Hacker News new | past | comments | ask | show | jobs | submit login
How to name your startup (joel.is)
55 points by joshuacc on Aug 11, 2012 | hide | past | favorite | 54 comments



A recent article made the case against disregarding domain names: while Square used to be SQUAREUP.COM, and Facebook used to be GETDROPBOX.COM, neither is today; both have acquired the bare domain name corresponding to their name and use them exclusively.

The issue here is then, once you make the decision to be "getFOO.COM", the better you do, the more expensive FOO.COM becomes. There is no guarantee the price for the "real" domain name will ever be reasonable. It may instead become intractably expensive.

If you're committed to "getFOO.COM" forever, the strategy works just fine. But when I see people talk about how Facebook used to be "thefacebook.com", I get itchy. How much would you have charged for "facebook.com" 4 years ago?


When you're a kid in a dorm room (even a Harvard dorm room), the cost of "facebook.com", low as it might have been at the time, is already intractable. When you're a funded Silicon Valley startup, the cost of "facebook.com" is higher in absolute terms, but relative to your buying power is more affordable.

Having such a successful startup that domain squatters will charge more for your domain is a good problem to have. The most any domain squatter will earn for the domain is the most they can get from you anyway--they can't sell it to a competitor since it'll be subject to immediate trademark dispute, and they can't sell it to anyone else since no one else will pay as much as you would. I don't think it would ever actually become intractably expensive--the business model doesn't even make sense that way.


Domain "squatters" buy domain names and keep them to sell.

As long as they aren't infringing on a trademark it is a legitimate and legal business.

While I don't really even like the word speculation that would more accurately describe what is going on. People might not like it or get frustrated that they can't buy the domain at the price they would like. That's simply entitlement. And they also assume that if the "squatter" didn't buy the domain it would be just sitting waiting for them to purchase for $10 for their startup. In a free market system that would simply not be the case.

Someone thought of an idea and purchased domains at the going rate (or more later on if not the original owner) and now the domain has value. If it was so apparent the domain had value to everyone then everyone would have done the same thing and they didn't. This is really no different than any type of gamble or investing someone does in business or in real estate.

You can come up with any scenario to try limit domains to "actual use" and it will fail on it's face. There is no way to have the growth of the Internet that we have and police how someone uses their domain. This is not gas or war time rationing. There are many options if the domain that you think is perfect for your company is taken by someone else. Find a different domain. Get over it.


You seem to be arguing against a comment I didn't even write.


I'm taking issue with the use of the word "squatters" in the way you used it.


It's the commonly used term for the business model. Sorry if you find it offensive, but you've done a masterful job of missing the damned point.


Yeah I don't understand what you're arguing against either. The fact that you think domain squatting is a great, legitimate business is sort of irrelevant. Maybe you don't like the term "squatting" because you keep putting it in quotes, well even it is the best business in the whole wide world, it's still squatting. You don't like the word "speculation." You seem to not like to call things what they are ... Anyway, the parent did not make a case for why they're bad guys. Get over it. :)


"Yeah I don't understand what you're arguing against either."

I'm arguing against the use of the word squatting to describe people who buy domain names with the intent of reselling who are not infringing on obvious and famous trademarks.

Squatting is identified commonly as:

"Squatting consists of occupying an abandoned or unoccupied area of land and/or a building - usually residential - that the squatter does not own, rent or otherwise have lawful permission to use."

http://en.wikipedia.org/wiki/Squatting

Specifically if we replace the real estate with the domain name we can see that the person who has acquired the domain name that they wish to sell does "own, rent or otherwise have lawful permission to use".

Certain words and expressions have negative connotations and "squatting" is certainly one of them. They are used incorrectly many times by writers to get people's emotions up and band them together against something the writer is trying to achieve.


I don't see the logic here at all. If I start an application and it's called "buffer" and I have no money and the domain buffer.com is being sold by some domain squatting pirate for $200,000 then obviously I can't buy it. The choice then is to find a name like squeebo.com or whatever that no-one has yet or go with a less popular TLD like .co (which seem to fill up just as fast as they're released).

However, if I go with bufferapp.com and get it for $8.95 or whatever the going rate for a domain is, then if/when my company blows up and becomes Buffer the unstoppable international juggernaut, how valuable is the domain buffer.com?

No-one's going to start a company called "buffer" any more. So the value is in direct type-in traffic and monetisation through AdSense revenue. No-one else is going to buy it - you've got a market of one buyer and one seller. If the buyer decides to take his ball and go home, you're shit out of luck and have to monetise the domain yourself.

What determines whether or not facebook.com will remain "intractibly expensive no matter how big you get" (ie. facebook.com now has more money than most countries) is whether or not the type-in traffic for the domain facebook.com can be monetised using ads for more than facebook.com is willing to pay for it.

I have no idea about the stats behind type-in traffic but a lot of stuff I've heard indicates that "mainstreamers" would tend to type "facebook" into Google more so than the address bar so I'd be very surprised if the value of the domain didn't peak pretty early versus what the owner could have done with the money had they sold for a million or so in the first few years that facebook started to blow up.

Overall, though, I'd say that the fact there is only one buyer for a domain once a company establishes itself, and the fact that the risk of choosing a shitty name to suit the shitty domains you can afford, or of spending a lot of money on an awesome domain before you've even been able to go to market (holler.com anyone?) are far greater than the risk of having to pay what is comparatively a pittance once your company is making money hand over fist and needs to "cross the chasm" in order to continue to grow.


it is not a one buyer really. a competitor would also be interested in your domain name.


It may be the case that in some weird domain squatting affiliate marketing universe, someone would be able to make money from basecamp.com if 37 signals still ran things on basecamphq.com, but are you honestly saying that if they weren't, then Atlassian would seriously be interested in buying basecamp.com for some nefarious purpose? That they'd have been able to somehow prevent basecamp from becoming the dominant player it has by redirecting basecamp.com to the Jira homepage so that people would try and find basecamp.com then just think "fuck it, well I'm here now, I'll just use this".

I don't buy it ... do you have any actual examples of this having happened?


> Facebook used to be GETDROPBOX.COM

Perhaps you mean "thefacebook.com?"


Not long ago I read on HN that tptacek is always right about everything. I think we can safely say that Facebook used to be GETDROPBOX


That's the really obvious reason that everyone I know is afraid to start a company without getting the .com first.

But the OP makes a good point that it hasn't been as big a barrier as we have feared. Square, for example, is a superb name and I think they made the right decision to proceed with Squareup.com, even though I thought it was foolish at the time. Turns out they were right, and I was wrong.

Square is a damned good name. I wouldn't have had the cojones to proceed with the name, but they did, and good for them.


"There is no guarantee the price for the "real" domain name will ever be reasonable. It may instead become intractably expensive."

Thanks for saving me the time of saying exactly that.

I really can't stress this enough from my experience in this area both on the buying (and assisting sellers) side.

OP: "Pick a great name, then add something to get a domain name. It really doesn’t matter all that much - whether you get the domain later or don’t. Then get building!"

This is scary advice. And when contrasted with the examples given is dangerous. As you say "If you're committed to "getFOO.COM" forever, the strategy works just fine."

But then if you are "bufferapp.com" don't brand as "buffer" in your logo on your website and marketing materials, brand as "bufferapp".


This is the company naming article you're probably referring to: http://thenextweb.com/entrepreneur/2012/04/22/before-naming-...


I would add:

- Distinctive

- Easy to spell

- Easy to pronounce

- Fairly unique on Google

- Abstract is better than meaningful

I love the suggestion that the domain name doesn't matter. A big mental shift for me, but the examples are compelling. Most people find sites with a Google search and not by direct navigation anyway. Easy to spell and pronounce are really important. If people aren't comfortable talking about your company, they wont talk about it. We named my first company Voila! Software, and that was a really dumb name. People were afraid to say or even type it.

Quantcast is the best company name I've used, and that was actually suggested by Godaddy. Imagine that.


> Abstract is better than meaningful

I think this point is lost on a lot of first-time founders.

Finding an abstract name that flows well and is unique is so much better than looking for random synonyms of a word you wanted (or even worse, mis-spellings of said word).

> Most people find sites with a Google search and not by direct navigation anyway.

As a technical person, it always throws me for a loop to see so many people open a web browser and type either domain names or the full URL into a search bar. I think that as browsers like Chrome continue to gain popularity, the line between the "address bar" and the "search engine" will continue to blur, until there's no difference at all.

In the startup world, a name isn't everything -- you can't succeed just because you managed to snag `awesome.com`.

That said, if you think "cinerjie.com" will ever work out, you're sadly mistaken. [1]

[1] It's supposed to sound like "synergy." I'm lame, and I apologize :)


Unique on Google I would see as sufficient, but not necessary. But, as was commented earlier, don't name & spell your project after something extremely common (e.g. Rastafarian).

