The data doesn’t provide insight into the question. Those percentages are right in line with what I would expect for people using the product normally.
The tone and phrasing of the question is kind of strange, too, as if the author expected something else and is disappointed.
The data he should look at to answer “what to do?” would be the user requests to do things differently, add features, fix bugs, etc. After 8 years there must be a bunch.
[Comment removed by author]
One of the comments on the original post mentioned that in a car the gas cap is used far less than the steering wheel, but arguably more important. The point being that numbers alone don’t tell the whole story. We have to understand the questions we are asking and their true meaning.