IOTA Case Study Workshop on “Automated Segmentation of Debtors for Boosting Debt Recovery Success”

The main objective of the Workshop was to provide IOTA member tax administrations with an opportunity to expose their own cases and share their best practices of tax debt/debtor segmentation with a particular emphasis on how to differentiate debtors better to improve tax debt collection and prevent overdue indebtedness. During the Workshop, a couple of selected case studies from IOTA member tax administrations such as Belgium, Estonia, Hungary, Ireland, Lithuania, Norway, Spain, and Sweden, were presented. These case studies were designed to showcase various debt segmentation aspects and models, including the strategic, organisational, and technical support used to support the segmentation process, as well as the methods for monitoring and measuring the effects of the process.  

Before exploring the technical details of the first case studies on the opening day, workshop attendees were all able to exploit their creativity through an interactive activity session where they briefly introduced themselves to one another, and shared their needs and expectations from this Workshop. This activity was followed by plenary sessions and participants examined case studies from Norway and Spain. These cases were discussed in smaller breakout sessions and offered the participants an opportunity to exchange their own views, ideas, and experiences regarding the various segmentation models, organisational structure supporting those models,  and risk analysis models that need to be developed for a successful segmentation strategy. On the first day of the Workshop, the following countries brought their insightful case studies to the plenary:

• Norway: “Transitioning from Claim-Oriented Collection to Behavioural Segmentation - Key steps and organisational challenges” presented by Joakim Døving Dalen;
• Spain: “Segmentation in Preventive Recovery” presented by Fernando Alonso Monferrer.

The second day of the Workshop was very intense and explored another group of practical presentations. Workshop participants become acquainted with cases from Ireland, Lithuania, and Belgium. The studies particularly focused on the behavioral aspects of segmentation, the automated segmentation of debtors, the modern technologies supporting this process, KPIs, and evaluation methods used for monitoring the effects of segmentation. Each case was discussed in the groups where participants worked together on highlighted issues on the cases and discussed approaches, benefits, opportunities, and possible difficulties to each case. On this day, the following subjects were covered:

• Lithuania: “Decoding Debt Risk: Key Insights and Solutions” presented by  Vitalija Burdeinaja;
• Ireland: “Segmentation, Categorisation & Automated Actions in Debt Management” presented by Bernice Ward; 
• Belgium: “Face the music: we have a different tune for each segment of debtors” presented by Arnaud Collard.

The closing day of the Workshop featured case studies presented by the IOTA member tax administration of Estonia and Hungary. As was in the previous days, these case studies have also been discussed by the workshop participants in smaller breakout sessions to exchange views, ideas, and experiences regarding the topic. In the final plenary session of the day, delegates explored the forward look into further developments in automated segmentation of debtors, based on the Key Presentation delivered by the Swedish presenter who joined the Panel Debate together with speakers from Belgium, Estonia, and Hungary. On the final day, the following speakers and panellists shared their practical examples and viewpoints with the plenary:

• Estonia: “Segmenting on the basis of behaviour information” presented by Liina Jõõts;
• Hungary: “How to increase the segment of „low risk” taxpayers? Behavioural insights in tax collection” presented by Katalin Kanizsai LL.M;
• Sweden: “Less Complexity, More Clarity: The Power of Small, Explainable AI Model” presented by Marcus Bräutigam;
• Belgium: Panel debate on “How to provide clean and reliable data for segmentation and the possible use of GDPR proof data”, Michael Roekarts.

Besides the speakers, IOTA is grateful to all workshop chairpersons, note-takers, and attendees who actively participated in the group talks and discussions of this 3-day technical event. All the materials and presentations are now available online on the event’s webpage. This workshop was led by the IOTA International Taxation Expert Ivana Donchevska with support from the IOTA Work Programme Manager Eugenijus Soldatkovas and was held in Budapest, Hungary.