Say you want to implement a new training program for sales reps. But every time you propose a new HR initiative like this, executives question whether it can work or say the company can’t afford it.
Next time, take a different approach. Ask to implement the program for only half of the sales force. Measure and compare the results before the test program begins and after six months. Employees who undergo training should show improved results, bolstering your case for funding to implement the program with the entire sales team.
The technique is called split-sample testing. Marketers use it all the time to evaluate which promotions work best. You should, too, to (1) figure out whether an HR initiative works and (2) prove its value to executives. The technique can improve HR’s reputation as a department that supports organizational productivity.
Split-sample testing can work on most, but not all, HR initiatives. Focus on areas of HR that are easiest to measure and most likely to yield the best results:
Sales: Compare the sales results of every salesperson before, during and after the test program. Ask salespeople what worked, what didn’t and if they received adequate training. Make adjustments accordingly and then roll out the program to the entire sales force.
Recruiting: Launch an employee referral program with a certain business department or unit. Compare the productivity and quality of employees hired through the referral program with those hired through normal channels in other departments.
Compensation: Offer financial incentives or pay-for-performance programs to some teams of employees, but not others. Make sure those choices don’t discriminate against any protected group of employees, such as minorities.
Retention: Target certain retention efforts at a group of employees and compare the results.
Final tips: Ask the CFO and department managers—who all have a stake in the results of the split-sample test—to help design it. That limits criticism when HR reveals the test results.
Prepare for the possibility that your split-sample test won’t prove that you should continue your initiative. That’s OK. It’s important to know what not to do, too.