I use a set of custom Perl scripts. They do several things but the main one goes through the event logs of each of the servers, extracts the warnings and errors for the last 24 hours, creates an Excel spreadsheet with the results and puts that in a folder I check each morning. This way I get the interesting bits in an easy to read format.
I'm currently considering the practicality of integrating event log monitoring with Nagios. With the right kind of ignore filters (e.g. I really don't care that a print job failed) I should only receive alerts that need to be looked into. That's a fair bit of work to set up but I only need to do it once and it will make my job easier long term. Alternatively, I may have another look at Zenoss.
I can recommend NOT using GFI event manager. It's a huge bandwidth hog (streams entire logs over the wire repeatedly) and if you have t1 remote sites, this is a big deal. Also, it's flaky, so it requires far too much upkeep for what should be a simple solution.
SecureWorks has a solution that we will be trying. It looks pretty good.