Precedent acquisition analysis is my favorite valuation method. I'm not sure why. Maybe it's because I can always find two or three golden nuggets, or maybe it's because it takes a lot of research and sleuthing to find the numbers. Using a pre-acq has its detractors. They say that the information is stale, no two deals are alike, the numbers are incomplete or why base valuation on past deals when at least half were incorrect. I still think that a pre-acq done right and in the right context is a valuable tool. Most buyers don't use pre-acqs. I usually use them defensively and then offensively.
Most pre-acqs I see were prepared by an investment bank and include 15 - 20 deals. The deals were pulled from special third-party databases by entering a lot of specific parameters like SIC code, country, industry, keywords, size, etc. This spits out about 100 deals which are narrowed, as more than half the deals lacked any data and the others "just didn't fit". So when you present this to me I focus on the parameters, and after a discussion will probably agree with your parameter selection. Now show me the "just didn't fits". This usually causes some fidgeting and your banker explaining that they're not relevant so they weren't included. Great, let's call that I-banking analyst and have him/her rerun the whole thing and send it right over. When I get the spread sheet, I find that many of the "just didn't fits" actually do fit pretty well and plenty of the 15 - 20 don't really fit at all. This exercise always causes the multiples to decrease dramatically.
Don't wast your time with these databases. It's the wrong approach completely. With pre-acqs less is more. All you need are two to four high-quality matches to make a case, and you probably know off the top of your head which deals those are. Then spend your time building intelligence around these matches, so you can make a strong argument as to why they're a good fit. If these matches are too low for your valuation expectations, deal with in a realistic and honest way.
The second mistake is lack of documentation. VCs try this a lot, and I've seen some buyers go along with it. When I get list of merger comps with acquisition multiples, I always ask to see the documentation - database sources, press releases, SEC filings, conference call transcripts, etc. The majority of time there isn't any. Sometimes this information just needs to be found, but often it isn't documentable. If you can't prove it, I wont use it. The funny thing is that a lot of the information is out there, you just have to take the time and know how to find it.
I once represented a buyer who was looking to make a defensive acquisition of a target whose two primary competitors had been acquired within the last 18 months. All three companies (target plus two competitors) were private and the leaders in their segment. In this case a pre-acq with only the competitors was not only important but crucial; however, not enough information was initially disclosed. Our valuation discussions were pretty far off, but after a week of digging I was able to find enough data to connect the dots and come up with documentable multiples. This was very compelling information to the target, whose banker presented 32 precedent acquisitions to back their valuation. We ended up doing the deal at a premium to our pre-acq multiples but for far less than the original value the target wanted.