Productivity measurement is hard. The author points to many of the difficulties. We (economists) have been aware of many of these difficulties since the 1930’s or earlier.
No one in this literature thinks “hedonic adjustments get it exactly right!” to take just one example. Instead they think more along the lines of: this is the best solution we can come up with to this very hard problem.
For an informed, recent survey of the difficulties and the importance of getting these questions right, try Syversen and De Loecker’s survey “an industrial organization perspective on productivity” from the recent Handbook of Industrial Organization, available for free via the NBER working paper series.
Why might you care? Here is one fact from that survey: within a narrowly defined industry category, the difference in measured output from the 5th to the 95th percentiles of the productivity distribution can be a factor of 2! To emphasize: two firms with the same measured inputs, but one produces twice as much output than the other. This is visible across many industry categories. It’s a fascinating problem and well worthy of study.
The original author should probably also read “the sad truth about happiness scales” from the Journal of Political Economy in 2019. Such scales are unidentified in quite a fundamental way and the results they give are essentially meaningless.