I've seen a few comments on here over the months that a 1-2 hour delay isn't that big a deal. However, I'm of the school of thought where the delay, when measured as a proportion of the flight time, becomes an issue if it's more than around 30% of the flight time.

So, take my regular UK domestic flight which is a 50-minute flight (elongated to 1 hour 10 minutes by BA to absorb 20 minutes of delay/taxiing on the ground). If this flight is delayed by more than 15 minutes (which is 30% of the actual flying time) then that is unacceptable to me.

Yet, these flights are regularly delayed by anything from 10 minutes to 2+ hours - that's more than 200% of the actual flying time for the 2+ hour delays.

Currently the delay is measured as an absolute value, but I think that a better way of assessing the impact of a delay is as a proportion of the flight time. If you're on a 10-hour flight and you land an hour later then that, psychologically, is less frustrating in my opinion.

So, I'm of the opinion that airlines should be using the relative delay values to measure their performance (if they don't do so already). These values should also be used for compensation claims, rather than ridiculous absolute values of 3 hours that are currently used (for EU compensation claims) - certainly for flights where the flying time is under 3 hours.