.net Is Verisign's Shrinking Sidecar: 12.5M Names Stalled While .com Grows — and Wholesale Now Costs More

Open speedtest.net. Ookla's bandwidth test—hundreds of millions of runs per month, infrastructure-grade traffic on a suffix that still means "network" in DNS folklore. Open tech.net. In mid-2026 you are more likely to hit a parked page or a broker landing than a shipped product. Same suffix, opposite evidence quality. The thin stub this guide replaces never made that distinction. It rounded Verisign's disclosed 12.5 million .net base to a lazy "~13M", claimed a peak of "~16 million around 2015" when DNIB's 31 December 2014 cut shows 15.0 million—and has been falling or flat ever since—recycled AIO boilerplate about ChatGPT citation dividends, and priced wholesale at $9.92 when the 1 February 2024 fee schedule moved the registry service fee to $10.91 ($11.66 with ICANN's $0.75). Unaudited, delete-on-sight.

The story worth telling is not another ".com's reliable understudy" brochure. It is how the internet's original network suffix became a legacy infrastructure TLD in structural decline—down roughly 17% from its 2014 peak while .com added 51 million names in the same window—yet still runs on the same Verisign authoritative DNS that resolves 600+ billion queries per day, with wholesale pricing that in 2026 costs registrars more than .com despite carrying a fraction of the brand default.

Two layers: "network" origin vs fallback folklore

In IANA, .net is a generic top-level domain (gTLD) created 1 January 1985 alongside .com, .org, .edu, .gov, .mil, and .arpa. The first name in the zone was nordu.net—NORDUnet's research-network identifier, predating symbolics.com's formal registration on 15 March 1985 (Wikipedia / Verisign blog). Intended use: network infrastructure—ISPs, backbone operators, DNS and routing vendors. No restriction was ever enforced; open registration followed .com's commercial path.

In user culture, .net became ".com's understudy"—the suffix you register when the .com is taken or overpriced. That folklore peaked during the dot-com saturation of the late 1990s and held through the 2000s. It is weaker in 2026: startups now reach for .io, .ai, .dev, or .app before accepting a .net compromise—and Verisign's own numbers show .net stopped growing while those alternatives and .com both expanded.

Google adds nuance, not a bonus. John Mueller's standing guidance: keywords in the TLD are not a ranking signal. A .net site ranks on content and links; the suffix does not whisper "network" into the algorithm.

Who runs it—and where the zone is going

Registry operator (2026): VeriSign, Inc., Reston, Virginia—same operator as .com, under the 1 July 2023 .NET Registry Agreement (ICANN signed agreement PDF). Verisign also maintains Root Zone Maintainer services and operates two of the thirteen global DNS root servers. .net shares NetDNS resolution infrastructure with .com—29 consecutive years of 100% availability through mid-2026 (Verisign Q2 2026 release).

The economics tell a split story: infrastructure prestige vs registration momentum.

| Metric | Source / date | Figure | |--------|---------------|--------| | .net domain name base | Verisign DNIB, 30 Jun 2026 | 12.5 million (DNIB Q2 2026) | | .net base (prior quarters) | DNIB Q4 2025, Q1 2026 | 12.5 million — flat three quarters | | .net historical peak | Verisign DNIB, 31 Dec 2014 | 15.0 million (Q4 2014 brief) | | Decline from peak | 2014 peak → mid-2026 | ~−2.5M (−17%) | | .com base (same window) | DNIB 31 Dec 2014 → 30 Jun 2026 | 115.6M → 166.6M (+44%) | | .com + .net combined | DNIB 30 Jun 2026 | 179.1 million (+5.1% YoY combined) | | .net share of combined base | Same cut | ~7.0% (was ~11.5% at end-2014) | | Rank among gTLDs | Mid-2026 | #2 after .com; ahead of .org (~10.5M) | | Verisign registry revenue (total) | FY2025 10-K | $1,656.6 million (.com + .net + minor) | | DNS query volume | Q2 2026 earnings call | 600+ billion/day (.com/.net combined) |

Read the table carefully: Verisign's combined .com/.net base is growing—but every net-new name is landing in .com. From Q4 2025 through Q2 2026, .net reported 12.5M three times straight while .com climbed 161.0M → 166.6M. Jim Bidzos's 2026 guidance (1.5–3.5% combined base growth) assumes .com carries the load. .net is not in freefall—it renewed legacy at scale—but it is no longer a growth TLD.

Industry analysts flagged .net hemorrhaging to new gTLDs as early as 2014, when the zone sat at 15.2M and new strings were entering the root (Goldstein Report). A decade later the diagnosis holds: .net lost share; .com consolidated default status.

