Bad Debt Reduction

Bad Debt reduction : improving your recovery performance

Cash collection is one of the most critical processes of a company. Not only does it drive the need for treasury, but it is also a critical moment for customer satisfaction - or dissatisfaction.

Whether supported by automation software or outsourced to AR companies,  recovery actions are always about waiting, monitoring and reacting to delays. The unfortunate part is that we only see the delays when they occur.

At that moment, hard choices must be made between protecting the customer relationship and securing the cash recovery.

Profiling and predictive analytics can help us to anticipate those delays and act accordingly. Taking recovery action in due time and thereby avoiding unnecessary recovery costs can drive significant financial contribution.

Bad debt wordcloud.png

Optimise and personalise your recovery actions according to the customer’s profile


Financial audit requirements stipulate very clearly documented dunning processes and rules. The rules dictate what the recovery strategy should be, broadly based on invoice amount and payment terms.

Tuning the rules is a real challenge for risk management systems and can have significant financial impact. Once fixed in the system, they are not necessarily capable of capturing customer specificities; nor are they able to  leverage any additional information you might have gathered about them. It remains impossible to answer which customers will exhibit a fraudster profile or which customers can be treated more gently.

These questions become even more complex to tune when the recovery is outsourced to specialised agencies that will charge according to the effort it takes to apply the rules.

This is where can play a key role.

An analysis of your past invoices can highlight the different recovery profiles and can propose for each invoice its probability of recovery and at which step in the recovery process.


PredicSis AI for cash recovery improvement: get individual predictions of likelihood to pay as early as invoice issuance

Payment behaviours follow deeply complex patterns. With’s ability to compute automatic profiles based on past observation, you can use the full range of your customer data to create risk scores and probabilities-to-pay at each step of your recovery process and as early as invoice issuance.

Through simply loading your historical invoice or payment defects data, will give you cause analysis of bad payments and will profile all your new invoices. This valuable information will help you to decide the best recovery actions to take in the quickest possible timeframe.