The Name Inspector has a good post today regarding 6 naming myths to ignore. I think it’s generally right on. Naming is so important and so incredibly hard, especially for consumer internet companies that not only have to find a good name but also get the URL. I am convinced that a big part of Twitter’s success, for example, is it has such a great name. Simple word, easy to spell, great imagery, and also evocative of what the product does without being overly literal.
I have been involved in naming a number of startups, including my two most recent companies: Hunch and SiteAdvisor. Each time it was a long and painful process. Here are some things I’ve learned along the way.
1) Probably the most important thing is that the name be easy to spell after someone hears it pronounced. I was involved in one startup before where every time you said the name the person says “what?” and then you have to spell it. Trust me, it becomes really tedious and also adds friction to word-of-mouth buzz.
2) You should have different naming goals for different products. For example, SiteAdvisor was a security product. You really can’t make security “cool” so we didn’t even try to bother to do that with the name. Instead we went for a name that helped explain in a very literal way what the product did. Before we came up with the name SiteAdvisor, I probably had 100 meetings where people said “I don’t understand what you are building – is it an anti-phishing toolbar, a spyware blocker or what?”. This included meetings with VC’s who focus on security and other experts. I knew the name SiteAdvisor was a winner when my father in law wrote the name on a high school blackboard and asked the kids what they thought the company did and one kid said “They advise you about websites” (and then he said ” … or construction sites” 
). Also we liked the name because we imagined in the future doing more than just security – for example warning about adult content. (Alas, we never got that far).
3) I tend to disagree with The Name Inspector about name length. Shorter is definitely better. In particular the number of syllables is important. SiteAdvisor, while good at describing the product, is really clunky to pronounce. I also tend to really dislike Latin-y portmanteau names like “Integra” “Omnitrust” etc. Sounds like a pharmaceutical product.
4) A few things I’ve learned about methodology. I think it’s very rare to have an epiphany where you come up with a great name. First of all, even if you do, the domain is probably taken and too expensive. For systematically brainstorming, I really like the Related Words function on RhymeZone. I try to make lists of words that are sort of related to the product and then look at all the related words, look at all those words’ related words, etc, making lists of words and word fragments that sounds good. Then I have a systematic process for checking domains to see if they are buyable. If you are super lucky (and picked a multiword domain name) you might get it retail, but at this point almost all .com names (yes, I think you still need to own the .com) are owned by someone and the question becomes whether they will sell it at a reasonable price. The best case is usually that it’s owned by a professional domainer and it’s not very monetizable via Adsense (domainers make a lot of money from Adsense on sites like candy.com so you’d need to offer them a tons of money to sell it).
Naming is tough!
