Please note: The algorithm descriptions in English have been automatically translated. Errors may have been introduced in this process. For the original descriptions, go to the Dutch version of the Algorithm Register.
Behavioural groups
- Publication category
- Impactful algorithms
- Impact assessment
- DPIA, IAMA
- Status
- In use
General information
Theme
Begin date
Contact information
Link to publication website
Responsible use
Goal and impact
The algorithm helps DUO to improve debt recovery. It categorises debtors into behavioural groups based on their payment behaviour and payment status. The algorithm’s main aim is to approach debtors who are in arrears in an appropriate manner. In this way, DUO hopes that these debtors will get in touch with DUO. DUO can then make arrangements with the debtors regarding what they are and are not able to pay, and ensure that no further repayment problems arise thereafter. Once DUO has made contact with the debtors, the group to which they have been assigned has no bearing on how DUO treats them. DUO also uses the groups to decide which debts cannot (temporarily) be recovered. Finally, DUO uses the algorithm to gain an overall picture of the payment behaviour of all debtors. This information helps DUO to make strategic decisions regarding the total amount of all debts. These are therefore not decisions relating to individual debtors.
Considerations
The algorithm helps with contacting debtors. It enables DUO to collect debts more effectively and helps debtors to prevent or reduce payment difficulties. The algorithm also helps DUO to gain an overview of the total amount of all debts and to exclude from recovery those debts that are (temporarily) uncollectible. The actions DUO carries out using the algorithm have little impact on debtors. DUO uses the algorithm only for debtors with payment arrears. To create the behavioural groups, the algorithm considers only information relating to the debtor’s payment behaviour and payment status. DUO does not use any other personal data for this purpose.
Human intervention
Debtors are automatically categorised into behaviour groups. The algorithm does not make any decisions regarding debtors, nor does it make any recommendations. A member of staff is always involved in all actions taken by DUO.
In the follow-up messages from DUO, the debtor is asked to contact a DUO member of staff. As soon as DUO has made contact with the debtor (for example, because they have responded to a message), DUO proceeds with the standard collection procedure. The next steps are independent of the behaviour group to which the debtor has been assigned. The DUO member of staff and the debtor work together to find a solution, such as a payment plan.
Risk management
The data used by the algorithm is sourced from DUO’s own systems. This includes information on the amount of the payment arrears and the rest of the financial situation held by DUO. DUO already has this information and does not request any additional information.
DUO continuously updates the classification of debtors into behaviour groups on a daily basis. This provides an up-to-date picture of all debtors. The classification is not shared with others and is not used for any other purposes.
The algorithm does not make any automated decisions that affect the debtor’s debts. A DUO member of staff is always involved in decisions concerning the debtor.
Legal basis
Data processing for behavioural groups takes place as part of the person-centred debt recovery process. This process is based on specific legislation, such as the Student Finance Act 2000 and the General Administrative Law Act. Debt recovery is a statutory and social responsibility of DUO.
The Student Finance Act 2000 sets out, amongst other things, how former students are required to repay their loans. Articles 6.6 and 6.9 of the Act contain information on the start of the repayment period and the monthly instalments. Chapter 8 sets out important information regarding debt recovery.
Links to legal bases
- Student Finance Act 2000, Article 6.6: https://wetten.overheid.nl/BWBR0011453/2025-01-01/?g=2025-01-01&z=2025-01-01#Hoofdstuk6_Paragraaf6.1_Artikel6.6
- Student Finance Act 2000, Article 6.9: https://wetten.overheid.nl/BWBR0011453/2025-01-01/?g=2025-01-01&z=2025-01-01#Hoofdstuk6_Paragraaf6.1_Artikel6.9
- General Administrative Law Act: https://wetten.overheid.nl/BWBR0005537/2020-04-01/
Elaboration on impact assessments
Impact assessments have been carried out for this algorithm, in which privacy rights and human rights have been weighed against its positive and negative effects. The conclusion drawn from these assessments is that the use of the algorithm is justified.
Impact assessment
- Data Protection Impact Assessment (DPIA)
- Human Rights and Algorithms Impact Assessment (IAMA)
Operations
Data
The algorithm uses the following data relating to the debtor:
- The payment arrears
- Date of the last payment received
- Information on the payment history and the repayment phase the debtor is currently in
Technical design
The algorithm is a rule-based algorithm that groups behaviours based on the specified data. A rule-based algorithm means that it operates according to fixed rules that have been defined in advance. The algorithm follows these rules step by step and does not learn independently.
Similar algorithm descriptions
- The algorithm automatically checks whether borrowers repaying their loan for integration purposes are eligible for repayment based on their ability to pay. ‘Means-tested’ means that borrowers repay a monthly amount that they can afford to set aside based on their income (or that of their partner). This is calculated on the basis of their family circumstances and their (and their partner’s) income.Last change on 29th of June 2026, at 11:49 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- IAMA, DPIA
- Status
- In use
- DUO makes every effort to prevent payment arrears from becoming time-barred. For debtors whose arrears are about to become time-barred and for whom no residential address is known, DUO may issue a public notice. Public notification means that DUO publishes the debtor’s name and outstanding debt in the Government Gazette. An algorithm is used to determine who is eligible for this.Last change on 23rd of June 2026, at 13:10 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- DPIA, IAMA
- Status
- In use
- The RIS matching algorithm detects payment arrears among residents and matches them with data to identify at-risk groups for problematic debts. It supports early signalling and debt relief by revealing debt risks in a timely manner so that the municipality can provide preventive help in time. It identifies payment arrears at utility companiesLast change on 27th of January 2026, at 11:36 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- DPIA
- Status
- In use
- The 'OB Balance Payables' algorithm helps the Tax and Customs Administration employee detect remittance differences. A remittance difference occurs when it appears that, based on the most recent income tax or corporate tax return, more sales tax was due than was declared for that year.Last change on 12th of May 2026, at 14:50 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use
- This algorithm helps Customs and its enforcement partners select goods for control on a risk-based basis. Among other things, it uses declaration data from companies and assesses whether or not there are risks of non-compliant non-veterinary feed, feed materials and feed additives.Last change on 2nd of April 2025, at 12:48 (CET) | Publication Standard 1.0
- Publication category
- Impactful algorithms
- Impact assessment
- Field not filled in.
- Status
- In use