A few years ago I finished building a website for a client, and days before launch he sent me a list of six domain names.
He had come to me with none at all. We had talked it through, I had suggested a few directions, we picked one, bought it, and built the whole site on it. As far as I was concerned the naming conversation was closed.
It was not. He was still not at peace with it. So one weekend he went and registered six more names himself, in his own account. The same words in a different order. A plural where ours was singular. One of them had a letter quietly doubled in the middle, the kind of thing you only catch when you read it out loud, slowly, twice.
Then he asked me to point all six at the new website.
I want to be fair to him, because I have watched people do this again and again since, and it always comes from the same place. He was not being careless. He was hedging. Somewhere in his head, owning the neighbours of a name felt like owning a little more of the name itself.
It does not work like that. But before I explain why, you should know who is telling you.
Where I am standing
I sell domain names. This site carries a portfolio of premium domains that I own, and a domain concierge service where I go out and buy a name on somebody else's behalf. I also run a separate business that sells ordinary registrations at ordinary prices. Three different ways of earning money from this one subject.
So when I tell you a name is worth paying for, you have every reason to take it with salt. I also own names I have not managed to sell. Treat me as somebody reporting from inside the trade, not somebody who has it worked out.
I should add one more thing here, and then leave it alone until the end. I have never bought a premium domain for myself.
Which is exactly why I want to start with the thing my own industry keeps getting wrong.
The claim that is not true
Search for whether premium domains are worth buying and read the first ten results. They will be written by registrars, marketplaces and brokers, which is to say by people who are paid when you say yes. They all arrive at the same answer, and most of them include some version of this line:
A premium domain improves your search rankings.
It does not. Not on its own, and not in the way that sentence makes you picture it.
Google has been saying this plainly for years. Keywords sitting in the domain name are not the ranking lever people think they are. There is no boost waiting inside the address bar. A name stuffed with search terms can still cost you, just not in rankings: it reads like spam to people, and it locks your brand to whatever you happened to be selling the day you registered it. What decides rankings is the same boring list it has always been. Content that actually answers the thing someone searched for. Other sites finding you worth linking to. Technical basics not being broken. Time. And people not bouncing straight back to the results page.
Buy the best domain in your industry, put a thin site on it, and you will rank nowhere. I have watched that happen.
So what does a domain actually do?
Two things, and both are real.
For a brand new site, a name that matches its subject helps a search engine work out what you are about faster. That is a head start on being understood, not a head start on ranking. It shortens the confused early phase, nothing more.
The second one matters more, and almost nobody writes about it, because it is not a feature you can sell in a checkout flow.
The real reason that client was stuck
Go back to the man with six extra names.
He had a real worry underneath the hedging.
There are two words that describe his entire industry. Not his company, the industry itself, the thing customers actually type when they want this service. A competitor was sitting at the top of the search results for exactly that phrase, and that competitor's domain was those two words with a third word attached at the end.
The clean version, just the two words and nothing bolted on to them, belonged to somebody else entirely. It was parked, it was for sale, and the price was not a weekend decision.
Look at what that combination does to a person. He saw a rival ranking first. He saw the rival's domain. He joined the two together and went shopping for names.
The rival was not ahead because of his domain. He was ahead because he had been in that market longer, was better known in it, and had been feeding a website for years while my client was still choosing one. The matching domain was a coincidence sitting on top of real authority, and it read like a cause.
That is the myth from a moment ago, playing out in a real person's head. An easy mistake, because the evidence is right there on the screen.
So we did the sensible interim thing. We could not have the clean name, so we took exactly the same two words on a different ending and launched him on that.
That is worth knowing on its own. An alternative extension works well as a bridge. It gets you trading under the correct words while you decide about the expensive one. What it will not do is settle the argument inside your own head.
Because look at what he actually owned by then. Six near misses, one working website on the right words but the wrong ending, and a competitor still ahead of him. All six pointed at the one live site, so all six worked. That was the problem. Because they worked, he treated them as interchangeable and started handing them out that way. One name to this customer, a different one to the next. He was not being careless. He genuinely believed each of them would show up in search on its own, and no amount of explaining moved that.
None of that is a ranking penalty. It is worse than a penalty. It is dilution. Every mention, every card, every customer who remembered one of the six instead of the one, all of it landing somewhere just off centre.
A year and a half later he bought the clean name. The price hurt. He paid it anyway.
What he ended up owning is the kind of name people mean when they say premium. Not his company's name. The name of the category itself, the two words and nothing else. There is exactly one of those in any market. Rivals can add a word to it, take a different ending, register the plural, and every one of those is a version of the same admission: the original was not available.
Today he is the one at the top for that phrase. The competitor who was ahead of him is still trading under two words plus a third, and the newer ones are arriving on other endings, doing what my client used to do.
