Recruiters know that data is the key to improving hiring. Over the years, what data is available and how to use it has changed.
If you have been around an exceedingly LONG time like myself, besides being “ancient,” as my daughters like to say, you also remember how exciting it was to have an automated applicant tracking system with standard reports for recruiting. The new applicant tracking systems (ATS) allowed us access to data with the click of a button. You could see how many applicants in a period applied to a specific location or how many employee referrals you hired over the year. The magic of having EEO reports at our fingertips was pure joy.
ATSs were building new standard reports at lightning speed. Companies were competing to see who could offer the most standard reports. Then recruiters began to demand custom reports. Suddenly, we had access to hundreds of standard reports and the ability to have custom reports. These reports enabled us to analyze more data faster, giving a good overview of our hiring process. Unfortunately, even with the availability of reporting and data, there wasn’t any impact on our future hiring decisions.
Once reports were no longer exciting, dashboards with graphical data representations were the new trend. A recruiter could start the day by opening a dashboard with all the critical data represented graphically. The graphs made it easy for recruiters to share metrics with management. The graphics could be quickly added to a PowerPoint presentation with a quick screenshot.
Soon benchmark reports were added to the dashboards. We could see how we were doing compared to others in our industry or how we did in the past. Benchmark reports aimed to identify problems quickly by using these comparisons.
We also saw the move to reporting meant to give us answers to questions like why “time to hire” was so long or why so many applicants declined job offers. With all this data so easily accessible, there were high hopes for improving meaningful business outcomes. Regrettably, all this data did not analyze itself. Recruiters were still missing key data points around new hires and employee data. Again, we could still not improve business outcomes like increasing retention rates or making better-quality hires.
A whole new idea emerged called Predictive Analytics. This new type of data science allowed us to look forward to using past data—a great idea if your past hiring decisions were without issues. But if you previously hired poorly, you don’t want to reuse that data. If you do, you will repeat those poor hiring decisions.
Over the last couple of years, the science of data analytics has continued to improve. It’s ushered in the era of business intelligence for recruitment - where augmented intelligence, analytics, and data modeling are combined in one model for data-driven decision-making. This type of data model is smart. It continues to learn as new data is available, improving the model and enabling hiring decisions to get better over time.
The one thing we’ve learned is that data must be easy to understand and apply in real-life situations to have any impact on business decisions. Reports are great, but recruiters need augmented intelligence, analytics, and modeling to make data-driven decisions.
Data science in recruitment is still a new field, and its applications are only starting to be understood. We should expect to see a continued focus on improving business outcomes with data science in the recruiting space.
Enjoyed this article? You may like "AI From a Layman's Perspective" or "Let's Get Analytical".