure you are counting the right thing. Lead your organization by counting and comparing only unique phone numbers that result in a right party contact.

When these suggestions are not heeded, your analysis may mislead you and, in turn, steer you to make a wrong decision for your organization. You may have to take a stand with your organization to change its thinking, but with aligned goals, your organization should reap the ultimate benefit of becoming a better business by being able to reach more right party contacts than you were before adopting these best practices.

For more information and more best practices on phone number testing, download the free whitepaper “Strategies for Effective Phone Data Testing” here.