Timing and Satisfaction

While most companies and researches do not acknowledge it, satisfaction scores can be directly linked to when they were run, rather than the actual satisfaction themselves. You need your data to indicate how satisfaction is “at rest,” when nothing is clearly altering the results one way or another in a way that is ultimately temporary. But both customer satisfaction and employee satisfaction can easily be influenced by time factors that may interfere with the value of the data. Two examples are described below.

Employee satisfaction after a major event can alter the data. For example, if you run an employee satisfaction survey on January 1 st , you’re like to get scores that are higher due to the holiday season, yearly bonuses, vacations, better mood from management, and other factors that can accidentally influence your data. It’s also possible that these factors are not consistent each year.

If bonuses went down year by a very small margin (enough to be noticeable, but not enough to be a long term issue), satisfaction scores are likely going to be down a bit compared to the year before, but only for a short time after the bonuses were provided. A coinciding satisfaction score would then yield inaccurate results.

The release of the iPhone 4 is a great example of how customer satisfaction can easily change as well. When the iPhone 4 was released it received extremely high praise and customer satisfaction would have been high. Then, just a few days later, it was found that there were connectivity problems if you put your hand near a certain spot on the phone, and satisfaction was incredibly low.

Then they gave out cases and satisfaction was medium. Now satisfaction is fairly high again. These scores certainly have meaning, but if you were trying to measure trends with Apple customers, the resulting differences would have been pronounced, with few of them necessarily indicating long term satisfaction.

There is no easy way to make sure that your survey was run at a time that is yielding the most accurate results. The best thing you can do is to always make sure you’re analyzing not only the data, but also all of the potential events that may influence the data. This can give you insight about how to read the results and ensure you don’t make any drastic decisions about data that may not be relevant.