I think it's also important to consider that satisfaction surveys tend to suffer from a sort of selection bias. You're only getting feedback from people who feel compelled to give feedback.
I worked for a big, big global company for a number of years. IT was run from call centres around the world, so that anybody working anywhere would always find somebody in a convenient timezone to work on the problem for at least a few hours, even if the ticket was submitted at the end of the business day for the person with the problem.
And if the IT people fixed the problem quickly, there was always a request to complete a satisfaction survey.
Curiously, if they didn't fix the problem quickly, there was no such request.
This kind of situation is not limited to customer surveys, it applies to all manner of Key Performance Metrics and other measurements.
Where I live right now, there is a straight section of road through a residential area.
There were complaints to the town police about cars being driven at excessive speeds, and requests for traffic calming measures. The town opposed the expenditure.
To prove that the measures were not necessary, a few police officers were given the task of measuring average speeds along the road. Of course, being deployed along a public road they had be be safe, so they wore high-visibility jackets at they pointed radar speed-detection guns at oncoming vehicles. And guess what? Not a single vehicle exceeded the limit during the operation. Ergo, no need to install expensive traffic calming measures.