A short history—peak, plateau, pivot

• 1 January 1985 — .net created; nordu.net is the first name in the zone.
• 1990s — ISPs adopt .net as standard namespace: earthlink.net, att.net, mindspring.net—millions of dial-up users type .net by habit.
• Late 1990s — Dot-com saturation makes .net the primary fallback when .com is taken.
• 31 December 2014 — Zone peaks at 15.0 million registrations per Verisign DNIB.
• 2005 — Verisign divests .org to PIR; retains .com and .net under one operator.
• 2012 onward — ICANN New gTLD Program launches .tech, .dev, .io (ccTLD fashion), and hundreds more—siphoning "second choice" demand.
• 2014–2026 — Gradual decline to 12.5M; zone flat since at least Q4 2025 while .com adds millions.
• 1 July 2023 — ICANN renews `.NET Registry Agreement through 30 June 2028; 10% annual wholesale increase right retained (ICANN decision paper).
• 1 February 2024 — Wholesale registry service fee rises to $10.91 (+10% from $9.92) per ICANN fee schedule.

Pricing: the counterintuitive toll

.net has two price systems—registry shelf and aftermarket—and a paradox at the shelf: lower prestige, higher wholesale than .com in 2026.

Registry shelf (everyone pays something):

| Tier | Typical range (USD, mid-2026) | |------|------------------------------| | Verisign wholesale service fee (.net) | $10.91/year (since 1 Feb 2024) | | ICANN fee (.net) | $0.75/year | | Total .net wholesale | $11.66/year | | Verisign wholesale (.com, current) | $10.26/year | | Verisign wholesale (.com, from 1 Nov 2026) | $10.97/year | | Retail new registration (1 yr) | $12–$18 | | Retail renewal (1 yr) | $14–$20 (often above .com renewal at same registrar) | | Transfer | $11–$17 |

Under the 2023 .NET RA, Verisign may raise the service fee up to 10% per calendar year—same legacy cap structure since 2005, with no NTIA price mediation unlike .com's 7%-in-four-of-six-years NTIA rules. Model renewal at retail, not the first-year promo. The next 10% step on $10.91 would land near $12.00 service fee (~$12.75 total wholesale)—illustrative, not announced.

Why wholesale exceeds .com: Verisign sets .net pricing under a separate ICANN contract without the NTIA cooperative cap that constrains .com. Lower demand does not mean lower toll—registrars pass through on renewal.

Aftermarket (thin liquidity):

| Tier | Typical range | vs .com equivalent | |------|---------------|---------------------| | Random/low-quality strings | $5–$40 | Similar junk pricing | | Brandable two-word combos | $100–$2,500 | Often 10–25% of .com comp | | Short brandables | $500–$15,000 | 10–30% rule of thumb | | Single dictionary words | $5,000–$100,000 | Headline sales rare; broker-dependent | | Premium keywords | $50,000–$500,000 | Fraction of .com ceiling |

Most registered .net names never trade. Treat broker "appraisals" as fiction until a NameBio comp closes. For macro TLD context, see Popular Domain Extensions in 2026—which uses the same 12.5M DNIB baseline.

.net vs .com: same registry, opposite trajectories

Both strings resolve through Verisign NetDNS. Strategically they diverge:

| Dimension | .net | .com | |-----------|--------|--------| | Registry operator | Verisign | Verisign | | Zone size (Jun 2026) | 12.5M | 166.6M | | Momentum (2014→2026) | −17% from peak | +44% | | Cultural default | Fallback / infrastructure | Global muscle memory | | Wholesale (mid-2026) | $11.66 | $10.26 (→ $10.97 Nov 2026) | | Price-cap regime | 10%/year ICANN legacy | 7% in 4 of 6 yrs, NTIA-mediated | | Intended semantics | Network infrastructure | Commercial (now: "default") | | Best modern fit | ISP/CDN/dev docs/defensive | Consumer brands, institutional email | | Typo risk | Users type .com first | Users type .com correctly | | Aftermarket depth | Thin | Deepest globally |

See .com guide for the default-namespace economics. .net is not a prestige downgrade from .com—it is a different era's compromise that new cohorts increasingly skip.

