Catalog record · Customer Experience
Stage as of Sep 9, 2026 · historical report, not live reconfirmation.
Business debt-collection journey segmentation
Primary workflow source. A named organization's workflow supported by a reviewed primary source. Stage is reported separately; this label does not establish measured outcomes or independent validation.
Last researched Oct 29, 2026. Source access and reported stage are recorded below; deployment evidence does not establish ROI.
What the research supports
HMRC's Business Propensity to Pay links business tax debts and uses calibrated gradient-boosted trees in SAS Viya to inform communication timing, frequency and content. It scores approximately four million customers monthly. The record describes temporal holdouts, cross-validation, monthly output/input-drift monitoring and annual model review; the score is only one input into the collection journey.
- Reported stage
- Production
- Technology
- Machine Learning
- Business function
- Not assessed
- Impact (legacy label)
- Not labeled
- Catalog date · unconfirmed
- Not reported
- Imported
- Oct 29, 2026
- Last researched
- Oct 29, 2026
Measures & evidence
No quantified measure recorded. This is a gap in the evidence, not a finding of no value.
Documented vendor relationships
No vendor relationship has been linked to this specific workflow in the curated overlay. This does not mean it was built without vendors.
See the company’s documented relationshipsDeployment evidence & limits
Production; ATRS published 2026-09-09 Status as of 2026-09-09.
Stage is as reported in the cited evidence. Current operating status has not been independently reconfirmed; the research date is an editorial review date.
The record gives evaluation methods but no accuracy, collection uplift or causal harm metrics. Human review is monthly aggregate review, not mandatory review of every model-driven communication. Source describes no direct appeal against the score. Tax enforcement powers and repayment incentives differ from lender collections. Engagement is not willingness or ability to pay, loss reduction or customer welfare. Do not reuse tax-trained scores or equate exclusion of protected attributes with fairness.
Potential FS application · inferred
Inferred: SME arrears outreach prioritisation and channel/timing selection, with calibrated engagement likelihood and human restructuring discussions.
Inferred: SME arrears outreach prioritisation and channel/timing selection, with calibrated engagement likelihood and human restructuring discussions.
Original industry: Tax administration. Transfer limits: Tax enforcement powers and repayment incentives differ from lender collections. Engagement is not willingness or ability to pay, loss reduction or customer welfare. Do not reuse tax-trained scores or equate exclusion of protected attributes with fairness.
Source register
- HMRC: Business Propensity to Paygov.uk · opens in a new tab
HMRC / GOV.UK · issuer · full-text · Published 2026-09-09
Research lanes: public_science_transfer · Batch global-discovery-2026-10-29. Coverage and remaining gaps