Project AURIONAI Use-Cases Research & Innovation
Telecom

Subscriber Personalisation, Next Best Offer and Churn Reduction

Recommends products and offers to each subscriber across apps and channels and targets retention offers at likely churners.

recommendationtabular
0named models
41candidates
1deployments
1in Indonesia

Named Models

The research names no model for this use case.

Where this comes from
Models the research names for this use case. Research, with sources on each model record.

Who Runs It

production, 1
TelkomselTelkomselCountry: IndonesiaISYANA in-house AI and analytics platform (collaborative filtering, matrix factorisation, custom churn models)Vendor: Telkomsel (in-house)productionDate–20 to 30 TB of data a day; 20.75 percent more programme take-up (company claims)Evidence: company statementConfidence: lowSource: info.tmforum.org
Where this comes from
Documented uses by named organisations. The evidence column says who made the claim.

Candidate Models

TabArena: All Tasks, no imputation, snapshot 2026-05-06, 59 entries

A general benchmark for this model class, not a test on this use case. Elo across tabular tasks, anchored RandomForest default = 1000; TabPFN-lab researchers are involved. Higher is better. Elo has no zero, so the axis starts at 1,200. Showing 10 of 13.

AmazonAutoGluon 1.5Amazon maintainer-run1,700Prior LabsTabPFN-2.6Prior Labs maintainer-run1,626Prior LabsRealTabPFN-2.5Prior Labs maintainer-run1,604INRIATabICL v2INRIA maintainer-run1,598INRIARealMLPINRIA maintainer-run1,514Layer 6 AITabDPTLayer 6 AI maintainer-run1,460YandexTabMYandex maintainer-run1,448Microsoft AILightGBMMicrosoft AI maintainer-run1,435YandexCatBoostYandex maintainer-run1,418DMLC communityXGBoostDMLC community maintainer-run1,376
1,2001,5001,800
DMLC communityXGBoostDMLC community, tabularReleasedAug 2026open weightshas scoreslast 18 monthsMicrosoft AILightGBMMicrosoft AI, tabularReleasedJul 2026open weightshas scoreslast 18 monthsPrior LabsTabPFN-2.6Prior Labs, tabularReleasedMar 2026open weightshas scoreslast 18 monthsYandexCatBoostYandex, tabularReleasedFeb 2026open weightshas scoreslast 18 monthsAmazonAutoGluon 1.5Amazon, tabularReleased2026open weightshas scoreslast 18 monthsINRIATabICL v2INRIA, tabularReleased2026open weightshas scoreslast 18 monthsPrior LabsRealTabPFN-2.5Prior Labs, tabularReleasedNov 2025open weightshas scoreslast 18 monthsSAPSAP-RPT-OSS (sap-rpt-1-oss)SAP, tabularReleasedJun 2025open weightshas scoreslast 18 monthsKuaishouOneRec-Foundation 1.7BKuaishou, recommendationReleasedJan 2026open weightslast 18 monthsKuaishouOneRec-Foundation 8BKuaishou, recommendationReleasedJan 2026open weightslast 18 months

Showing 10 of 41. All models of this class

Where this comes from
Derived, not researched: models in the directory whose class matches a class this use case uses, excluding family records and models shut down or deprecated. Ranked by how many fit facts they meet, then by release date. Each fit fact is read from the model record.

Sources

  1. info.tmforum.org/AI-powered-service-hub-boosts-monetization-an... https://info.tmforum.org/AI-powered-service-hub-boosts-monetization-and-customer-engagement.html
  2. googlecloudpresscorner.com/2024-06-14-Indosat-Ooredoo-Hutchison-and-Goog... https://www.googlecloudpresscorner.com/2024-06-14-Indosat-Ooredoo-Hutchison-and-Google-Cloud-Deepen-Strategic-Alliance-to-Deliver-AI-driven-Digital-Experiences-Across-Indonesia

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Subscriber Personalisation, Next Best Offer and Churn Reduction", https://aurion.semai.id/usecases/tel-subscriber-personalisation-churn/, records of October 2026.