Datamining lists double the yield of URSSAF audits of small firms

The Cour des comptes found that URSSAF audits of very small firms picked by data mining reassessed 12.6% of the contributions audited in 2018. The rate across all such audits was 6.1%.

  • 2 an organization

    Audits of very small firms picked by data mining reassessed 12.6% of the contributions audited in 2018, against 6.1% across all such audits.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported benefit.
Evidence

foundstudiedreportedclaimed

An independent audit report supports the benefit.

Compared with

All URSSAF accounting audits of very small firms in 2018, including those picked by data mining. Measured.

Without the system
6,1 %
With it
12,6 %
The system’s part
Not isolated. The Cour compares audits drawn from the data-mining lists with all audits of the same kind, which include them. It writes that the Acoss sends lists of firms with high risk scores. It does not describe how the scores are built or say whether they learn from past audits.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
Acoss data-mining risk scores (outils de datamining) that list small and medium firms (since 2015) and very small firms (since 2016) for URSSAF accounting audits
Made by
Agence centrale des organismes de sécurité sociale (Acoss)
Used by
URSSAF network
Where
France (national)
Stage
In routine use
Date
Kind
Aggregate result
Broad benefit
Analyzing Data or Information, Organizing, Planning, and Prioritizing Work
Collections
Scrutiny and findings
Last revised
31 October 2026

Sources

  1. Audit report, by authors with no tie to the system.

    Cour des comptes. La fraude aux prélèvements obligatoires: évaluer, prévenir, réprimer. Communication au Premier ministre, November 2019, published 2 December 2019, pp. 128-129, Tableau n° 14.

    Ainsi, le taux de redressement des cotisations par ciblage grâce au datamining, mesuré sur le champ des TPE, ressort à 12,6 % en 2018, soit deux fois plus que le taux constaté sur l'ensemble des CCA.

Details

AI system

Workflow

The Acoss scored firms by risk and sent the URSSAF offices yearly lists of firms with high scores. URSSAF inspectors audited firms from these lists alongside audits planned by other means.

Follow-up

Conditions for replication

URSSAF offices must take up the lists when they plan audits, as they did for 10% to 20% of accounting audits in 2018.

Entry details

Sector

Social contributions enforcement (employer audits).

Purpose

The Acoss built the data-mining tools to score firms by risk and list those that URSSAF offices should audit.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Datamining lists double the yield of URSSAF audits of small firms (AIBD-0920). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0920.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions