That they are still picked by the company. That makes them suspect to begin with.
Let me give you one sample of what DD entails:
If you're going to buy a factory with a warehouse in another town you could believe they have a warehouse in another town, or you could go there to look at the warehouse, make sure it's up and running and that there is stock roughly in accordance with what you were told. You check the ownership of the lot and the building if the company claimed it is their property. You count the number of employees, how many trucks there coming and going during your visit and what got loaded / unloaded.
Of course 99% of the time you'll find that everything is in order and that any errors are accidental or immaterial. But that 1% (actually, probably more than 1% but never mind) of the time where things are not as they should be you realize what DD is for.
It's not a rubberstamp, it is a verification that the picture the company painted was error free and complete.
Asking the company what the address of the warehouse is would in my opinion not be enough, even though you could assume that all is as it should be.
I've had the weirdest stuff surface during DD's, and I've had a few eyebrows raised for stuff like 'ok, I'll visit your warehouse tomorrow please alert them' (or other stuff in that vein) but for real companies with actual products and honest management this is never any problem, at most an inconvenience. Whenever there is significant pushback to such a request you can bet there is trouble and it usually doesn't take long after that to find something really bad (which typically signals end-of-deal).
Letting the company pick the examples or the results in the case of a company that provides test equipment is not an option, you'd want to walk in with your samples have them analyzed on the spot and follow the whole chain from entry to results and then you'd compare those results to a control that you possess. Anything less would not do.
Otherwise, why do DD at all? If you're going to go on what the company provided then you may as well skip DD and become a true believer.
If you present results that are cherry picked, or problematic in how they were obtained, how much of a practical difference is there to your example?