Advanced CSAT Data Analysis

Today we're following up on another article (Thought on Customer Satisfaction Data Part 2) that, at this point, many of you will have already forgotten. How many follow ups will there be? Stay tuned! But in all seriousness, in the past few posts, we've discussed the idea that your company should focus both on averages and data distribution when it comes to analyzing and sharing the results of your data.

Averages don't always tell the whole story, and data distribution is where you can see the differences/changes that your customer satisfaction efforts provided. It's with that in mind that we'd like to offer another suggestion: considering both types of analysis when you create employee incentives.

The most effective way to make employees interested in customer satisfaction is to offer incentives with regard to satisfaction scores. For example, if your current satisfaction level is 5 on some arbitrary scale, you offer a bonus or some type of reward if the company reaches 6. This is a useful way to ensure that employees care about the satisfaction scores of the company.

With that in mind, it may be a good idea to create rewards that, while a little complex, are also fair to both the employees and the reality of these data. Consider offering a reward based on either an increase in average score or a change in distribution.

For example, perhaps you want the number of people that score within the 7 to 10 range to go up (without worrying about the lower scores), or you want there to be fewer low scores but are less concerned with the highest scores. This would make the bonus payouts fairer, because they would represent things of real importance within the company. It's not necessarily important to go from a 5 to a 6 in average if there are noticeable changes within the range, such as a 10% increase in customers within the 7 to 10 scores.

The average overall may not move if there is a corresponding decrease in scores on the other side of the table, but depending on your priorities this may still be valuable. It's something to consider when you're deciding on your employee reward systems. It may be more complex to involve both average and data distribution, but in the end it would likely be a system that rewards employees more accurately for their hard work.