Introduction
The Q2 2026 Domain Intelligence Report analyzes aftermarket sales pricing from Dynadot's own marketplace data, April–June 2026, benchmarked against Q2 2025. It leads with medians and reports shares, ratios, and index values, making every finding citable.
Q2 saw typical sales remain flat, with value concentrated at both ends of the market and premium names defined by their structure rather than their subject matter. This highlights the importance of domain pricing rather than overall market growth. It's written for domain investors, naming professionals, and analysts seeking transparent pricing benchmarks.
This report covers:
- Median vs. average pricing, year over year
- Value concentration by price band
- TLD, length, and naming signals
- The four listing types
- What premium sales had in common
- Quality signals that moved price
Executive Summary
Q2 2026 was shaped by pricing and sales mix rather than overall growth, highlighting typical sale prices, where aftermarket value was concentrated, and the traits shared by the highest-value domains. Five findings define the quarter:
- The typical sale price held flat year over year. The median was $18.87 almost exactly where it was in Q2 2025 - $18.85. The average fell 26.5%, from $63.85 to $46.93. The listing mix shifted toward lower-priced inventory; the typical domain did not get cheaper.
- Value is extremely concentrated. 88.9% of orders closed under $50 and produced 33.9% of marketplace value. The 0.5% of orders at $1,000 or above produced 30.3% on their own.
- The four listing types behave as four separate markets. Median prices ranged from $2.11 in user auctions to $24.99 in backorders, a spread of roughly 12x on the same marketplace in the same quarter.
- Premium sales share a distinct profile. Among sales above $5,000, not a single domain was hyphenated or internationalized, and 57.5% were five characters or shorter, compared with 13.1% of sales under $50.
- AI-keyword domains showed no price premium. On a word-boundary match AI-keyword domains averaged 1.03x the market price and their median exactly matched the market. Backlinks did little for the typical sale either, the signal that actually marked the premium tier was cross-TLD registration.
Key numbers at a glance
| Signal | Reported figure |
| Median sale price Growth, year over year | $18.87 vs $18.85 (+0.1%) |
| Average sale price decline, year over year | -26.5 (mix-sensitive) |
| Share of orders closing under $50 | 88.9% |
| Share of value from orders at $1,000+ | 30.3% |
| .com share of aftermarket orders | 70.6% |
| .com share of sales above $5,000 | 60.0% |
| Sales above $5,000 that were 5 characters or shorter | 57.5% |
| Median cross-TLD registrations, $5,000+ sales vs all sales | 26 vs 2 |
Methodology
Marketplace data
This analysis draws on more than 54,000 completed aftermarket sales across four Dynadot listing types in Q2 2026:
- Expired domain orders : auctions of lapsed registrations.
- Buy-It-Now (BIN) : fixed-price seller listings.
- User auctions : seller-initiated auctions.
- Backorders : pre-orders to capture a dropping name.
How to read the numbers
Medians lead throughout, because averages move whenever the listing mix moves. Averages are labelled wherever they're quoted.
Age, backlink, and cross-TLD data are not supplied for backorders, so backorders are excluded from every quality analysis (this is stated where relevant). This edition reports no buyer geography, because the dataset contains no reliable location signal.
Premium-sale characteristics describe sales that occurred, not causes of price. Nothing in this report is advice on what to buy or register.
1. Domain Pricing On Our Marketplace
The typical domain sold for the same price it did a year ago: the median moved just one tenth of one percent, from $18.85 to $18.87.
The average sale price, meanwhile, fell by a quarter(or-26.5%). When those two numbers pull apart that sharply, it means the mix of what sold changed, not the prices themselves: more low-priced inventory entered the market, and the very top thinned out. The share of value carried by orders at $1,000 or above fell from 44.3% to 30.3%. In short, Q2 2025 had a heavier tail of very large sales; Q2 2026 had a broader base of small ones.
The table below shows the year-over-year price comparison, led by the market median, with listing-type medians arranged from highest to lowest.
| Measure | Q2 2025 | Q2 2026 | Change |
| Median sale price | $18.85 | $18.87 | +0.1% |
| Average sale price (mix-sensitive) | $63.85 | $46.93 | -26.55% |
| Median: expired domain orders | $18.85 | $18.87 | +0.1% |
| Median: BIN orders | $4.13 | $5.00 | +21.1% |
| Median: user auctions | $3.00 | $2.11 | -29.7% |
| Median: backorders | $24.99 | $24.99 | 0.0% |
| Share of value from $1,000+ orders | 44.3% | 30.3% | -14.0pt |
Median sale price by listing type, Q2 2025 vs Q2 2026
2. How Order Price Shapes Aftermarket Volume and Value
Fewer than 1% of aftermarket orders generated nearly a third of marketplace value.
Here's an overview of how each price band contributed:
- Orders at $1,000 or above were 0.5% of all orders and produced 30.3% of value.
- Orders above $5,000 were roughly seven in ten thousand and carried 11.9% of value on their own.
- The 88.9% of orders under $50 produced 33.9% of value.
The marketplace floor and the marketplace summit contribute about equally, from bases three orders of magnitude apart. The aftermarket reads as a volume business and behaves as a long-tail one; both are true at the same time, and most commentary picks only one.
Where orders live vs where value lives, Q2 2026
3. TLDs, Length, and Naming Signals
The choice of the TLD matters as well as the length of the domain name, but some naming signals are more important than others.
.com dominated volume but priced below the market
.com took seven in ten orders but priced below the market average. 354 distinct TLDs traded in Q2 2026. Volume and price rank almost inversely: .com holds 70.6% of orders at 0.96x the market average, because it is where the cheap end of the market transacts, while the highest-priced extensions are all under 0.7% of volume.
.io is the honest premium story here,with a median of $61.49, more than three times the market median.
.us shows a 2.74x average but a $14.95 median; that average is driven by a few large sales and shouldn't be quoted on its own.
| TLD | Share of orders | Median | Avg vs market |
| .com | 70.6% | $18.87 | 0.96x |
| .org | 6.4% | $18.52 | 0.88x |
| .cc | 4.1% | $16.79 | 0.53x |
| .net | 3.9% | $20.51 | 0.94x |
| io | 0.6% | $61.49 | 2.93x |
| .us | 0.6% | $14.95 | 2.74(tail-driven) |
TLD demand looks different in each acquisition channel
Split by listing type, several extensions turn out to trade almost entirely through a single channel.
| TLD | Expired orders | User auctions | Backorders |
| .com | 75.0% (1.00x) | 76.5% (1.02x) | 74.6% (0.99x) |
| .org | 6.4% (0.97x) | 4.6% (0.70x) | 9.1% (1.40x) |
| .cc | 3.8% (1.17x) | 0.7% (0.23x) | 0.0% (—) |
| .cn | 0.0% (—) | 0.0% (—) | 2.9% (10.1x) |
| .co | 0.8% (0.78x) | 1.0% (0.99x) | 3.0% (2.85x) |
| .online | 0.1% (0.32x) | 2.6% (12.6x) | 0.0% (—) |
| .ai | 0.0% (—) | 1.1% (17.1x) | 0.0% (—) |
Table note: within-channel share of orders; index in brackets vs the three-channel combined mix. Buy-It-Now is excluded because bundled BIN orders cannot be attributed to a single TLD per sale.
- .com is channel-agnostic — within two points of the same share in all three channels.
- Country codes are channel-specific. .cn and .ca sell almost only via backorders (~10x over-index): buyers plan for the drop. .cc is the mirror image, an expired-auction phenomenon with no backorder placements. .ai and .online are seller-supplied, surfacing almost only in user auctions.
- Backorder competition is TLD-selective. Medians sit at each TLD’s base fee; about 4% of backorders closed over 10% above it — roughly one in four for .co and one in six for .cn and .io, versus under 3% for .com. Contested backorders settled at a median 2.5x the base fee.
- Expired auctions carry the premium tails. 6.3% of expired orders closed at $100+, rising to 24.9% for .io and 21.0% for .us, versus 5.4% for .com.
Short names raised peak prices, but not the minimum prices.
Length was one of the quarter's clearest pricing patterns, but it worked in a more specific way than "shorter is more expensive." Here's how the numbers break down
- Names of 1 to 5 characters averaged 1.65x the market price — the strongest length premium in the dataset.
- The curve falls steadily as names get longer , bottoming out at the 11-to-15 character bucket (0.76x).
- Past 15 characters, the curve flattens , and 21+ character names actually priced slightly above 16-to-20.
- The short-name premium varies by channel: it's steepest in user auctions (4.04x) and gentlest in expired auctions (1.87x).
- The median was $18.87 in every length bucket — a nuance most length analyses miss.
What does that last point mean in practice? Averages rise for short names, but the typical sale price doesn't. Brevity doesn't make the ordinary sale more expensive, it raises the odds of an exceptional one. Short names lift the ceiling of what a domain can sell for, not the floor of what it usually does.
Numbers priced high, hyphens priced low
Two naming patterns stood out clearly in Q2, one in each direction.
Pure numeric labels were rare but valuable. Domains made up entirely of digits accounted for just 0.6% of orders, yet they averaged 2.45x the market price.
Hyphenated names were the clearest discount signal in the dataset. Domains containing a hyphen made up 5.0% of orders and averaged just 0.70x the market price. Buyers consistently paid less for names broken up by punctuation.
Among keyword themes, just three priced meaningfully ahead of the market:
- green energy at 1.36x
- agents and bots at 1.17x
- general tech at 1.14x
Every other keyword theme is priced at or below the market.
AI keywords carried no premium
AI-keyword domains showed no price premium in Q2 aftermarket sales. They averaged 1.03x the market price, and their median, $18.87, was identical to the market median.
One note on how we counted them: simply searching for "ai" in a name catches far too much, and words like chair, email, and domain all match. We used word-level matching instead, which still catches genuine AI names like aitools.com. Measured properly, whatever premium AI naming once carried, it doesn't show up in these sales.
4. How Each Listing Type Behaves Relative To The Market
The four listing types don't just differ, they behave like four separate markets, with different price distributions, TLD mixes, and odds of a premium sale.
| Expired | BIN | User Auctions | Backorders | |
| Share of orders | 71.0% | 15.7% | 4.9% | 8.4% |
| Share of value | 68.6% | 20.5 | 3.4% | 7.5% |
| Median price | $18.87 | $5.00 | $2.11 | $24.99 |
| Average price | $45.30 | $61.51 | $32.08 | $42.20 |
| Average / median ratio | 2.4 | 12.3 | 15.2 | 1.7 |
| Median price, year over year | +0.1% | +21.1% | -29.7% | 0.0% |
| .com share | 75.0% | 46.6% | 76.5% | 74.6% |
| Short-name premium | 1.87x | 2.56x | 4.04x | 3.39x |
| Share of sales above $5,000 | 27.5% | 60% | 2.5% | +10% |
| Median age / links / cross-TLD | 2 / 20 / 1 | 0 / 10 / 2 | 0 / 14 / 1 | not supplied |
What distinguishes each, in one line apiece:
- Expired domain orders are the baseline. Seven in ten orders, the tightest price distribution, the oldest inventory, and the channel that single-handedly holds the market median flat.
- Buy-It-Now is where the extremes live. The lowest routine prices sit here alongside the largest single sale of the quarter and most of the premium tier (60% of all $5,000 sales), and its TLD mix is unlike anywhere else, with far less .com.
- User auctions are a different market entirely, with the lowest median ($2.11) and the steepest length curve.
- Backorders run the highest median, with prices clustered at a standard fee point. They were also the only channel where value grew faster than volume within the quarter ( +62.2% vs +14.8%, April to June ) , a shift worth a product-side explanation.
5. Premium Sales Common Characteristics
The quarter's most expensive domains looked remarkably alike: short, clean, and unhyphenated.
The quarter’s largest sales
| Growth rank | Domain | Price | Listing type | Month |
| #1 | tenvels.com | $25,000.00 | Buy-It-Now | June |
| #2 | betzz.com | $15,210.88 | Expired auction | April |
| #3 | personio.ai | $13,023.61 | User auction | April |
| #4 | 2290.com | $12,909.88 | Expired auction | May |
| #5 | (omitted: brand-safety rule) | $12,125.01 | Buy-It-Now | May |
| #6 | 5jl.com | $10,211.02 | Expired auction | April |
| #7 | glws.com | $9,999.00 | Buy-It-Now | June |
| #8 | 4439.com | $9,910.88 | Expired auction | April |
| #9 | tgh.com | $9,772.80 | Backorder | May |
| #10 | dyno.org | $8,888.00 | Buy-It-Now | May |
One sale is omitted under the report’s brand-safety rule; it remains in every cohort statistic below.
Shared characteristics by price cohorts
Among Q2’s largest sales, not one domain was hyphenated, and more than half were five characters or shorter.
Every gradient below moves in one direction as price rises.
| Characteristic | All orders | Under $50 | $1,000+ | $5,000+ |
| Average label length | 10.2 | 10.4 | 8.6 | 6.0 |
| 5 characters or fewer | 14.0% | 13.1% | 26.8% | 57.5% |
| Pure numeric | 0.6% | 0.6% | 1.4% | 5.0% |
| Hyphenated | 5.0% | 5.2% | 3.4% | 0.0% |
| Internationalized (IDN) | 0.4% | 0.4% | 0.3% | 0.0% |
| .com share | 70.6% | 71.6% | 61.9% | 60.0% |
| Sold via Buy‑It‑Now | 15.7% | 16.5% | 39.5% | 60.0% |
| Median cross‑TLD registrations | 2 | 1 | 7 | 26 |
What changes as prices rise: Characteristics by price cohort, Q2 2026
Reading the table from the strongest signal down:
- Brevity is the clearest marker. The share of five-character-or-shorter names more than quadruples between the floor and the top tier.
- Hyphens and IDNs disappear entirely above $5,000 - they don't just thin out.
- Premium names were already registered in 26 other TLDs, against 2 market-wide.
- Buy-It-Now carried 60% of the top tier against 16.5% of the floor.
- .com share slightly falls as price rises, from 71.6% to 60.0%. That one is counter-intuitive and citable: the premium end of the market is more TLD-diverse than the floor, not less.
6. Age, Links and Cross-TLD Demand: Which Quality Signals Actually Moved Price
This section covers expired, BIN and user auction sales only. Backorders are excluded because age, backlink and cross-TLD data are not supplied for that channel. Medians per bucket, not averages, because the price distribution is extreme enough that averages mislead.
Quality signals and the typical sale price, Q2 2026 (backorders excluded)
Age barely moved the typical sale
A domain's age made almost no difference to what it typically sold for. From 1 year old to 20, the median price sat exactly at the market median, no gradient at all. Only two ends of the range behaved differently: brand-new or recently reset names sold at 0.63x the market price, and names older than 21 years carried a modest premium at 1.32x.
The takeaway is that age raises the odds of an exceptional sale, not the price of an ordinary one. An older domain has had more time to accumulate whatever makes a name valuable, but by itself, age isn't what buyers paid for.
Backlinks were nearly worthless as a price signal
Zero-link names are priced at 0.89x, and past that threshold, the movement is very slow. A name with ten thousand backlinks carried a median just 5% above one with fifty.
The premium cohort agrees, with a median of 14 links among $5,000+ sales against 18 market-wide. For a market that habitually quotes link counts in listings, that is a genuinely contrarian result. Backlinks barely tracked price at all.
Cross-TLD registration is the signal that works, and it is a threshold, not a slope
Cross-TLD registration counts how many other extensions the same name is already registered in. In effect, a measure of how many other people wanted that string.
It behaves like a threshold rather than a gradual slope. Below 26 other registrations, the count tells you nothing useful about price. Past that point the median starts climbing, and names registered in more than 100 other extensions sold for nearly twice the typical price.
It's also the only quality signal that keeps working at the very top of the market. The median $5,000+ sale was registered in 26 other TLDs against 2 market-wide, thirteen times as many. What premium names have in common isn't their own history; it's demonstrated demand for the string itself.
Which signal dominates?
None, everywhere - and that is the finding. Each signal works in a different part of the market, which is why no single metric prices a domain on its own.
Across the broad market, age tracks price most closely in rank terms (+0.43), largely because brand-new names cluster in the cheap tiers.
At the premium tier, age and links stop distinguishing entirely and cross-TLD demand and brevity take over: the median $5,000+ sale was no older and no better-linked than the median $19 sale. It was half the length and wanted thirteen times as many extensions.
Length is the strangest of the four signals. It shows no correlation with price overall, because the median is flat across every length bucket, yet it produces the sharpest shift of any characteristic at the premium tier. Length does not price the typical sale; it selects for exceptional ones.
Conclusion
Q2 2026 was a story about price and composition, not growth. The typical domain sold for almost the same price as a year ago (a median of $18.87) while the falling average simply reflected a mix that shifted toward cheaper inventory.
Beneath that flat headline, the market split in two. The 88.9% of orders under $50 and the 0.5% at $1,000 or above produced roughly equal value, and the four listing types priced so differently that these findings can help you decide which channel to focus on.
What separated the top of the market wasn't topic or history. Premium names were short, clean, and already registered across many other extensions, and backlinks did little for the typical sale. Structure and demonstrated demand for the string, not the subject matter, defined the quarter's most expensive domains.




