I've never had the impression that they're corporate shills or bought off or anything like that.
But what I have noticed is that the results of their studies aren't weighted by the same things that I might value. For example, when buying a washing machine, they'll weight stuff like energy/water efficiency very highly, and I (personally) would rate cleaning performance and reliability much, much higher than efficiency or noise, for example.
To some degree, that's an IT issue in that they could easily enough devise a web interface that let you rank the data elements as you saw fit and come up with your own rankings, but that would steal a lot of their own thunder in providing "Recommended" models and what-not.
Or worse, they use the "Normal" cycle on things like dishwashers and washing machines for testing, when the very clear implication from the owner's manual is that they optimized the "Normal" cycle for the EPA efficiency tests, and that the "Heavy Duty" cycle is the better one for most uses. I know CR thinks they're leveling the playing field, but it also distorts the actual results in situations like this.
The other thing I've run across is that their selections are often somewhat limited, or aimed at a different market segment than I'm in. For example, they review and rank pillows, but they're all name-brand pillows. There are no Walmart or Target house-brand pillows in that mix.
So to sum up, I think it's useful information, but more along the lines of a few more data elements rather than a review I can necessarily take to the bank. Still better and more rigorous than some boobs on YouTube with videos showing tests of questionable methodology though. (Project farm, I'm looking at you.)
