Not to be a sage here, but "every research is an average taken in the past on other people".

What works for the average is likely a good starting point. It is not necessarily the best for you and a given case (almost for sure different from the experimental ones).