Matching Loyal Employees

Employee loyalty is an important measure. Since the costs of replacing even an "adequate" employee can far exceed the internal costs of improving that same employee's loyalty, ensuring that employees plan to stay with the company is crucial to long term business success. Of course, employee loyalty is only valuable when the loyal employees are those that bring in revenue. Employee loyalty among easily replaceable (or soon to be replaced) employees may be less important, depending on how your company operates – at least comparatively.

Still, one of the problems with any employee research is that employees lie when they think the company is going to judge them based on their answers. If the data you receive isn't valuable, you can't hope to make good decisions based off the information. If you attach the results of the survey to the name of the person that filled out the survey: The person is less likely to give honest answers.

Truly honest answers can hurt the relationship between the employee and management. The data collection itself may be unethical. That’s a considerable challenge, and one that doesn’t have an easy answer.

But one potential solution is to use a coding system that happens behind the scenes and is only viewed by one or two people limited to human resources. The employees can fill out the survey anonymously, and each given a code. That code can then be linked to (further anonymized) employee evaluation data, or those that perform evaluations can fill out information on each employee, and that data is then connected to their code.

Either way, this will allow you to run analyses that will tell you the amount of loyalty of the employees which you can then check against the rating system of employee productivity.

You will not be able to know exactly which great employees are disloyal, or vice versa, but it will give you a great starting point for knowing what to target, and from there you try different methods of improving that loyalty in general. This may be the better method anyway. You are risking your research ethics if you label each employee by name, and you are especially likely to get inaccurate responses if the employees found out. So this method gives you a useful overall score that you can use to target all employees and try to improve loyalty out of your best ones.