Case Study

Service: Artificial Intelligence

Industry: Pharma

Department: Global Business Services

Size: 15.000+ employees in over 40 countries

Revenue: 5+ bn CHF (2020)

Credit collection management – predict when your customers are going to pay

Questions about this topic? Ask our Specialist!

Chris Griessmann
Business Intelligence

Chris Griessmann, Business Intelligence and Analysis of Business Processes Contact

Challenge

A large multinational biotech company was looking to optimise its cash collection process. With hundreds of customers ranging from Fortune 500 companies to SMEs, this company wanted a more efficient way to identify high-risk invoices requiring special attention.

An essential step in the invoice-to-cash process is the collection of accounts receivable (A/R). Effective management of A/R is a vital issue for any business as it directly affects the cash flow and the ability to mitigate the risk of aged or bad debts. The ability to predict whether an invoice will be paid on time or late is valuable in a plethora of industries and supports decision-making processes in most financial workflows. Sereviso has prototyped a custom AI solution to support collectors in predicting the probability of an invoice being paid on time or late, as well as the amount of delay.

Solution

The team at Sereviso identified an opportunity to use artificial intelligence to improve the client’s cash collection process. The first step was to single out the relevant data points needed to predict the anticipated payment delay of each invoice. We recognised that invoice-level accounting data and the historical payment behaviour of each customer were vital for generating accurate predictions. We also identified additional uncaptured relevant data points that could increase our prediction accuracy and recommended drawing up methods to collect this data for possible use in the future. 

Moreover, we created a proof of concept to validate our hypothesis and gather client feedback. After receiving approval from the client, we proceeded with developing our invoice payment prediction solution. We leveraged advanced analytics and machine learning to create a predictive algorithm using available client data. Our model took account of dozens of variables such as net due date, risk rating, and credit limit.

We also migrated on-premise client data to Azure Cloud for more robust, flexible, and performant data pipeline management. Additionally, we built cloud analytics services to facilitate the development of the invoice payment prediction solution.

The final outcome of the invoice payment predictions was displayed in dashboards within the client’s broader accounts receivable reporting solution. The solution targeted our client’s business operations as well as their overarching finance department.

Results

Following implementation, the A/R solution provided a series of wins for the client:

  • A 94% accuracy rate in invoice payment predictions
  • A reduction in days sales outstanding
  • Lower risk of bad debt write-off
  • Improved prediction of expected cash flow
  • Lower cost of capital

The solution’s accurate classification of high-risk invoices enabled the credit collection team to better identify and prioritise customers that required preventative attention. Our solution’s dashboards provided key information and insights on each customer. Newer team members were enabled to become more easily acquainted with each customer’s specific needs and situation.

This led to increased efficiency and better performance of the team. It also improved the overall financial health of the client’s business. Sereviso’s team continues to work with the client to further enhance the available data and identify opportunities for optimisation throughout the entire order-to-cash process.

 

 

Interested in more?

If you are interested in the technical part behind it, you can read about it here.


More Case Studies