Philips stopped just auditing suppliers and predicted who needed help

Traditional supplier sustainability audits tell a buyer what went wrong, but they are expensive to scale across thousands of suppliers. Philips needed a way to improve performance with limited data, not merely inspect more factories.

The CARE framework reframed the problem as collect, assess, react and enhance. Machine learning inferred sustainability levels from general supplier information such as location, size and sector; optimization then chose improvement actions.

Philips used the framework in its supplier program. The published study reports a 24% year-on-year sustainability-performance improvement in 2021 for suppliers that entered the program in 2020.

Turn audits into improvement routing

Supplier self-assessments are cheap enough to cover a much larger base than repeated audits.

Prediction from general supplier attributes lets the buyer focus scarce expert attention.

Optimization turns risk scores into action choices rather than dashboards.

The program measures improvement, not just compliance failures.

A supplier program scales better when analytics decides where help changes behavior, instead of only finding violations after audits.

The CARE approach gave Philips a scalable alternative to audit-first supplier management. Its reported 2021 improvement among enrolled suppliers was 24%, supporting the case for proactive analytics-led supplier development.