# What the domain aftermarket learned in thirty years
> Brokers, escrow, quiet sales and public auctions: the DNS aftermarket built its practices by trial and error. Which lessons carry over to onchain TLDs.
Source: https://hush.partners/lessons-from-the-domain-aftermarket
Category: Market & ICANN
Published by Hush Partners.
---
The secondary market for domain names did not arrive with a rulebook. It grew out of a simple observation: registration fees were low, and the names that carried meaning were not evenly distributed. Anyone who held a short, memorable name that described a whole category had something other people wanted. From that gap, a market formed, and the market formed its own practices, most of them learned the hard way.
This article looks at how those practices developed, why they took the shape they did, and what they reveal about the mechanics of any aftermarket built on a namespace. The DNS story is not the only possible story. But it is thirty years of trial and error, and it carries lessons that every TLD holder and every professional advising on onchain TLDs should understand before assuming the field is new.
By the end you will be able to identify which DNS practices transfer cleanly to onchain TLDs, which transfer with modification, and which do not transfer at all. That sorting matters because copying familiar tools without examining the conditions underneath them is exactly where the DNS aftermarket went wrong, more than once.
## How a secondary market forms from scarcity
The domain aftermarket came into being because names are rivalrous: once a name is registered, nobody else can register it. As the commercial web took hold, companies and individuals recognized the marketing and branding potential of a compelling domain name. That demand gave rise to a secondary market in which names registered at standard prices changed hands for much higher amounts.
The mechanism is straightforward. The primary market, registration through a registrar, offered most names at a flat, low fee. It did not price for demand. The secondary market did. The gap between those two prices is what made the aftermarket possible.
An aftermarket purchase is therefore a different act from a registration. The buyer negotiates with the current holder of a name that carries brand value, a history of visitors or the plain meaning of a keyword. Every attribute that made a name valuable was, in principle, negotiable. The price was whatever two parties could agree on, which meant that names that looked alike could sell for very different amounts depending on who was in the room.
This early phase had one dominant characteristic: the price was invisible. Negotiations were private, and outcomes were known only to the parties involved. That opacity was not a defect at first. It reflected the size of the market. Private deals suited a small, fragmented community in which serious buyers were few, high-value names in circulation were fewer, and the reputation earned in one transaction followed a person into the next. When the market outgrew that scale, the same opacity became a structural problem. Newcomers could not tell a fair offer from a poor one, and holders could not tell a serious buyer from a curious one.
## The rise of intermediaries and what they actually did
Intermediaries appeared because the gap between buyer and seller was not only a price gap. It was an information gap, a trust gap and, in many cases, an identity gap. A buyer who needed a particular name had no reliable way to reach the holder. A holder who wanted to sell had no reliable way to find a buyer without signaling a willingness that weakened the holder's own negotiating position.
A broker, in the plain description, is someone who has completed many of these transactions and knows which details delay or derail them. That description is accurate, but it understates the function. The broker was, at the core, a trust intermediary. The broker created a channel that neither side would have built alone, and absorbed the friction of the negotiation so that the principals did not have to expose their positions directly.
The intermediary also solved an information problem. Experienced brokers and investors carried knowledge that sat in no public record: sales that were never reported, valuations that rested on comparables seen behind closed doors, relationships with holders who had no public presence. All of this lived in the heads and address books of practitioners, not in a database.
Professional habits followed. Brokers came to rely on legal counsel they trusted, on contracts whose clauses had been tested by earlier disputes, and on escrow services to safeguard funds in transit. This professionalization happened gradually. The early market had no standard contract and no standard escrow. Both evolved from accumulated bad experience.
The identity function of an intermediary turned out to be as important as the deal-making function. Acting for a buyer without naming that buyer, and sometimes for a seller in the same way, became a recognized service, particularly ahead of a product launch or in any negotiation where the name of a party was itself sensitive information. That function appears again when we reach onchain TLDs.
One more lesson belongs here, because the market took a long time to state it plainly: an intermediary is useful only when both sides know whom the intermediary works for. A broker mandated by the seller owes the seller the best outcome. A broker mandated by the buyer owes the buyer the same. Negotiations in which that question was left vague tended to end in resentment, even when the price was fair. The practice that survived is a written mandate from one side, with the fee paid by that side.
## Why escrow became the non-negotiable standard
Escrow, in which a neutral third party holds funds until agreed conditions are met, is the most durable structural innovation the domain aftermarket produced. It is also the one that took longest to become a real standard.
The underlying problem is not complicated. The buyer has to trust that the seller controls the name, that the seller will move it, and that no one else has a claim on it. The seller has to trust that the buyer can pay and will not reverse that payment after the asset has been handed over. Those two trust gaps point in opposite directions, and they cannot both be closed by good intentions.
Escrow closes both gaps at the same time. The buyer funds the escrow first, which proves the money is real and removes the seller's fear of handing over a name for nothing. Only after the funds are confirmed does the seller move the domain. The escrow agent verifies that the transfer happened, then releases the money. Neither side has to go first on faith, because the neutral party stands between the two acts.
The sequence matters, and the market kept learning again what happens when it breaks. A name is never transferred before the funds are secured in escrow. Verbal promises and screenshots of a payment are not security. Once control of a domain is lost, recovery is difficult. A great many sellers learned that lesson in person before "always use escrow" became a norm instead of an aspiration.
A subtler version of the same error involved a buyer who had funded escrow in good order but then pressed the seller to complete the transfer outside the agreed process to save time, or pressed for the funds to be released before the buyer's own inspection, citing a deadline. The sequence exists to protect both sides. A counterparty who asks you to break it may be careless or may be dishonest. You cannot tell which in the moment, so treat both the same and hold the line.
## Public listings and private sales: different tools for different situations
Many secondary sales of domain names take place in private, are never disclosed, and cannot be verified unless one of the parties chooses to reveal the details. This surprises professionals who come from markets where transaction data is easier to obtain. The visible aftermarket, made of auction results and published sales records, is not the whole market, and nobody can say with confidence what share of it the visible part represents.
Public auctions serve a specific function. When several bidders compete in the open, the price is set by competition and not by the persuasion of one party. An open auction is a sound mechanism for price discovery when the seller has no firm view of what the name is worth, or when the holder needs to show that no individual buyer received preferential treatment. It has limits too: it only discovers the demand that happens to be watching during the auction.
Private sales serve different purposes. They are typically arranged through a broker or by direct approach, often under a non-disclosure agreement, with confidentiality as an explicit part of the transaction. The reasons on both sides are practical. A seller who offers a name publicly commits to a visible process and can lose control of timing. A buyer who approaches publicly reveals an interest that the seller, and the buyer's competitors, can then act on.
In the early market, almost every transaction was a quiet affair, begun through direct outreach, a personal network or a simple inquiry sent to a holder whose identity was often only partly visible. What changed over time was not that private sales disappeared. Practitioners became better at choosing which tool suited which situation.
The split that emerged is roughly this. A public auction works when the name has broad appeal and the seller's goal is a market price reached quickly. A private sale works when either party has a reason to manage the flow of information: a seller who does not want to appear eager, a buyer who does not want to provoke a competing offer, or two parties who want the transaction to pass without legal or reputational noise. The choice of channel is a strategic decision, never a default.
## When organizations became the dominant buyers
The character of the domain aftermarket changed substantially when large organizations became regular buyers instead of occasional participants. The change was not merely one of scale. Organizations brought a different set of constraints and a different vulnerability.
Revealing the identity of the buyer during a negotiation can raise the asking price, attract competitors and expose brand strategy. For an individual buyer, that is a negotiating inconvenience. For an organization, it can be a material financial exposure. Once a seller knows who wants the name, the price can climb and the timeline can stretch, and the interest can reach competitors before a launch, a merger or a trademark filing has been made public.
Organizations also brought internal process. A large purchase might involve a broker on one or both sides, a lawyer reviewing the purchase agreement and, on the buyer's side, a finance function that has to approve the payment. That chain of approval means the deal cannot move at the speed of a private individual. Sellers who did not understand this read the delay as a loss of interest or as bad faith, and negotiations collapsed that had no reason to.
The need to protect identity drove organizations toward specialist intermediaries. That anonymity keeps the price from inflating is obvious in principle. It becomes an operational commitment only when it is built into the engagement from the beginning and not attempted partway through. Once a name has been connected to an organization in any correspondence, that connection is difficult to undo.
The organizational buyer also changed what a clean transaction meant. A brand protection team that acquires a domain for defensive purposes has a different tolerance for risk than a speculative investor. It needs to know that the seller is entitled to sell. It needs to know the name was not used in a way that carries trademark liability. It needs a documented chain of transfer. These requirements pushed intermediaries toward more formal due diligence, which in turn raised the professional floor for the whole market.
## Mistakes the market made, and kept making
Thirty years of a market produce a short list of recurring errors. These are not exotic failures. They are the same failures, repeated by different people in different settings.
**Transferring before funds were confirmed.** This was the first error of the DNS aftermarket, and it never fully disappeared. A buyer who proposes to pay outside escrow to save the cost of it is offering a trade that does not make sense, because that cost is the price of insurance against total loss. Dishonest counterparties use haste to get around protection. The seller who agreed to step outside the sequence almost always regretted it.
**Treating an opened escrow as a closed deal.** The pattern is common: the transaction is opened, both sides agree, and then the buyer never funds it. Opening an escrow commits nobody. A deal is not real until the funds are confirmed. Sellers who stopped talking to other buyers on the strength of an opened but unfunded escrow lost time and occasionally lost a competing offer.
**Failing to define what was being transferred.** A domain sale can include or exclude associated intellectual property, content and other names. Failure to document exactly what was sold caused a recurring category of disputes after closing. The terms of a transaction should set out the price, the name and every other asset included, and they should do so before funds move, not afterward.
**Mistaking published sales for the whole market.** Professionals who built valuations mainly on published sales data misjudged specific names in both directions, because the visible data was not representative. The same belief led observers to overestimate how transparent the market was and to misjudge its size. Valuation practice improved when public data came to be treated as one input among several, not as a complete picture.
**Announcing an acquisition before the transfer was final.** An announcement made while a transfer is still in progress leaves the buyer exposed if a technical or legal issue on the seller's side delays or blocks it. The pressure then comes from several directions at once, and the seller knows it. The rule that emerged: nothing is announced until the name is under the buyer's control and any agreed inspection period has passed without dispute.
## What transfers to onchain TLDs
The table below sorts the core practices of the DNS aftermarket by whether they carry over to onchain TLDs. The sorting reflects the structure of the transaction, not optimism.
| DNS practice | Transfers? | What changes, if anything |
|---|---|---|
| Escrow as the closing standard | Yes | Escrow is onchain; the sequence is the same |
| Intermediary as identity shield | Yes | Organizations need the same protection |
| Private channel for sensitive names | Yes | The registry record is public, so the channel matters more |
| NDA covering every exchange before closing | Yes | Same principle, same risk if skipped |
| Diligence on prior use and associations | Yes, modified | The onchain record is public and can be checked directly |
| Valuation before any approach | Yes, modified | Different inputs, same logic |
| Verification before announcement | Yes | The public record replaces the registrar check |
| Open auction for price discovery | In part | The strategic context differs |
| Expiry and renewal management | No | A Freename TLD carries no annual renewal fee |
| UDRP as the dispute channel | No | The policy applies inside the ICANN system |
The most important transfers are structural: escrow, identity management, and the discipline of not announcing before the asset is safely in hand. These practices are not artifacts of DNS technology. They are responses to the conditions of any high-value secondary transaction in which the asset is intangible, the parties may be strangers and the transfer is hard to reverse.
Escrow is the clearest case. The [escrow and transfer stage](/escrow-and-transfer) of an onchain TLD sale closes through onchain escrow on an aftermarket. The instrument is different from the escrow agent of a DNS sale. The principle that neither side goes first on faith is identical, and so is the rule that nobody steps outside the agreed sequence.
Diligence transfers, and in one respect becomes easier. In the DNS, establishing who controlled a name, and since when, could mean piecing together records kept by several parties. For an onchain TLD the registry record is public, and anyone can check it on a block explorer, a website that displays the contents of a blockchain. What the record does not tell you is how the TLD was presented and used, which names were issued under it and what those names were associated with. That part of diligence is the same patient work it always was.
[Valuation](/valuation) transfers with modification. In the DNS, a valuation rests on the strength of the extension, keyword relevance, length, traffic history and comparable sales. For onchain TLDs there are fewer public comparables and the inputs differ, but the logic of building a range from several inputs, instead of asserting a single figure, applies without change. So does the lesson about published data: a handful of visible sales is an input, not a market price.
The mandate transfers as well. The DNS market learned that an intermediary must work for one side, in writing. Nothing about a token changes that. A [holder who wants a buyer found](/sell-side-mandates) and an organization that wants [a holder found and approached](/buy-side-acquisition) are giving two different mandates, and each should know which one is in play before the first exchange.
## What does not transfer
Not every DNS practice belongs in an onchain TLD transaction. Some are irrelevant. Some would actively mislead.
**Renewal management has no equivalent.** A significant part of DNS aftermarket activity exists because registrations expire. Monitoring expiry dates, placing orders for names about to lapse and bidding on expired names are established activities in the DNS world. A TLD on Freename is a token recorded onchain and held in the holder's wallet, and it carries no annual renewal fee. The annual deadline that drives so much DNS activity therefore has no counterpart, and neither does the practice of waiting for a name to lapse. A professional who assumes there is a clock running on the holder is importing a DNS assumption that does not apply.
**The UDRP is not the dispute channel.** The Uniform Domain-Name Dispute-Resolution Policy is the procedure through which trademark owners challenge domain registrations. It binds registrants through the registration agreements of the ICANN system. An onchain TLD exists outside that system, so a brand protection team that expects to resolve an onchain conflict through the UDRP will find no procedural path. Trademark law itself is not confined to the DNS. What remains is a negotiated acquisition, which is the work of a desk such as ours, or enforcement under the law that applies, which is still developing for blockchain assets and varies by country. Your counsel decides which route fits a given situation.
**Privacy through a WHOIS proxy.** In the DNS, a proxy service can replace the registrant's contact details in the public record. Nothing comparable sits in front of an onchain record: the wallet address that holds a TLD is visible to anyone on a block explorer, although an address is not a name. Discretion therefore has to come from how the parties conduct the negotiation and who speaks to whom, not from a privacy layer on the record. The protection is upstream, not downstream.
**Traffic metrics as valuation inputs.** The value of a DNS domain is partly a function of direct visitors, inbound links and the search history attached to the name. Onchain names resolve outside the DNS, so an ordinary browser does not open them without support, and that support is partial and changes. [What works with an onchain TLD](/whats-live) is a different list from what works with a domain, and a valuation built on DNS traffic assumptions will be wrong in its foundations. The relevant inputs lie elsewhere: what the namespace can be used for, how closely the string matches a brand or a sector, and what issuing names under it can return. Only the holder of a Freename TLD can issue names under it, and the registry runs a royalty program that pays the holder a share of each registration fee for those names, advertised at up to half and subject to the registry's current terms and activation conditions. That return is not free of cost, and a valuation should net it: the absence of a renewal fee on the TLD does not extend to the program, which carries an annual cost set by the registry. Some TLDs were registered under earlier terms that include the program for life, and whether such terms pass to a buyer is a point to confirm with the registry before closing.
**The instinct that a registered name is a rented one.** A domain is held for as long as the registration is renewed, and much DNS contract practice is written around that fact. A TLD held as a token is controlled by whoever controls the wallet. That shifts operational risk from the calendar to the custody of the wallet: who can authorize a transfer, and what happens when that person leaves. A multisignature wallet, which requires several approvals before anything moves, is the kind of arrangement an organization will want to consider with its own advisers.
## Sorting the practices before you need them
The value of reviewing the history of the DNS aftermarket is not nostalgia. The market already ran the experiments. The practices that survived are the ones that solved real structural problems, and those problems are not unique to the DNS: two strangers exchanging an intangible asset for a large sum, across borders, with no physical delivery.
If you hold a TLD and are considering a sale, take four things from this history: a valuation before any outreach, an NDA covering every exchange, a closing through escrow with a strict sequence, and no announcement until the transfer is confirmed. The page written [for TLD holders](/holders) sets out how the desk works with them. If you act for an organization considering an acquisition, add three more: decide how your identity is handled before the first contact, treat an unfunded escrow as no deal, and review the public history of the TLD before funds move.
The practices that do not transfer need to be removed from the mental model deliberately, not left in place on the assumption that they do no harm. Renewal deadlines, UDRP mechanics, WHOIS proxies and traffic-based valuation each carry an expectation about how the other side will behave, and each expectation is wrong here. An experienced DNS professional brings a toolkit that is mostly right. It needs to be sorted before it is used.
Hush Partners works within the part of that history that carried over. The desk acts under a mandate, either from a holder who wants a buyer or from an organization that wants a holder found and approached. Every exchange is covered by a standard mutual [NDA](/nda), and closing runs through onchain escrow on an aftermarket. To see how that is organized in practice, read the [rules of engagement](/rules-of-engagement) and then [how a file moves from mandate to transfer](/how-it-works).
## Questions people ask
### Does an intermediary replace legal counsel in an onchain TLD transaction?
No. The DNS market settled this early: the broker runs the negotiation and the process, and counsel reviews the agreement and advises on the law. The same division holds for an onchain TLD, with the added point that the law on blockchain assets varies by country and is less settled than the law on domain names. Nothing in this article is legal or tax advice, and your own counsel decides.
### Why does discretion need more care with an onchain TLD than with a domain?
It matters in both, but the onchain record cannot be masked the way a WHOIS record can. Anyone can see on a block explorer which wallet address holds a TLD, and when that changed. The address does not name its owner, but a transfer is visible to anyone who is watching, including a competitor who knows what to look for. An organization should therefore settle who makes the approach, and under what confidentiality terms, before the first contact and not after it.
### When does a public sale make more sense than a private one for an onchain TLD?
The DNS lesson applies in general terms. A public sale reaches a broad audience and lets competition set the price, which is useful when the holder cannot tell who the likely buyer is. A private sale reaches a narrower, deliberately chosen audience, keeps both parties out of view and leaves the seller in control of timing. For a TLD that plainly matters to an identifiable class of organizations, such as `.yoursector`, a private approach to that class is usually the more efficient route. For a TLD with broad appeal and no obvious buyer, an open process may find demand that a private approach would miss.
### What should a brand protection team review before acquiring an onchain TLD?
Start from the public record: which wallet address holds the TLD and how it has moved, both of which can be checked on a block explorer. Then review which names, if any, have been issued under the TLD, since those names carry their own associations. Where names issued under `.yourbrand` conflict with the trademarks of third parties, that history does not disappear on transfer and may leave residual risk. Your counsel assesses the specific exposure and the terms that should cover it.
### Does the absence of a renewal fee change how a holder should time a sale?
Yes. In the DNS, a holder who delays a sale keeps paying to renew, and a missed renewal can end the registration. That creates a soft deadline which shapes negotiating behavior on both sides. A Freename TLD has no annual renewal fee, so that deadline is absent. A holder can take a longer view on timing and negotiate from a more neutral position. For buyers and sellers alike, pressure tactics that rely on a renewal date have no basis here.