Given that, Python seems to have done remarkably well given these constraints. Having a superior project speaks volumes, apparently.


Easy to pronounce is pretty important. One startup I was at, I had to spell the name every time I mentioned it to someone new.

"I work for Wayfarer Communications."

"Wafer?"

"No, /Way Fare Er/"

"Wayfarfar?"

"Double-ewe Aye Why Eff Aye Are Eee Are."

"Oh . . . what does it mean?"

Not a /bad/ name, but not optimal, either.


At least "wayfarer" is a word. I worked at "Adapx". Don't feel bad, everyone except the employees pronounced it wrong. The fact that the marketing company that charged them for it wasn't laughed out of the room amuses me. The decision makers had to have asked, "how do you pronounce it?" And they wrote a check anyway.

("Adapts" or maybe "adaps" is the intended pronunciation, I'm still not exactly sure and I worked there.)


"My current startup is named Buffer, but the domain name is bufferapp.com. My previous startup was named OnePage, but the domain name was myonepage.com."

Best of luck to the OP with their strategy and advice. But I wouldn't say "myonepage.com" was a big success by any stretch.

http://siteanalytics.compete.com/myonepage.com/

I'd also like to know specifically why the OP feels they have the long term experience to offer advice in this area? I'm seeing someone who is a few years out of college and doesn't really have the depth of experience in this area to write an article titled "How to name your domain name". Maybe "my experience with how I named my startup" would be more appropriate.

Add: Some of the things he says I agree with somewhat (stick to two syllables) and some I don't (see reply to tptacek below which hmm, by the way I can never remember how to spell lest you think naming doesn't matter) The problem with advice like this is that if you know nothing you won't know what is correct and what isn't.


I don't claim OnePage was a success by any means, in fact in many of my other articles I use OnePage as a reference to countless mistakes and learning which helped me with Buffer. One thing I would say, however, is that the reason OnePage didn't succeed was not because the domain wasn't onepage.com ;)

We have a long way to go with Buffer but I'm very happy with the growth so far: we're a top 4000 website (alexa) and with almost 300,000 users we see 130,000 shares per day via the platform and are fast approaching a $1M annual revenue run rate through our freemium model. We also have some top investors on board in our seed round we closed at the end of last year. We've done all this with the domain bufferapp.com.

I'm not an expert of anything, this is just how I would approach naming a new startup if I were to start something today. I agree your suggested title is perhaps more appropriate. That said, I did draw attention to a few other successful startups, and some of them have been running for a while now.


You are clearly attention whoring by choosing titles that are more likely to generate foot traffic than to accurately reflect the contents. This is disingenuous. You are, as GP said, in no position to offer authoritative advice, so you might try and respect your target audience and puff your cheeks less.


Most of those how-to guides lack meaningful data and/or studies and/or logic.

Just one example: "The domain name doesn’t matter"

How on earth can one come up with such a broad statement without studying the answer for... years?

Oh, he or she "learned it". Now, you can call me a believer!


Exactly. This applies, conservatively, to a good half of HowTo articles hitting HN's front page lately.


The most crucial point is to not obsess over names.

Obsess over product, not name. Companies prosper or fail because of their product, not their name. Consider Amazon. It outshines competitors in e-commerce, even though one rival -- Buy.com -- owns the perfect name for online shopping. Would you care if Google renamed itself to Moogle? Probably not. Google's search engine is the best, independent of name. To paraphrase billionaire investor Vinod Khosla, brands are nothing more than proxies. Meaning, great names cannot hide poor products -- especially in the information age.

There are many examples of ordinary names representing extraordinary businesses. Apple. Four Seasons. Amazon. These names evoke excellence because the underlying services are excellent. Strong brands today will fade tomorrow once quality suffers. Think GM, Dell, and Sears. In short, search for a cool domain, but remember a name is nothing more than a proxy. Build something people value, and value will flow to the name.


~>Amazon outshines competitors in e-commerce, even though Buy.com owns the perfect name for online shopping

I'm not sure i agree with this analysis.. amazon is a great name; i don't think sharing a name with something real is a disadvantage, there's pre-existing recognition (no need to spell it out, doesn't sound weird) but no real chance of confusion. By not being 100% generic (as buy.com is), it has a chance to build a brand. 'Generic'.com ironically makes it unprofessional, as they are handicapped in making a strong brand, imo. About.com is pretty much the only generic name that i could honestly attribute any benefit and success to- definitely not anything commerce related.

