Now, I often pondered (here and here, for example) why people chose to build something to do log management and then realize that vendors are in business of doing it for a good reason :-) Namely, that their stuff works perfectly ... until it doesn't. And then mayhem ensues!
Our brilliant field engineer, Dimitri McKay, has to face such situation pretty often and he is frustrated as well: "So tell me why day after day I see administrators and engineers hard nosed about building their own solution? Why do I have to show them day after day that there’s a much more efficient way of managing those logs which will free up resources for a zillion other things."
I would guess that for every million of dollars paid to log management vendors, there are 10 (ten!) millions flushed down the drain by building home-grown logging tools that then get abandoned and decay there on the side of the road ... sad indeed.