The advice all reviewers give is don't go off just one or two reviews, use half a dozen at least. There will always be outliers and individual reviewers have different methodologies and different biases.
Synthetic workloads might have the "downside" of not being "Real World use", whatever the hell that is, but they do have the upside of being much more consistent.
Also, one reviewer's work is only valid against their own testing. As long as you are comparing the Tiniest One with his own work, you have a valid comparison. Not so much when compared with others who might have different controls and different variables.
yeah i know. and i don´t want to badmouth the 980 pro.
i have 4 of them and santa just brought me 2 more.

i bought them because compared to the 970 pro they offer a good value (209 euro for 1TB here).
but the thing is, to have a real insight into how a SSD performs you need more than synthetic benchmarks.
Synthetic workloads might have the "downside" of not being "Real World use", whatever the hell that is,
let say your a photographer that fills up the SSD quite often when he returns from a shooting.
in a way that the SLC cache runs out. how well perfoms the SSD then?
there are quite some differences how badly SSD models will behave.
how are actual application load times?
for most gamers the great looking numbers of PCI 4.0 SSDs yield next to nothing (on a PC) in real world usage.
how many gamers now that... not many from my experience.
they look at 3500 vs 7000 MB/s and think it will be twice as fast.
some would actually be better of to buy a bigger and slower SSD than a smaller and faster SSD.
i could give a few more examples but i think it has become clear what i mean.
running a few benchmarks is something everyone does.
my grandma can do it.
if you want to give your readers (as a youtuber/reviewer) more "value" then you need to dig a bit deeper.
you don´t judge a car only by how it fares on perfectly prepared racetrack.
and while the 980 pro is sure a nice SSD... some of the changes they made have negative effect on performance. if/how you are affected is a question of your usage... but you need to know about it first.