Extra careful, is not redundant. If your load matches up good with published data that seems to make sense, loading to the max listed is no big deal. If your stuff does not match real good, either get data that does match, or back off and do not go to the max listed. I added the qualifier about the load data, because I really like to compare all the data I can get (more on this below).
Using a chrono can work Ok as long as your gun matches up pretty close with the test gun. For revolvers, using a chrono is much less effective than most other gun types. Revolvers have lots of variables that can make a given gun shoot slower or faster than other guns.
The Quickload software is not perfect, but it is very predictable and completely repeatable. When one load from a published manual shoots a good deal below (velocity and/or pressure) Quickload and the next load down in the manual shoots above Quickload, I get suspicious of the published data. I have been around enough data gathering efforts to understand that any test data can potentially be off. I have seen plenty of variations in max charges for the same bullets in the same cartridges from different sources.
I recommend not using the hottest data you can find or pushing limits unless you know your gun has plenty of margin. I personally prefer getting a bigger and/or more powerful gun when I want more power instead of taking silly chances with a gun that will work just fine if not pushed too hard.
Again, the biggest risk is for specific combinations of cartridges and guns. A lot of rounds are not assured to show "pressure signs" at under SAAMI ratings. Combine this with guns that are not known to be high margin guns and you are in the risk area. This really is just common sense. I continually get amazed by the pushback from people that either do not like it or do not get it.