We use it about every day for our own products, specifically for finding performance issues in development. For example, one of the things we found in this release was that we had more updates than we expected compared to the previous version. We learned it because we saw a peak in “updates” in the row activity graph in the Enterprise Monitor. We highlighted the peak in the updates, and we figured out exactly which part of our code related to these updates through the Query Analyzer, and finally we fixed the problem. This process would have taken much longer if we did it manually because it was actually a certain set of updates that were increasing at a certain point of time, while other normal updates were happening at the same time. If we just looked at the logs, it wouldn't have been as obvious which one was causing that increase.
In addition, our support team also runs the Query Analyzer on internal servers, and they actually found some problem queries. In some occasions they solve performance issues that they didn't know the root cause; in other situations they see the performance drop and have actually been able to tune some queries that they didn't even know were problematic.