If you have ever said your own domain name out loud on a call and then had to spell it, you know the moment. Borets Stamenov, co-founder of SeekFast, described it precisely:
We picked a clever, short domain that no one could spell right. It looked great on paper, but every time we said it out loud, we had to follow up with 'That's with a K, not a C' or 'Two Ls, one Z.' We lost traffic, emails bounced and worse -- people Googled the wrong thing and found competitors.
Startup knowledge is supposed to compound. That is the whole premise of the postmortem, the founder essay, the accelerator library. Someone makes a mistake, writes it down, and the next generation skips it. On hiring, the notes circulate. On pricing, on fundraising, on churn... the notes circulate.
On naming, they don't. Every founder in the last forty years appears to have learned this lesson individually, from scratch, at full price. I know because they said so, on record. My team collected over 140 of these statements from founders, CEOs and operators. Read together, they stop being anecdotes. They become an archive of the same tuition, paid separately, on very different invoices.
The cheap version of the lesson
The lesson starts small. Matt Holmes, founder of Handshaking:
I'm glad I was able to get Handshaking.com with the 'g', as many people had been emailing me, matt@handshaking.com, and before I was not receiving those emails when we only owned Handshakin.com.
Emails to an address you don't own are not a hypothetical risk.
Stella Morrison, founder of The Stellastra Effect, on what customers conclude when the obvious address doesn't resolve:
A company called All Things Fans might use their full name across all their branding, but choose the URL ATFFans.com because it's available and cheaper. Naturally, people are going to try AllThingsFans.com first, and when there's nothing there, customers will assume All Things Fans was out of business.
Not "customers will be mildly inconvenienced." Customers will assume you are dead. That is the reading a mismatched namespace produces, and the worst part? No one inside the company sees it happen.
The expensive version
James Currier, now a general partner at NFX, got his invoice in 1999:
I woke up to the power of company names when we screwed up my startup's name in 1999. We called it Emode which in hindsight was hard to spell, impossible to remember, and a dated name within 24 months. With coaching, we changed it to Tickle. Traffic shot up 30%, journalists started writing about us, our ad spend became 20% more effective, and no one forgot our name.
Ben Carmitchel was paying his in monthly instalments before acquiring DataRecovery.com:
We were spending around $10,000 per month on Google AdWords and other pay-per-click services, which wasn't really a great investment. As soon as we acquired the domain name, we saw a significant increase in focused traffic, apparently from direct visitors.
Darpan Munjal, after moving to Atom.com:
The difference was immediate. What surprised me most was how fast the impact showed up. Organic traffic doubled, up 129%. Our cost-per-click even fell by 18%. The brand was doing work that ads alone couldn't.
And Alex Vetter on what Cars.com who played it smart:
We get a majority of our traffic directly and organically, which allows us to really invest in other things, which to us is product innovation as opposed to what a lot of our peers have to do is, is they have to keep spending marketing every dollar every day just to stay relevant.
Matt Barrie of Escrow.com and Freelancer.com compressed all of the above into one line:
Consider a top domain as a permanent discount on your marketing budget.
The pattern across these four companies is not that a domain is a nice to have. It is that a weak one is a recurring tax, billed monthly, itemised as "marketing."
The version investors read first
Charlie Bullock raised for Scan.com when the company had almost no cash:
What I knew was that this name would allow us to be taken a lot more seriously by future investors because it is a lot easier to raise venture capital when you've got a domain name like we have because the first thing you see on the deck is the title, and if it says Scan.com it is making them turn to the next page.
Waseem Daher made the same call at Pilot, in less glamorous circumstances:
At that time, we were six people in an office above a fried chicken restaurant. $400K was a lot of cash. We only went for it because we had raised a $3M seed round, and I believed that the long-term branding benefit outweighed the short-term cost. I definitely don't regret it.
Paul Graham, who has watched more pitch decks than anyone quoted here, on what the gap signals:
The problem with not having the .com of your name is that it signals weakness. Unless you're so big that your reputation precedes you, a marginal domain suggests you're a marginal company.
Investors read the namespace before they read the deck. Founders tend to discover this after the meeting.
The version that arrives at exit
In 2013, Paul Graham emailed the founders of a startup called Yhat:
Your product is great, but your URL is yhathq.com. You need yhat.com. It's $19k right now -- a big expense, I know. But if you succeed, the price will skyrocket. And if you fail, the cost won't matter anyway.
Austin Ogilvie, Yhat's founder, on what happened next:
We listened and bought Yhat.com for $19,000. When Yhat was acquired by Alteryx in 2017, that $19k purchase added over $300,000 to the cash consideration in the deal.
Nathan Barry, after ConvertKit became Kit:
Revenue growth is up substantially and we've landed some massive deals. Just in the 6 months after rebranding we signed Tom Brady, Matthew McConaughey, Dua Lipa, Morgan Freeman, Lil Jon, Deepak Chopra, Ellen, Chef Nick, Stephen Bartlett, and 10,000+ creators. And the whole rebrand cost less than 5% of our annual revenue.
Same lesson. The only variable is when the invoice arrives, and the invoice grows the longer it waits.
The sceptic
If you are reading this thinking the sample is self-selected... people who spent money defending the spend... Greg Isenberg started where you are:
I used to think buying a premium dotcom domains were a waste of cash. I'd hear stories of people spending $500k on a dotcom domain and would silently laugh to myself. If only you spent that money on paid ads, viral videos etc. That 500k might be 100k customers! You're just lighting the cash on fire for ego. But I was wrong. I recently started actually acquiring some dope domains. So, I ran an experiment. What if I took the same product and slapped on a new premium dotcom domain. A dot co for a dot com. And I started seeing the conversion rates jump 2-3x Same EXACT product, just a new domain, more customers.
Same product. Different domain name. Different conversion rate. It is the closest thing to a controlled experiment in the data we collected, and it was run by a man who expected the opposite result.
Why the lesson never compounds
None of these founders was careless. Aaron Patzer, who spent three months and $182,000 on Mint.com, put his finger on how the mistake actually gets made:
Most people choose their name because the domain is available. That's a really bad idea.
The name gets chosen at the moment the company knows the least, by whoever is around, under the logic of what's free. There is no owner for the decision, no benchmark to check it against, and no industry that has bothered to explain the stakes in language founders use. So the knowledge sits in scattered interviews and podcast transcripts instead of in the diligence checklist, while each generation of founders pays again.
Paul Graham, one more time, with the numbers:
100% of the top 20 YC companies by valuation have the .com of their name. 94% of the top 50 do. But only 66% of companies in the current batch have the .com of their name. Which suggests there are lessons ahead for most of the rest, one way or another.
Lessons ahead, one way or another. The data says the lesson itself is fixed. What varies is the price, and the price is set by how long you wait to learn it.
Which version of it is your company paying for right now?