You can't buy someone almost as big as you and get away with leaving it alone. You have to integrate it or Wall St. will fry you. If one plus one equals two instead of three, then what was the point in the first place? Only giant holding companies can get away with buying smaller operating companies and letting them run on their own. Warren Buffet's Berkshire Hathaway is one such player. The companies that are purchased and left alone live and die on their own, with leverage from daddy. Those that are integrated cause the lions share of problems in the digestive tract.

Veritas had to be integrated by Symantec; it was a merger of equals essentially. There was no point in paying a premium to own Veritas unless you could take all the revenue and lose a bunch of costs. The devil, alas, is in the details. Symantec spoke about having problems integrating ERP systems (see ""). Imagine, two software companies having trouble integrating software systems -- maybe God is getting even. It would be like two GPS companies merging and blaming a blown quarter on everyone getting lost going to work, or two of the Big 5 joining up and having to restate earnings due to accounting irregularities.

Devouring a company is never easy, even under the best of circumstances -- cultures clash, human nature and ego's interfere, and politics are played. The trick is to keep the good and clean away from the bad, and that's not so easy. EMC has acquired a thousand companies, or so it seems, but it hasn't fired a whole heck of a lot of folks, which means they are keeping them, paying them, and not necessarily getting any incremental leverage from them (see ""). VMware is a resounding success - that was kept separate. Documentum, RSA and Legato were huge deals, and it is not easy to see from the outside if these deals will be home runs or bombs. FilePool was a technology buy that turned into Centera, and that's a billion dollar business now. With all the acquired cultures running around EMC these days I can't tell if it reminds me of the U.N. or the bar scene from Star Wars.

IBM, HP, Netapp, etc. all buy folks, and all have the same challenges along with the same variety of successes and failures. Many have made outright bad moves, either buying technologies that didn't fit or companies that couldn't fit. Oracle usually has to overpay because everyone on the selling side knows that Larry is going to gut the place on day one. If cost and operations are the issue, then that's the right play - as long as you don't need folks to hang around and help with any lingering issues.

I don't know that I can point to any one company in the IT space that has mastered the art of acquisition. Some have used it as a hedge (IBM), some for growth (CA), and some for unknown reasons altogether (Sun) (see ""). Many combine them all.