From a trademark perspective, you don't want to be "books.com" - it's too generic.
From a mindshare perspective, you also don't want to be "wrejiorgsdf.com".
The sweet spot is just off-center of the topic. E.g. Amazon - the world's biggest bookstore. Your brand is distinct from (and hence more memorable than) the thing you sell, but it is EVOCATIVE of a trait that you are bragging about.
Many folks go with Latin and Greek roots (because they are the basis of much of English) to play the "evocation" game.
The names of my two startups are decent ,but not insanely great.
SmartFlix - the "smart" brings out the educational component (and also reassures customers that it's a wise decision to rent a video from us), and Flix has been a synonym for movies for almsot a century.
HeavyInk - the "ink" references the artwork and the history of comics, and "heavy" has a slight outsiderish / authentic connotation (heavy metal, "deep" , "profound", etc.) that also helps make the subtle point that we carry every comic that's available (far more than your typical bricks and mortar store).
And, of course, the important thing is that both domains were available in the .com TLD.
Travis J I Corcoran, President SmartFlix
-- http://SmartFlix.com/ web's biggest how-to DVD rental store
* SEO. Along with the fact that being too generic means you won't have access to good trademark protection, it also means that you'll never get anywhere in search results. A fantastic amount of people still type the name of what they're looking for into Google (instead of just <name>.com). So, if your bookstore is named "Books", you'll never be found. That's not a battle you likely need when first starting.
However, if you run a site that is going to depend on Google (search engines) for traffic, you also might want to think hard about including some terms people will search for into your domain. Search engines weigh domain names heavily, so it does help. You can overcome this by getting linked to early and often, but if you really need Google, you might want to think about this.
The flip side is that using a made up word will ensure you are the _only_ result in Google and don't have to fight for position. So, if you're planning on getting your traffic elsewhere, don't want to focus on SEO, and just want to ensure that you are at the top for your name, making up a word can be good.
* Spell-able. Making up names leads me into my second thought. If you make it up -- or even if you don't but it's not a commonly known word -- make sure it is spelled how it sounds. Think about Google. Correct spelling: Googol; but no one would've ever done that correctly.
Openomy was the perfect example of what not to do. No one has any idea how to pronounce this word correctly, let alone spell it. Links would get botched, people couldn't get to the site, every time I talked to someone about it they would ask "how do you pronounce it", etc.
3. Don't be afraid of prefixes/suffixes. With domain availability running thinner every day, you will have to be creative. This will probably mean some combination of splicing words together (NetFlix) or using prefixes/suffixes. Don't be afraid of this, just careful. Some suffixes I like are -ly, -ware, -ish, -ize, -er, -it.
4. Talk about it and sleep on it. The way I came up with my latest project name is by writing down all the words that I could think of which were either directly or indirectly related with the field, as well as all the prefixes and suffixes I could remember. Then I went to town coming up with combinations of these things until I found a few I liked. I took that short list to my partner and we decided on one. We preemptively bought the domain name, but we didn't necessarily settle on it. We kept using it as our "name", making sure that even a few days later we liked it. Sleeping on it and using it as if it were real, without being religious about it, helps ensure you actually like it.
Then I started asking a couple friends/family about it. (1) Did they like it, (2) could they spell it for me, (3) what did it mean to them? If you can pass these tests, and you're happy, you've probably found a decent name.
younar youner yuner
I guess it depends on how you say it and how people pick it up - ask 20 people you know to spell it for you after you say it just once to them and see what the results are like ...
If you can't read your name over the phone without clarification, pick another one.
ps. I'd vote against going with Yunar. Besides having 7 plausible spellings, it just sounds bad.
The goal is to come up with variations or combinations of words or related words. Use sites like morewords.com which can show you words starting with a word, ending with a word, containing a word, etc. You can try Google Sets, but it's not very useful for coming up with new stuff. Also do general Google searches for your words and see what other words pop up on the result pages. During this period, carry pen & paper around so you can write down anything you think of.
For most good names you'll come up with the .com won't be available, but that's okay, just keep trying. Be prepared to check 20 domains to find one that is available. Be sure to use non-evil domain checkers that don't automatically register the name.
A good name should be easy to pronounce, and based on the pronounciation, easy to write down, even for foreigners (names with vowels removed, used at the height of the web2 bubble, are no good). It should be related to your project (eg. Amazon is not really a good name for a hypothetical startup bookstore).
You should have 5-10 good names (which may be close to each other) at the end of the process, consult with your partners and pick the best one.
Good luck!
;)
http://www.breadlinedesign.com/naming_machine/index.htm
Just something to keep in mind when trying to avoid naming pattern cliches.
In the end way came up with a company name via a lot of lateral thinking. I spent a long time writing down all the attributes of our service and then trying to find related concepts and words and finally narrow down a list of hundreds of possibilities to a few possibles.
We then used purewhois.com to check domain names and uspto.gov to check trademarks.