The honest version of that story is not that the domain ranked him. The domain did not rank him. The domain let him stop splitting himself seven ways, so that all the ordinary work of running a business finally collected in one place instead of leaking out of six.
That is what you are actually buying. Not an algorithm trick. Consolidation.
So where does the price come from?
There are two completely different products hiding behind the word domain, sold at two completely different prices.
One is a registration. You are renting an unused string of characters from the registry, a year at a time, for a small annual fee.
The other is an aftermarket sale. Somebody already owns the name. You are not paying a registry, you are paying a person, and that person sets the number.
What pushes a name into the second category is fairly predictable. Short beats long. Real words beat invented ones. Words that spell themselves when you say them aloud beat anything you have to repeat twice. Dot com still beats everything else, not because it is technically better but because it is what people type when they are not paying attention. Names that already receive visitors typing them directly into the address bar carry that traffic with them. And sellers look at what similar names have sold for, the same way property is priced.
All of that sets a rough band. But the actual number comes from somewhere else entirely, and this is the part worth understanding.
A domain is worth what it is worth to one specific buyer.
There is no market of thousands of people who need the exact name of your business. There is a market of one. You. That is why the same name can be cheap for a decade and then expensive the week you decide you want it, without anything about the name changing.
I have watched this from the selling side too. A name sat in my own portfolio for years, two plain words that together described one narrow technical job. In all that time it was worth nothing, because nobody who needed it knew it existed. Then I found the handful of companies that do exactly that job and wrote to one of them. A few emails later it was his, at a price that had not existed the week before.
The name never changed. The market of one arrived.
It cuts the other way as well, and honest sellers will tell you this. A name that means everything to you may mean nothing at all to the next person in the queue. Which brings me to the most misunderstood purchase I have seen.
The domain you buy and never use
A client of ours runs an online store that we built and still look after. It sells a physical product, it has been ranking for years, and there is real money going into ads pointing at it.
The store sits on a longer domain. The short, obvious, brand name version of it belonged to somebody else.
He went and bought that short one. Not through me, I should say. He did not even know I work with domains, and I only found out afterwards. Then he parked it, and for a long stretch it did nothing at all.
From outside that looks like money burned. It was not. There are two reasons he could never put the store on it, and the second one is the one almost everybody misses.
The first is the boring one. That store had years of search history behind it, links pointing at particular pages, and an advertising account tuned around it. Moving a business like that is not a rename. It is a controlled demolition you hope you rebuild correctly. Redirects mapped page by page, rankings wobbling for months even when the work is perfect, every old link now travelling through an extra hop. Sometimes it is worth doing. Very often it is not.
The second is about what a name is actually for.
Take the word office. On its own it is one of the most valuable words there is, and you cannot sell a single chair on it. Nobody arriving at a name that broad is looking for furniture. They could be looking for anything.
Now put two words together. Office chairs. Uglier, cheaper, and completely unmistakable. Build the shop on that one and it will outsell the expensive word every single day, because everybody who types it already knows what they came for.
That is exactly what happened here. The store sells one kind of part, and the domain it trades on says so, with the product word sitting right in the middle of it. What he bought was a person's name. Not a brand somebody invented for a company, a real person, and in that world a name people know. Putting the shop on it would have swapped a name that says what is for sale for a name that says who is behind it.
So he bought it for a completely different job. Today it is not a store at all. It is the brand's home. The history, the story, the work, all living on the name people say out loud, quietly pointing visitors towards the shop that does the selling.
That is a premium domain doing the thing nobody advertises. Not traffic. Identity.
Two things fall out of this, and they are worth more than anything else here.
Buying a better domain and moving your site onto it are separate decisions, and the second is not automatic.
And a name can be too big for the job. Breadth is not usefulness. The right domain for a shop usually says what the shop sells.
The detour costs more than the domain
You have watched this happen for years without noticing it.
Twitter spent its early life on twttr.com, and Evan Williams has said twitter.com cost them $7,500. Facebook was thefacebook.com for its first year and reportedly paid $200,000 to lose the word the. Slack ran on slackhq.com, and the real name is reported to have cost $60,000. Dropbox ran on getdropbox.com, and Drew Houston has said the real name cost $300,000 in cash, after turning down an offer of stock instead that would be worth an absurd amount today.
Four companies with better funding, better lawyers and better advice than most of us will ever have. Every one of them took the long way round. Launch on the version nobody wanted, grow, then go back and buy the one you meant all along.
If it works like that at the top, it is not a small business problem. It is just what wanting a name looks like.
Another founder, another shape of the same story.
His company ran on an abbreviation. Over the years he had collected the longer versions of it, the ones with a service word bolted on the end, the ones with the plural. All fine. None of them the clean one, which sat publicly listed for sale, for anyone to see. He looked at that listing for months.