Verified roofs—and the brochure names to delete

Skip Mad Libs (startup.net, brand.net, tech.net as "notable sites" without live products). These mid-2026 checks return mission-aligned surfaces:

speedtest.net — Ookla's speed test; among the highest-traffic .net properties globally.

battle.net — Blizzard's gaming platform; .net as persistent consumer-facing infrastructure brand (rare).

php.net — Official PHP documentation; developer gravity on the suffix's original audience.

behance.net — Adobe's creative portfolio; behance.com redirects—.net still carries the product.

slideshare.net — LinkedIn presentation hosting; B2B content staple.

sourceforge.net — Legacy open-source hosting; still active at scale.

nordu.net — First name ever in the zone; still resolves as NORDUnet's site—living DNS museum.

att.net / @comcast.net / @charter.net — Millions of ISP email addresses; .net as infrastructure residue in inboxes, not startup fashion.

cloudflare.com / stripe.com — Counterexamples: modern infra unicorns lead with .com, not .net. Do not treat "network company → .net" as 2026 law.

behance.com → behance.net pattern is exception; the norm for new brands is .com first, then niche TLDs—not .net.

SEO and AI search: parity in algorithm, penalty in habit

Google: No TLD ranking bonus for .net or .com. John Mueller: TLD keywords are not a ranking signal. Where .com wins measurably is CTR and direct navigation—users click and type what they trust. Industry tests consistently show .com outperforms .net on CTR for identical snippets—not algorithmic punishment, habit.

AI surfaces (ChatGPT Search, Perplexity, Google AI Overviews): Models cite pages with authority and freshness, not suffix semantics. speedtest.net appears because Ookla is a known tool, not because .net whispers "network" to model weights. No audited AIO preference layer exists—delete schema-template paragraphs expecting citation dividends.

For naming mechanics that actually move needles, see Does Your Domain Name Affect SEO?.

Risks registrants actually face

1. Structural decline, not zone collapse — .net is not going away; Verisign's contract runs through 2028 with renewal options. But liquidity and mindshare are shrinking relative to .com and modern alternatives.
2. "Second choice" perception — Users, investors, and acquirers read .net as "couldn't get .com." Real in diligence, fair or not.
3. Typo leakage to .com — 5–15% of direct-entry traffic intended for .net may reach the .com equivalent. Mitigation: own both or pick a name where .com is inactive.
4. Higher wholesale than .com — Paying more at registry shelf for less default trust is the 2026 paradox.
5. Compounding 10% hikes — Legacy RA permits annual increases; portfolio holders feel this at renewal scale.
6. UDRP / trademark — Standard ICANN UDRP applies. Defensive registration only helps if you actually use or monitor the name.

One routing question—not a who-should matrix

Skip symmetric columns. Ask:

Are you building on legacy infrastructure credibility—or settling for a suffix your audience will mentally autocorrect to .com?

Choose .net when you operate ISP, CDN, DNS, hosting, or network tooling where "network" still reads intentional, when you need a widely recognized gTLD and the .com is prohibitively priced in the aftermarket but .net is available at shelf, when you are defending a .com you already own, or when php.net-class developer documentation makes the suffix part of the product identity.

Skip .net when launching a consumer brand (type-in tax), an AI-native product (use .ai or .io), a developer tool where .dev or .io carry stronger 2026 cohort signal, or when .com is available at retail—the default is worth the $12/year even if .net is cheaper today.

Alternatives worth comparing:

• .com — Default namespace; growing zone, lower wholesale in 2026.
• .org — Mission/community signal via PIR; similar legacy scale (~10.5M).
• .io — Developer I/O fashion; separate ccTLD sovereignty stack.
• .dev — Google-operated, HTTPS-enforced, developer-native.
• .tech — Broader technology positioning without 1985 "network" baggage.

FAQ

How many .net domains exist in 2026?

Verisign DNIB: 12.5 million at 30 June 2026—flat since 31 December 2025. Not "~13M" and not growing.

When did .net peak?

15.0 million at 31 December 2014 per Verisign's Q4 2014 DNIB—not "~16M in 2015."

What is the wholesale price?

$10.91 registry service fee + $0.75 ICANN fee = $11.66/year since 1 February 2024. Verisign may increase up to 10% annually under the 2023 .NET RA.

Is .net cheaper than .com?

Not at wholesale in mid-2026. .net is $11.66 vs .com $10.26 (rising to $10.97 November 2026). Retail spreads vary; renewals often favor .com at the same registrar.

Is .net good for SEO?

No special algorithm bonus. Indirect disadvantages via CTR and direct navigation vs .com are real.

Who operates .net?

VeriSign, Inc. under ICANN contract effective 1 July 2023, term through 30 June 2028.

Should I register .net if .com is taken?

Only with eyes open: you are buying legacy recognition at declining momentum, paying premium wholesale, and accepting typo leakage. Often better to rename or buy .io/.ai for cohort signal—or budget for .com aftermarket.

---

Deciding between a $12/year .net compromise and a six-figure .com acquisition? Our domain search checks availability, premium tiers, and alternates in one pass—or read Brand Name vs Domain Name before you treat a shrinking legacy suffix as permanent identity.