Everyone has their way of measuring things. Quite often the way you measure something could be different to the way I measure something. And the way I measure something might not conform to generally accepted principals. Is that wrong?.
The answer to that is both yes and no.
If you seek to compare your measures to Industry benchmarks then you will need the same computational methodology to make the comparisons relevant. Otherwise you will be comparing two unrelated metrics even though you may call them by the same name. If your dashboard includes comparisons to industry benchmarks then calculating those numbers in the same way to Industry Standards is obviously important to obtain any value out of those metric comparisons.
Internally it is not so important as long as you are consistent in your approach each time you measure the metric. In fact you can quite easily create a key performance indicator that is basically by normal standards. This could be by design or by error. But it matters little. Once you establish a benchmark measure and apply consistency to that measure moving forward, your decision making capabilities on that measure will not be compromised. You have actually just set a new "innovative" metric benchmark.
There is no right way or wrong way, it's your business, measure your metrics anyway you like. Just ensure consistency in arriving at those metrics and your management ability will not be effected.