When he finally asked me to approach the seller, I did, and I came back with a workable number. Then it stopped. His team told him it was a waste of money. Nobody types domains any more, they said. People find us on Google. We already own three, they said. He was the founder. He could have simply decided. He did not.
Six months later he called me and said, in about this many words, buy it, whatever it costs now.
Nothing had changed except him. No competitor had grabbed it. There was no deadline. He had spent half a year quietly not being at peace with his own company's name, and one day the discomfort outgrew the invoice. We bought it, everything folded onto the clean name, and a few months later he came back for a second name that happened to be sitting there at ordinary registration price.
Two things stayed with me.
The first is that the price of waiting is rarely the price of the domain going up. Sometimes it does, sometimes it does not. The actual cost was six more months of business cards, email signatures, proposals and ad spend pushed through a name he did not believe in.
The second is uncomfortable, so I will say it gently. A domain is a founder's decision, not a committee's. I do not mean founders are always right. I mean the people advising against it are usually correct about the arithmetic and blind to the thing being bought. It is nobody in that meeting's name on the wall. The person lying awake about it is the person who should decide.
The backwards way that actually works
One more, and this one is my favourite because it inverts the whole problem.
A friend of a friend spent roughly a year and a half unable to start his company. Not because of funding or product. He could not settle on a name. Every good one was taken, every available one felt like a compromise, and the whole thing stalled at that step.
Eventually he gave up on imagining names and started shopping for them instead. He went through marketplaces and looked at what was actually for sale, at prices a new company could survive. It took a week. He found two ordinary words sitting next to each other that described exactly what he intended to build, available, priced like a piece of office equipment rather than a house.
Then he named the company after the domain.
That order feels wrong to people. You are supposed to have a vision, name it, and then acquire the matching name. In practice that route ends in either a compromise domain or a bill you cannot pay.
Choosing from what you can actually own is not settling. It is the only version where the name, the company, the logo, the email address and the URL all agree with each other on day one. He never had to explain a mismatch to anybody. He never registered a single near miss.
The ones who skip all of this
Not everybody takes the long road, and an article full of detours can start to sound like detours are compulsory.
One client arrived at my desk having already bought his domain. A short single word, the kind you can say once on a phone call and nobody asks you to spell it. It fitted his business exactly. He had paid a premium price for it out of his own pocket, before there was a website, a logo, or a launch date, and he had done it without asking anybody's opinion including mine.
We built his store on that name and still look after it today.
No near misses. No bridge extension. No year and a half of quiet dissatisfaction. Nothing to unpick later.
He is the cheapest version of everything written in this article, and the only thing separating him from the others is that he believed the name mattered at the beginning rather than at the end.
What actually happens when you go and buy one
Since I have spent this long on why, here is what it actually looks like.
Taken turns out to mean three different things, and they behave nothing alike. A name parked with a for sale sign on it is a conversation waiting to happen. A name with a working business behind it is a closed door, worth one polite question and then leaving alone. And a name registered years ago by somebody who has long since forgotten about it looks like the easy one, and is the hardest of the three, because there is usually nobody left at the other end to answer.
Owners themselves are mostly invisible now. Their details sit behind privacy services, and whatever contact you find just forwards your message into an inbox nobody reads. So enquiries travel through a form, or a marketplace, or a person acting on somebody's behalf.
That last one exists mostly because of what an ordinary email address gives away. Picture the message arriving at the other end. It comes from an address ending in a company name, and it asks about a domain that is that same company name with one word missing. Nobody has to investigate anything. The seller knows within seconds that this is not idle curiosity, and that walking away is going to be difficult for the person asking.
The name did not become more valuable. You did. Sellers are not really pricing domains at all. They are pricing how badly somebody wants one, and the enquiry itself did the telling.
The rest is quieter than people expect. Whoever says a number first has handed the frame to the other side. Money never moves straight between two strangers, it goes through an escrow service that holds the payment, waits until the name has landed in the buyer's account, and only then pays the seller. Replies take days. Negotiation takes days or weeks. The transfer can sit still for reasons nobody ever explains properly. A month is a fair guess for the whole thing to feel finished, and building a launch date around it is asking for trouble.
The one part I would settle before starting anything is the number at which you stop. Written down, while nothing is happening and nobody is waiting on a reply. Almost every good outcome I have watched in this came from somebody who could genuinely have gone home without the name.
When not to buy one
I would rather say this than have you find out later.
Do not buy a premium name to fix a business that is not working. It will not. It buys clarity, not customers.
Do not buy one because it contains a search term you want to rank for. That is the myth this article opened with.
Do not buy one that carries somebody else's trademark. You will lose it, and possibly worse.
