Case Study: Appen cuts support resolution from two weeks to under 24 hours with Provectus

Appen resolves 80% of support tickets automatically with Provectus

Appen, a global AI training data company, faced a challenge in supporting its vast network of over one million contributors. Its small, two-person support team was overwhelmed by more than 11,000 tickets per month, leading to resolution times of roughly two weeks. This slow response threatened contributor satisfaction and retention, which are directly tied to Appen's revenue and data quality. Appen partnered with vendor Provectus to use machine learning to automate and improve this process.

Provectus built an ML-powered support system on Amazon SageMaker that automatically classifies, routes, and resolves tickets. The solution cut resolution time from two weeks to under 24 hours and automatically handles approximately 80% of all tickets without human intervention. This resulted in a 10% increase in contributor satisfaction and improved retention, allowing Appen's support team to focus their efforts on complex issues that require human judgment.

Appen