But i agree with the sentiment that if you get a name that is not awful, it will not hold you back.


"Consider Amazon."

Using Amazon, Yahoo, Google, ebay etc., backed by years of free traditional media publicity (90's style) to validate a strategy I certainly take issue with that. I'm not saying it's not possible to develop a made up name but using those legacy companies and the advantages that they had at the time is not in itself justification for thinking any startup can do the same.

Similarly news.ycombinator.com is very successful in spite of the fact that the domain is highly unusual. It's an outlier.

And obviously you have to build something of value. But naming does matter and certainly deserves a great deal of attention.

"Four Seasons" - This contradicts your point. Do you really think any old name could be used by a luxury hotel brand? What if it was called "Roomsville Hotel"?


i'm not sure we're in disagreement. names matter and warrant attention. the less direct your name (e.g., amazon), the more you need to invest in marketing and educating consumers about your company/product. the macro point, which i suspect you agree with, is that ultimately the product mattes more than the name, and that is where people should obsess. good names will not lift companies into prosperity, though terrible names can sink companies per your point and my edit.


> Companies prosper or fail because of their product, not their name. Consider Amazon. It outshines competitors in e-commerce, even though one rival -- Buy.com -- owns the perfect name for online shopping.

While true, I would hesitate investing with a group called "Bernie Madoff Groupies" or purchasing a DVD from an Ebay company called "KnockOff".


Great point. "Or fail" should be removed from the line.


Wow--I didn't even consider that. Now I completely agree with the point.

Language is lost on me sometimes.


A recent example of a terrible product name is Fit Sliq Jeans by H&M. Fit sounds very similar to "fitta", the Swedish word for cunt and Sliq sounds like "slicka", i.e. (to) lick. You would think that in a global company such as H&M, which has its roots in Sweden, there would be at least one person who would notice the error.

http://adland.tv/content/unfortunate-name-or-genious-marketi...


Thanks for the edit!


Have to disagree with this article. There are three qualities that override everything else for a good name:

1) Being memorable

2) Easy to spell when you hear it said out loud (and conversely: easy to pronounce when you see it written)

3) Not being cheesy or otherwise embarrassing to say when you introduce yourself to people

I just can't believe it every time I hear a company name that, if I remember it a few days later and want to go look them up, is impossible to spell. Missing a letter is fine. But having silent letters, with k's instead of c's and other non-phonetic spelling? That's just crazy.


Yeah, I think those factors are a bit more important than counting syllables. Shorter is better, but clarity and memorability trumps that.

As a random example based on the tabs I have open, RockPaperShotgun is probably the most popular blog that focuses specifically on PC games. That's a lot of letters and five whole syllables, but it's also easy to remember. It's a good name.


Interesting tidbit about my ebook products, piracy, and naming:

I sell a few ebooks. Let's say they're all about growing plums in your personal garden.

Two of them are brandable, unique-on-Google names. Like "Plum Thumb" or "Plumateer". The other is a generic name. Like "The Plum Guide". plumthumb.com, plumateer.com, and theplumguide.com. All my ebooks are $12.99. I imagine the piracy rate increases substantially once you Photoshopize the price.

My plumthumb and plumateer products are super-easy to pirate. Googling "plum thumb rar" or "plumateer ebook" brings back a whole page of ebook warez links. On the other hand, Googling "plum guide [rar|ebook]" brings back pages of unrelated plum links that aren't my product. The name is just too generic. (Note: theplumguide.com is the first google result for "plum guide").

My "plum guide" ebook serves a broader niche than the other two ebooks thus is more heavily trafficked and has more sales volume. Most of my sales volume comes from PPC->sales-letter funnels. Using free link-back tools, plum guide has a proportional word-of-mouth backlinks to sales ratio.

I don't know how useful this information is. Does forcing someone to use google-fu or fish through Google results to pirate an ebook increases sales? Does a little obscurity-through-genericness provide any benefit when your sales are driven through Adwords anyways? Who knows.

If general behavior is "can't pirate it? then forget it" instead of the wishful-thinking "can't pirate it? fine, here's my credit card", then isn't it better to get pirated? Audience begets more audience and there will always be shrink.

It's fun to think about.

I hope I didn't obscure what I was trying to say with the arbitrary plum examples.