Check what the name was used for before you. A domain with a history of spam or something worse behind it arrives carrying that history, and cleaning it up costs more than the name saved you.
Do not buy one if the money is genuinely tight and the current name is merely imperfect rather than actively confusing. Imperfect is survivable for years.
And if you already rank, already advertise, and are simply irritated by your longer domain, look very hard at that decision before you move anything. Buying it can still be right. Migrating onto it is a separate question with a much higher bar.
My own name took two years
I have spent this whole article describing other people's detours. Mine was longer than all of them.
I built my first WordPress site for a client in 2010. By 2012 I had decided the work needed a proper name of its own and started looking for one. I did not buy the domain I use today until the middle of 2014.
Two years. Not saving up, not waiting for anything. Just looking for a name.
Some of that was fear. Before all this I had launched a service on a dot net carrying a word that belonged to somebody else's trademark. Nothing ever came of it, and it easily could have. That scare is most of the reason I then went so far in the opposite direction. The rest was me. My friends learned long ago not to ask me where we should eat.
I already had a name I loved. It was my niece's name, and it carried a meaning no generator was ever going to produce. It also did not sit right on a technology company, and I knew that even while I was refusing to let go of it.
So I paid a branding agency, cheaply, to solve the problem for me. They came back with more than forty options and I picked none of them. Then came the generators, every business and domain name tool Google would give me, months of them, nothing surviving until the next morning.
The name I finally used had been sitting in that agency's list the whole time. They had taken the family name I loved, cut a piece off it, and welded it to the word technology. I walked straight past it the first time. A year later it was the only one still standing.
Its dot com was gone, of course. Registered in 2005 by a company on the other side of the world, still theirs today, still in use. I wrote to them. They never replied, which is the ordinary outcome this article already warned you about.
And here is the part I want you to notice. I did not know the aftermarket existed. Everything written above, the whole apparatus of brokers and escrow and walk away numbers, was invisible to me. If somebody had explained it to me back then, I would probably have gone into debt chasing a name I did not need.
So I did the only thing left. I ran the name through the generators once more and let it come out slightly changed. Two letters moved. That was it.
What came out was better than the thing I had been chasing. It cannot be mispronounced, which the original could. And over the years it quietly collected meanings I did not plan. A shift in technology. A shift that still holds its balance. Even the way it sounds when somebody says it too fast.
I did not buy a name that meant those things. I grew into them.
I registered it in June 2014 and have used it ever since, though for twelve years there was no company website on it at all. I was too busy doing the work to build the shopfront. There is a small joke buried in that. Look up the company behind all of this and the only date you will find on paper is August 2017, when the private limited company was incorporated. Not 2010, when the work started. Not 2014, when the name arrived. Records are tidy like that. Everything before the paperwork lives in old invoices and in my own memory, which is true of most businesses and worth remembering the next time you judge one by its founding date.
I know what my own name is worth by the standards of this article, and it is not much. It is invented. Plenty of people who meet me say it wrong, and I stopped correcting them years ago. Say it once on the radio and a fair number of listeners would type something else, a flaw well known enough in this trade to have its own name, the radio test. Anybody who values domains for a living could look at my name, then at what I do for a living, and have a good afternoon at my expense.
They would be right about the name and wrong about the point. It was never a purchase. It was a decision about what I wanted to call the thing I was going to spend my working life on.
So, the confession. I sell premium domains. I go and buy them for other people. I have never bought one for myself, and I do not want one.
Not because they do not work. This whole article is a record of them working. It is because somewhere in those two years the hunger went out of me. I stopped needing a name to be perfect and started needing it to be mine, and those turned out to be different things.
If you are in the middle of your own two years right now, that is the only part of this I would actually press on you.
You are not behind
There is a version of this article that would make you feel late, and I do not want to write it.
Plenty of businesses I admire did not own the name they wanted on the day they opened. They started on whatever was available, sometimes on something genuinely awkward, and got on with the actual work of being good at the thing. The name came later, once it was obvious the business was going to exist long enough to deserve one.
That order is completely fine. A good domain removes friction. It does not create demand. Nobody has ever bought anything because the website address was tidy.
So if the number in front of you is a real problem right now, leave it. Note the name down, keep an eye on it, build the business, and come back when the money is boring instead of frightening. Most of the people in this article did exactly that, and the only thing they would change is how long they spent pretending they had stopped thinking about it.
What I have actually learned
Almost everybody ends up on the right name eventually.
Some get there in a week, by shopping for what they can own. Most get there the long way, through the near misses, the extra extensions, the doubled letter, the six months of being talked out of it. Then they buy the name they wanted at the start, usually for more than it would have cost them, and always after paying for the detour in ways that never show up on any invoice.
The domain is rarely the expensive part. The years of not having it usually are.