How to read domain auction results

Closing bid against confirmed sale

Every number attached to a domain is making a claim, and the claims are not equally strong. Before comparing two figures, work out which rung each of them is on.

Each rung is useful. They are just not interchangeable, and the mistake that costs money is treating a lower rung as though it were a higher one: quoting an asking price as market value, or handing a seller a model estimate as though it settled the argument. Ours does not, and a seller who has been in this a while knows it.

Bids, bidders, and false competition

The Activity column is the one field whose unit changes between rows, and it is the most misread number on the page.

A bid count is how many bids were placed. A bidder count is how many accounts took part. One determined bidder can place thirty bids on their own, so a bidder count is almost always the lower figure, and the two are never comparable as a single number. A row reading 12 bidders describes a far more contested auction than a row reading 12 bids.

Which one you get depends on the source, because marketplaces publish one measure or the other and a few publish both. NameJet, for one, publishes no bid count at all. We label every row with the measure it actually carries rather than quietly standing one in for the other.

The other thing to know is that not all activity is open competition. On drop-catch platforms the auction is often between people who backordered the same name, which is a closed pool formed before the auction started rather than a market discovering a price. It is real competition. It is just a different shape of it, and a small bidder count there can mean more than a large one somewhere else.

Why the marketplace matters

Marketplaces are not interchangeable venues that happen to sell the same thing. They attract different inventory and different buyers, and some of them are not really marketplaces at all so much as shop windows onto other people's.

That last part matters more than it sounds. When an aggregator lists another registrar's auction, the same auction reaches us twice at an identical price, bid count and close time. Counting both would report one auction as two closes, inflate every figure on the page and make a quiet week look busy. We treat a domain and a close time as one auction, and where several sources carry it we keep the operator's record over the aggregator's view of it.

Generated from the same source metadata the auction page uses, so the two cannot drift apart.

The practical consequence: a result from one venue is not automatically the same evidence as a result from another, and a shift in which sources contributed rows is enough to move a market figure on its own. Reading a trend covers that failure in detail.

Check the archive before you use the price

This is the section that changes the most decisions, and it is the one most people skip.

A domain and the website that used to run on it are two different assets that happen to share a name. When bidding runs up on an ordinary-looking string, it is worth asking what the other bidders could see that you cannot. Often the answer is a decade of a real publication, a backlink profile, or traffic that still arrives every day out of habit.

The workflow itself takes about two minutes per name:

1. Open the capture calendar rather than a single snapshot. The shape of the history is the useful part.
2. Scan which years hold captures and where the gaps are.
3. Open three: the earliest useful capture, one from the middle, and the most recent.
4. Ask whether the topic and the branding stayed the same across all three.
5. Look for abrupt changes of subject, language, or design. Those usually mean the domain changed hands.
6. Decide what it actually was: a business, a publication, a parked page, a redirect, an affiliate site, or an empty shell.
7. If the history looks unexpectedly thin, try the www and bare forms, and both http and https, before concluding anything.
8. Write down a one-line verdict. If you cannot write one, you have not finished looking.

Two limits worth stating plainly. Archived pages are not yours to republish: the capture is a record of what was there, not a transfer of the copyright in it. And an empty calendar means the archive could not answer the question, not that the answer was clean. Sites go uncrawled, get blocked, sit behind passwords, and are removed later. No captures means unknown.

The full history check goes past the archive into backlinks, ownership and trademarks, which is where a material bid should end up.

Turning a result into a comparable

A comparable is not "another domain that sold". It is a name close enough to yours that its price carries information about yours, and most results fail that test.

Work down the list. Anything that does not match is a reason to adjust or to reject:

• Extension. A .com result does not price a .io.
• Structure. Word count, length, and how it reads out loud.
• Category and buyer pool. Who would want it, and how many of them there are.
• Channel. An investor auction is a different market from a brokered sale to an end buyer.
• Period. Prefer recent, but an older exact match is better than a recent loose one, adjusted for the market since.
• History. Did that name carry a website, and does yours?
• Evidence quality. Which rung of the ladder above the number came from.

Then write the verdict down, in this shape:

One strong comparable beats five weak ones. And if the only comparables you can find are weak, that is itself a finding: thin evidence means a wide range, and a wide range means anchoring low.

Three results, read three ways

Three closes, all plausible, all requiring a different conclusion. The domains and figures are invented; the readings are the ones that matter.

Example

Invented names and figures, chosen to show three readings of the same kind of row.

Setting a ceiling before you bid

Decide the number before the auction, and write it down. Once bidding is running, every counter feels like progress, and progress is what makes people overpay.

There is no formula that produces the answer, but there is an order to the subtraction:

1. Start with what you expect to sell it for, through the channel you will actually use.
2. Subtract renewals and holding cost for as long as you expect to hold it.
3. Subtract marketplace or brokerage fees.
4. Subtract a discount for uncertainty, and widen it when the comparables were weak.
5. Subtract an allowance for anything the history is going to cost you to clean up.
6. Subtract the return that makes the trade worth doing at all.

What is left is your ceiling. If it comes out below the opening bid, that is a useful afternoon rather than a wasted one.

One exception, and it is the one people get wrong in both directions. If you are the end buyer, the name matches a business you already run, and the alternative is a worse name for your own company, then the ceiling is what the name is worth to you, which is usually higher. Just be honest with yourself about which of the two you are before the bidding starts, because both types tell themselves they are the other one.

The sixty-second check

Before you use a result as evidence for anything:

• Is this a final recorded bid, or a confirmed sale?
• Is the activity figure bids, or bidders?
• Does the source run this auction, or is it showing someone else's?
• Have I opened the archive?
• Am I pricing a name, a former website, or both?
• Does the history carry legal, search, or reputation risk?
• Is the result structurally and contextually comparable to my name?
• Have I set a ceiling, in writing, before bidding?

Open the auction record and read a row against this list. It takes a minute, and it is the difference between a number you have seen and a number you can use.