We also involved everyone in the company in the process taking suggestions from everyone.
I am not a "domainer" but by necessity of having tried to find some goods ones over the years I've accumulated around 50 that I hold on to "just in case". I am sure there are others here in a similar position as myself. I'm not saying HN is the best place for asking for domain name sales (it may greatly annoy some here), but maybe we are a friendlier community for having a few interesting ones in our pockets as opposed to the cold open markets?
Pick a code name and start coding. You can always pick out a domain name 15 minutes before launch.
- write down a list of words closely related to the subject
- get thesaurus and add the words I find as synonyms to the list
- make a list of permutations that make sense
- ask my registrar to do searches on all of them for .com, .org and .net
- rinse and repeat until I find a good set, if the .com is free but the rest is not then that is ok too, but not preferred.
The process can take up the better part of a day before I find a domain that I'm happy with.
Drop the E, Add an R
If the domain with a c isn't available use a K
Find obscure country TLDs to represent the last two letters
Doublee the lastt letterr in the domainn.
If all else fails, start with the name of your favorite Star Wars Character and work from there.
Names are delimiters; In your case you're delimiting your product in someone else's mind. And while names may be arbitrary, humans understand things in terms of things we already know.
Think of baby naming. I have a two standard names stuck together. It's an Anglican/Christian English name. Though I think Christopher may be derived from Greek. Maybe it's too generic. Compare that to Shanacqua-Tibea. Nobody would know how to pronounce it, and it's too unfamiliar to remember immediately. For social reasons, you might have to remember it (to prevent embarrassment), but that pressure isn't there when someone searches in Google.
I like a mix between the familiar and different. You need to hook them in with something they know, while adding a twist to it.
I would start with writing down descriptive works and just keep playing with them, creating different combinations, looking them up in the thesaurus, etc. I haven't had any luck with latin, greek, etc names, the approach always sounds good, but I find the results unrelatable (if that's a word). You want something people easily understand, starting with foreign languages is starting from the complete opposite side.
Also, don't expect to pick it in 20 minutes. Give yourself a week or more where you toss all kinds of ideas around in your head, let different variations sink in, bounce ideas back and forth between friends, etc. Keep coming back to it over and over, even if it's just for 5 minutes.
And one thing: Be brave with your name.
Then I opened up Google Translate and converted the words to Italian, Spanish, etc to see if any interesting variations popped up. And, finally, with this new list in hand, I started determining which ones were available.
Overall, I spent no more than an hour on the whole process and came away with a new domain name in hand.
IMHO, spend your time worrying about whether your startup actually solves a real problem for your potential users. If you can actually do that, the name won't matter... that much.
New features: http://news.ycombinator.com/item?id=483333
I'm also in the process of finding a name and so far it's been both painful and fruitless. I've tried to look for Greek and Latin names but most of those I tried were already taken.
I even used a Hawaiian online dictionary: http://wehewehe.org/ but again haven't come up with something decent. Hawaiian can sound pretty childish very easily...
1. Look for a metaphor - like for instance, "digg" because you're uncovering stuff, or "YCombinator" because it's a company that spawns other companies
2. Be descriptive - like "RemindsYou" or "SalesForce" or "FriendFeed"
3. Kind of like JasonLBaptiste said - drop some or all of the vowels can work sometimes as well... so like "RCRDLBL" or "MSTRKRFT" or "PIPL" or "TWTPoll"
With the 2 names I've had to choose one was brainstorm -> short list -> check trademarks -> check domains. This lead eventually to "Barefoot Ceramics" (a paint your own pottery studio, http://www.barefoot-ceramics.co.uk). One of the top names "Ty Pots" (pun on teapots and "house of pots" in Welsh) I really liked but we discounted in because it's just too clever and can't be dictated easily. Barefoot is sometimes misspelt Bearfoot, but rarely - it's a term in pottery as well as being a reflection on the relaxing environment and earth friendliness we've tried to build.
The other was choosing a name that was a double-entendre from my location and the nature of the activity, I chose early and then scouted around for domain variations using different TLDs. This lead to BridgeHosting.net (web design, SEO, internet hosting, http://BridgeHosting.net). The pun is "connecting people" and that my location has lots of bridges.
My personal webspace is at http://www.alicious.com - I chose that because it can be an ending for various subdomains, eg "web.alicious"; noone can spell it or pronounce it!
Here are some words I just came up with using a random word generator script I once wrote (in Python): zenthanooo, uroratou, arerent, malimanged, wastsal, matiri, rasanto ... you get the point.
They might not be the best names for your startup, but they might provide a starting point for brainstorming.
http://www.igorinternational.com/process/naming-guide-produc...
(A bit surprised it hasn't been linked to in this thread already.)
Then, the name just hits me.
Why? Because it's hard enough to build consumer awareness of one name, especially with a limited marketing/PR budget. Doing it simultaneously for more than one name is even harder.