The policy used to arbitrate domain name disputes (UDRP or Uniform Domain Name Dispute Resolution Policy) dictates that building a company on a site like bufferapp.com doesn't give that company the right to the corresponding .com. (buffer.com) Attempting to sue to get the name is called "reverse domain name hijacking" and the arbitration committee is very very skeptical of those types of request.

Case in point: http://www.thedomains.com/2012/06/16/rick-schwartz-wins-the-...

That being said, the price of the .com will track the perceived value of your company. It would be an awful waste to have to use a significant percentage of some round of funding to buy a domain name. Also, it seems like a significant percentage of your type-in traffic would be lost to the buffer.com. Even 10% of your type in traffic is probably worth the $xx,xxx you will pay for a decent .com name.

The arguments against paying for a decent .com seem weak.


Good luck finding a good name that is still available as a .com domain, or that can be bought for less than $2000.

(Actually, I hate the idea of paying any amount to a domain squatter.)


I prefer guidelines for choosing bad names because they are much fewer than guidelines for good names.

* Make the name offensive in some language.

* Make the name difficult to pronounce.

* Make the name difficult to spell.

* Make the name non unique or confusingly similar to another name.

As long as you DON'T FOLLOW these guidelines you'll be ok.


Speaking of names, can anyone identify the current trend of $arbitrary_number + $arbitrary_noun? I saw '80Legs' in a different thread, then there's stuff like 20Cans, 32Tractors, 99RedBalloons, and god knows what else.


The pool of unique names pretty much had ran dry after 37signals. It's the concept that is memorable, and it is quite shallow here.


I haven't heard of those before, however 37signals comes to mind with that naming convention.


Most likely Signals and Folders? And http://www.youtube.com/watch?v=9whehyybLqU for the balloons..


Like my username? orangethirty. In my case, it has a real world personal meaning. At least for me.


I think this forgets the MOST important point of the name. Easy to remember and easy for people to spread it virally. For example, facebook.com or even "thefacebook.com" is not too hard to remember.

I have a systematic way to evaluate this by actually telling over 100-200 people via an audio or video the domain name and then asking them later to remember and write down the name. You will be shocked at how difficult it is for people to remember most names and also at how easily they misspell.

I also totally disagree that the .com name does not matter. It matters a huge amount for referral dynamics, for the brand. The companies listed in the article were mostly forced to buy the .com name at great expense. If it didn't matter they wouldn't have bothered.

In these cases listed in the article the companies succeeded DESPITE the handicap of not having a great .com domain.

It's better to start with a cheap domain to save costs if you don't have the money, but if you can get a memorable .com domain then great.

Whatever you do don't get a .co domain. These are proven by overstock to be total duds (they tried to launch o.co and after a huge and expensive launch they realised that a large percentage of people will remember it as .com and they lost all that type in traffic). A .net .me .biz is much better than a .co


I agree. I ran a company with a name similar to ChicagoSpanishTutor and literally four out of five people remembered it incorrectly. No use fighting, just find a better name or make some change on your part that makes it easier to remember.


You only the mistake of having a bad name once. I had a copywriting side gig I tried to name and got caught up with the idea of a red pen. So we ended up with RedWrit - not only impossible to say and spell (if you know how to say it), but the red pen allusion was completely lost.

This time I don't think we made the same mistake. GrassWire. It comes from grassroots newswire, but you don't need to know that - everyone knows how to read it, when someone says it you know how to spell it, it's unique and the domain was available. That's all that matters.


How do you like Schadeaux Technologies (pronounced shadow) for a business that creates censorship evasion hardware? I always thought it would be cool because "schad" was a German prefix and -eaux was a French suffix, but I then I learned that "schad" means harmful in German and -eaux forms plural diminutive male nouns, usually an animal, so it would really mean "a little harmful," but that can't be right because -eaux is only for nouns. Perhaps it would mean "a little harmful animal." :D


Your logo should be just the shadow of some small, invasive, but endearing or sympathetic animal, to get the oppressed-and-underground-but-not-defeated overtone. I want to say a mouse or rat, or a mole, but none of those are quite right. You might get some kind of insect to work, like a wasp, depending on what kind of vibe you're looking for.


Facebook used thefacebook.com to begin with, because its original name when founded was Thefacebook. This was changed a year later to Facebook when it purchased facebook.com.


5. hire a copywriter


All of these companies paid for a better domain name.

GetDropbox.com

TheFacebook.com

Twittr.com




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: