Continuation of the proposal above: the sections that did not fit the post length limit, with their original numbering.
From §2 — the installed base
The installed base, from the registrar’s records big map:
| Length | Names ever registered | Active today | Expiring within 12 months |
|---|---|---|---|
| 1 char | 36 | 0 | 0 |
| 2 char | 5 | 3 | 2 |
| 3 char | 1,628 | 81 | 39 |
| 4 char | 4,955 | 447 | 210 |
| 5+ char | 138,912 | 24,389 | 10,598 |
| Total | 145,536 | 24,920 | 10,849 |
From §6 — what the short tiers are worth
What the short tiers are worth. Repricing only the 5+ tier and leaving short names at today’s prices produces $20,000 / $28,100 / $34,200 across the same three scenarios. Including the short tiers adds $9,300 / $17,200 / $27,700 — about 38% of projected income in the central case, and 47% of the uplift over today, from 9% of the transactions.
4.2 Users lost, against comparable Tezos applications
Unique addresses that paid for a buy or renew, against unique addresses transacting with five long-running Tezos consumer applications over the same windows. None of the comparators changed its pricing.
| 2024 H1 | 2025 H1 | 2026 H1 | 25H1→26H1 | 24H1→26H1 | |
|---|---|---|---|---|---|
| Tezos Domains | 2,702 | 1,416 | 864 | −39% | −68% |
| objkt.com Marketplace v2 | 10,011 | 3,439 | 1,950 | −43% | −81% |
| fxhash Marketplace v2 | 2,942 | 1,463 | 796 | −46% | −73% |
| Teia Community Marketplace | 3,627 | 2,264 | 1,482 | −35% | −59% |
| SpicySwap Router | 1,142 | 298 | 195 | −35% | −83% |
| Plenty | 208 | 66 | 49 | −26% | −76% |
| Comparator median | −35% | −76% |
In the year after the price tripled, Tezos Domains lost 39% of its paying users. The comparators, with no price change at all, lost between 26% and 46%, median 35%. Over two years Tezos Domains lost 68% against a comparator median of 76% — that is, over one year it sat inside the range the ecosystem lost anyway, and over two it retained users better than the ecosystem, while charging three times as much.
5. Why these particular short-name prices
Three independent lines of evidence, all pointing above where we are.
5.1 The market has already paid these prices — repeatedly
Expressing what buyers actually paid in USD at the exchange rate of the day:
| Period | 3-char price/yr | 3-char purchases | 4-char price/yr | 4-char purchases |
|---|---|---|---|---|
| 2021 H2 | $563 | 79 | $141 | 703 |
| 2022 H1 | $320 | 155 | $82 | 499 |
| 2022 H2 | $152 | 1,299 | $38 | 3,348 |
| 2023 H1 | $96 | 47 | $24 | 199 |
| 2023 H2 | $75 | 49 | $17 | 203 |
| 2024 H1 | $100 | 14 | $25 | 46 |
| 2024 H2 | $68 | 3 | $17 | 45 |
| 2025 H1 | $79 | 6 | $14 | 41 |
| 2025 H2 | $31 | 17 | $8 | 72 |
| 2026 H1 | $18 | 9 | $4 | 56 |
The 2021–22 figures are a bull market and should be discounted. 2024 H1 should not be: it was a bad market, roughly comparable to now, and 4-character names sold 46 units in six months at $25/year, with 261 renewals across that year. Since then the effective price has fallen by 84%, to $4, and half-year volume has gone from 46 to 56 — a 22% gain for an 84% price cut, while the 5+ base fell 77% over the same span.
Reading that honestly: short names do respond to price a little more than long ones, but nowhere near enough to justify the discount. $40 for a 4-character name is above the highest recent proven point ($25) and a quarter of what the market paid in 2021. $150 for a 3-character name is between the $96 and $152 points that both cleared volume, and well below the $563 of 2021 H2.
5.2 The comparison set
| 5+ char | 4 char | 3 char | 2 char | 1 char | |
|---|---|---|---|---|---|
| ENS | $5 | $160 | $640 | not sold at fixed price | not sold at fixed price |
| .tez today | $0.66 | $2.66 | $11.06 | $44 | $177 |
| .tez proposed | $5 | $40 | $150 | $400 | $1,000 |
The proposal matches ENS at the base tier and deliberately sits at roughly a quarter of ENS on the short tiers, on the grounds that the Tezos ecosystem is smaller and the names are worth correspondingly less. It does not attempt ENS’s 32× and 128× multiples of the base price; the proposed multiples are 8× and 30×.
5.3 The two shortest tiers are a special case
One character. All 36 single-character names (a–z, 0–9) were held with a reserved registration that expired on 10 December 2025. They have been openly purchasable at 800 XTZ/year ever since — through a period when that was worth between $450 and $177 — and not one has been registered. Nearly nine months of zero at a falling price says the constraint is not the price; it is that nobody knows they are available. The proposed $1,000 target is therefore set to hold the floor rather than to clear the inventory: it prevents 36 permanently scarce names being captured cheaply, and it prices them sensibly if and when they are marketed properly. Actually releasing them — an auction, or an announced release window — is worth doing, but it is a separate proposal with its own mechanics.
Two characters. Six registrations covering five distinct names since the tier opened in August 2025 (gm, ai, in, me, 69; three still active). At 200 XTZ the tier cleared roughly one name every two months at a USD price that ranged from $162 down to $42. There are 1,296 possible two-character labels and five have ever been taken. The $400 target roughly restores the tier to its opening USD price. Revenue either way is small; the point is not to sell the scarcest inventory in the registry for $42.
4.3 The chain underneath
| 2024 H1 | 2025 H1 | 2026 H1 | |
|---|---|---|---|
| Contract calls (chain-wide) | 14,764,652 | 6,390,301 | 5,240,888 |
| New user accounts (chain-wide) | 369,899 | 117,550 | 30,487 |
| Tezos Domains actions | 9,237 | 4,345 | 2,251 |
| Tezos Domains new customers | 834 | 408 | 223 |
New-account formation on Tezos fell 92% across the two years; Tezos Domains’ new-customer count fell 73%. Contract calls are the weakest of the three tests — the figure is dominated by oracles, bots and automated routing rather than people — and it is included only for completeness.
9. Implementation
Prices live in the TLD registrar’s config map (KT1Mqx5meQbhufngJnUAGEGpa4ZRxhPSiCgB), as a standard price per day in picotez (10⁻¹² XTZ). Key 2 is the default (5+ characters); keys 2001–2004 override it for lengths 1 to 4.
| Key | Length | Current value | Current XTZ/yr | New value | New XTZ/yr |
|---|---|---|---|---|---|
2001 |
1 | 2,191,780,821,918 | 800 | 12,328,767,123,288 | 4,500 |
2002 |
2 | 547,945,205,479 | 200 | 4,931,506,849,315 | 1,800 |
2003 |
3 | 136,986,301,370 | 50 | 1,863,013,698,630 | 680 |
2004 |
4 | 32,876,712,329 | 12 | 493,150,684,932 | 180 |
2 |
5+ | 8,219,178,082 | 3 | 63,013,698,630 | 23 |
The new values above are computed at $0.2213, the 90-day average used for valuations throughout this proposal, and rounded to a round number within 2% of the USD target. The values actually submitted must be recomputed against the 30-day average on the day the multisig proposal is prepared, per the rule in §7, and re-verified before signing. The rate used becomes the reference against which the ±25% band in §7 is measured thereafter.
Execution: the registrar’s owner is the DAO treasury KT1BzeXvLtPR83aj5FHemXmia6DmdXkeV3Uk, so the change is a treasury multisig proposal (create_proposal → sign_proposal → resolve_proposal) carrying an execute_lambda that calls admin_update on the registrar with the five config writes in a single operation. No contract upgrade is required. The multisig needs four of the seven keyholders to sign, so the time from the vote closing to the change taking effect depends on their availability; past executions took four to six days from the first signature. No execution date is announced: the change is live when the multisig proposal resolves.
After execution: verify the new values on-chain, update the pricing page and documentation, and publish a short post stating the change, the rate used, the USD targets, and the conditions under which prices can be re-expressed.
Appendix — method and sources
All figures were derived on 31 August 2026 from public chain data via the TzKT API; nothing here relies on internal records.
Transaction data. Every buy and renew operation against the registrar proxies KT191reDVKrLxU9rjTSxg53wRqj6zh8pnHgr (buy) and KT1EVYBj3f1rZHNeUtq4ZvVxPTs77wuHwARU (renew) from April 2021 to 31 August 2026 — 214,365 applied operations. Name length is the character count of the UTF-8 decoded label parameter; the implied annual price is amount ÷ (duration ÷ 365).
Installed base. The registrar’s records big map (pointer 1278), 145,536 keys, each carrying an expiry timestamp. “Active” means expiry later than 31 August 2026.
Treasury income. Transfers into KT1BzeXvLtPR83aj5FHemXmia6DmdXkeV3Uk, 1 September 2025 to 31 August 2026, split by sender.
Exchange rates. api.tzkt.io/v1/statistics/daily?quote=usd — daily USD close per XTZ. Historical USD prices in §5.1 use the monthly average for the month of each transaction. All other USD conversions use the 90-day average, $0.2213 as a valuation convention; the 30-day average, which is the basis the §7 trigger uses, was $0.2104, and spot $0.2348. The §7 table was produced by replaying each candidate rule over daily XTZ/USD closes from May 2021 to September 2026, assuming each reset executes three days after its trigger.
Prior decisions referenced. P-013, P-016, P-017, P-021, P-023, P-024.
Caveats. Auction settlements (settle, 6,402 operations) are excluded from the per-length income table; they are 2% of registrar income and almost entirely historical. Multi-year registrations are counted at full value in the period they were paid, which is why 3-character income (7,100 XTZ from 64 actions) is larger than a per-year reading would suggest. The projection in §6 assumes the mix of registration durations is unchanged.