I think Michael Gilmour has one of the sharpest minds in the domain business, but I am going to play devil’s advocate to his post today about buying based on revenue multiples. In today’s post, Michael provides some guidelines about what someone might expect to pay based on the type of name they are buying:
“I’m really going to stick my neck out here and state some revenue multiple ranges for different types of domains. For the purposes of this example let’s imagine that roysfood.com has a trademark and is a small business in Utah.
Type of domain Example No. Months
Direct TM infringing from heavy TM defending company microsoftword.com 0-3
Direct TM infringing from non-defending company roysfood.com 6-12
TM typo of heavy defending company micorsaft.com 3-9
TM typo of non-defending company rosyfod.com 12-18
Typo of a generic multi-word domainperking.com 36-48
Typo of a generic single-word domain.com 48-60
Generic multi-word domainparking.com 60-72
Generic single-word domain.com 72+”
In my time in the domain business, I have never purchased a name nor have I sold a name based on any type of revenue multiple. Incidentally, I had a long conversation with a successful domain investor today about this, and I believe that buying or selling based on a revenue is very risky and shouldn’t be done by anyone but a domain expert and/or domain actuary.
Reasons why I think you shouldn’t buy (or sell) based on a revenue multiple:
- It’s very difficult to determine how much a name can earn based on different parking companies, different revenue shares, different landing pages…etc. Whose revenue do you use for the multiple? Do you count on someone else’s revenue share which might be considerably higher than yours? It would be in the buyer’s best interest to have revenue be as low as possible during a traffic test. Likewise, it’s in the seller’s best interest to earn as much money as possible while the domain name is being tested. If the buyer can’t replicate the exact conditions the seller had when he was selling, the buyer may never see anything close to the quoted revenues.
- If the name is dependent upon search engine placement, what happens if Google/Yahoo update their algorithm, causing traffic (and consequently revenue) to plunge? The buyer could be screwed in this situation. This is especially difficult if the new owner changes DNS or does something that could catch the attention of search engines. Even a change in the Whois or registrar could possibly impact it. The reality of the situation is that the search engines are powerful and mysterious. We don’t know exactly how they work, but we hope things we do can help boost rankings.
- How does a buyer know if the traffic is “real” or if the seller is asking his buddies to do a little clicking on the PPC ads. What happens if/when the traffic dies? Hypothetically, a domain name that earns an extra $1.00 per day is worth around $3,000 more on a ten year revenue multiple.
Maybe I am wrong, but I don’t think there are many people out there willing to sell their revenue producing generic domain names simply based on a 72 month revenue multiple as suggested by Michael, or even a 120 month multiple. If there are, I would be suspicious, just because it sounds like it could be too good to be true.
As I said in my post about the Art of Pricing a Domain Name, the most important factors for me in determining a price to sell and to buy are the following:
1) Traffic the name receives
2) Revenue the name receives
3) Google listings for the “bracketed term”
4) Advertisers on Google
5) Comps of recent sales
6) Gut feel
Revenue is certainly important, but no way would I buy only based on a revenue multiple. It’s good to know what kind of revenue potential the domain name has based on what people are looking for when they navigate to the site, but it’s not close to being the main factor.
To me, buying or selling a domain name simply based on a revenue multiple is a losing proposition for both parties.