A large telecommunications company wanted to find new ways to reduce employee turnover and improve talent retention. Over 90,000 people’s sensitive data needed to be collected and analyzed to identify patterns in employee churn. However, legal regulations locked up the data, and manual anonymization of datasets for each analytics project took 6 weeks on average.
The HR Analytics project took off when readily shareable synthetic copies of siloed HR data assets were created and distributed. The resulting synthetic data was no longer classified as personal data and so it was exempt from legal regulations. The synthetic copies were statistically highly representative of the original data, enabling the analytics team to find the same insights they would have found in the original.
Using synthetic repositories, the analytics team:
- detected patterns leading to employee churn
- identified employees most at risk
- developed interventions to retain talent
Knowing when and why people typically quit will enable the company to take extra steps to identify and retain high-risk employees in time. Reducing turnover rates by only 0.5% could already result in double-digit million cost savings and positively impact team productivity