IPv6 Adoption Statistics 2026: UK Telemetry and IPv4 Market Costs

The 2026 UK IPv6 State of the Nation: Eyeball Reach vs Enterprise Inertia

Network telemetry confirms that the United Kingdom operates in an active dual-stack environment. APNIC measured more than half of its UK test samples as IPv6-capable and IPv6-preferred; the result does not identify the providers or causes behind that measurement. However, aggregate country-level traffic figures mask potential variation across sectors: country-level telemetry cannot establish how consumer, enterprise or B2B environments differ; organisations should measure their own networks and services.

This divergence creates an operational paradox for IT leadership. Because residential and remote mobile workforces initiate sessions over IPv6-enabled carrier connections, unmanaged dual-stack behaviour is already occurring on enterprise endpoints. Meanwhile, internal datacentre fabrics, legacy line-of-business applications, and private campus subnets are often maintained on IPv4-only configurations. Addressing this gap requires understanding how traffic is actually measured at the carrier edge versus the internal enterprise boundary.

UK Telemetry and the Adoption Gap: What APNIC Measurement Reveals

The most comprehensive public measurement of UK protocol readiness is provided by APNIC Labs, which uses automated measurement techniques to evaluate end-user capability and protocol selection across millions of real-world client sessions. APNIC’s seven-day measurement for 26 October–1 November 2026 recorded 54.45% IPv6 capability and 53.55% IPv6 preference in the UK across 49,400,931 samples, demonstrating widespread dual-stack reach at the client edge.

In APNIC's analytical framework, capability reflects the proportion of tested endpoints that have a functional IPv6 path to the wider internet, while preference reflects the proportion of dual-stack endpoints that select IPv6 when both protocols are viable. The 0.90-point gap indicates that capability and preference were close in this measurement; it does not by itself identify the client algorithm or explain causation.

Crucially, connection telemetry must be distinguished from organisational adoption. As noted in the methodology, these metrics capture client request behaviour across automated test scripts rather than the proportion of UK corporate leased lines, enterprise internal applications, or public-facing enterprise services that have deployed IPv6.

This eyeball telemetry is heavily driven by residential and mobile ISP deployments rather than enterprise adoption. In the UK, major consumer providers like Sky and BT/EE have deployed native dual-stack IPv6 across millions of broadband and mobile connections, and the academic research network Janet (Janet6) has maintained mature IPv6 routing for over a decade. In contrast, Virgin Media's customer-edge IPv6 rollout has lagged behind peers, creating an uneven protocol footprint when remote employees connect back to corporate systems.

The Real Cost of IPv4 Dependence: Transfer Pricing, Premiums, and Leases

While organisations delay internal IPv6 rollout, acquiring additional IPv4 space to support expanding infrastructure incurs substantial financial outlays. The global secondary IPv4 market continues to reflect scarcity, with transaction data demonstrating marked per-address pricing premiums for smaller block allocations that enterprise buyers typically purchase.

According to the October 2026 ip4.market index published by DCXV, the 30-day median transfer price for /24 blocks (256 addresses) stood at $25.34 per address across 26 transactions, closely matching its 90-day median of $25.25 across 79 sales. In contrast, larger blocks traded at lower unit rates due to volume dynamics: /23 blocks (512 addresses) recorded a 30-day median of $21.48 per address, /22 blocks (1,024 addresses) traded at $20.00, /21 blocks (2,048 addresses) stood at $16.98, and /20 blocks (4,096 addresses) recorded $15.38 per address. DCXV’s global ip4.market benchmark reported a /24 30-day median of $25.34 per address across 26 sales; treat prefix-level figures, especially the /20 result based on two sales, as indicative rather than UK quotations.

Broker and transaction records from earlier in 2026 illustrate broader market volatility. CircleID reported that blocks sized /16 and larger rose from approximately $10 per address in May 2026 to roughly $14 in September 2026, while average prices for /22 to /24 blocks averaged $23. IPv4Center’s 2026 table reported an average /16 transaction value of $857,211 ($13.08/address), an average /20 value of $68,731 ($16.78/address), and a /24 average of $26.70/address; these are source-specific averages, not a single market-wide quote. Furthermore, DCXV separately reported mid-2026 /16-or-larger levels of about $10–$13 per address and H1 2026 deal-flow ranges of $15–$22; these should not be treated as identical measures, with IPbnb listing /24 blocks around $30 and /16 blocks around $17 in November 2026. For practical budgeting across these divergent benchmarks, enterprise network planners should use an indicative envelope of $15 to $30 per IP (£12 to £24 per IP) depending on prefix size, broker fees, and transfer timing.

For UK IT buyers, secondary market pricing must be evaluated through international commercial realities and currency exchange. The cited public benchmarks are denominated in US dollars, reflecting international RIR transfer norms. At an indicative exchange rate of £1 = $1.25, the headline median of $25.34/IP equates to roughly £20.27 per address, meaning a standard /24 block (256 addresses) costs approximately £5,189 ($6,486.80) before accounting for UK VAT, broker commission, escrow charges, and RIPE NCC transfer administration fees. In contrast, an indicative lease benchmark of $0.30–$0.50 per address per month equates to roughly £0.24–£0.40 monthly per IP (£61–£102 per month for a /24). While leasing avoids immediate capital expenditure, multi-year leasing creates ongoing operational expenditure that increases the total cost of maintaining IPv4 dependence.

The 'Shadow IPv6' Threat: Unmanaged Dual-Stack on Corporate Networks

A network may not be IPv4-only merely because the infrastructure team has not deliberately configured IPv6. In practice, modern operating systems generally include IPv6 support and may configure addresses automatically when an IPv6-enabled network advertises a usable prefix; verify endpoint and network policy rather than assuming routable IPv6 is present.

This unmanaged connectivity creates 'shadow IPv6', where corporate client devices establish external connections outside the visibility of IPv4 security appliances. If IPv6 is not covered by firewall, VPN and monitoring policy, IPv6 traffic can bypass intended controls; verify the specific deployment rather than assuming this occurs. Audit controls for protocol parity to prevent split-tunnel VPN or unmonitored local IPv6 connections from communicating directly outside corporate inspection.

Security operations teams can audit for shadow IPv6 by monitoring network segments for ICMPv6 Router Solicitation and Router Advertisement packets (type 133 and 134) using standard packet capture tools. On managed endpoints, command-line inspection using standard tools confirms whether operational interfaces have acquired global unicast IPv6 addresses (outside the fe80::/10 link-local range). Network administrators must ensure that perimeter firewalls explicitly mirror security policies across both protocols, or enforce a deliberately tested edge-blocking policy where IPv6 is not yet supported.

Architecture Strategy: Dual-Stack, IPv6-Only with NAT64, and Edge-Proxying

Enterprise network engineering teams face three primary architectural patterns when addressing protocol modernisation: dual-stack deployment, native IPv6-only with translation bridging, or perimeter edge-proxying.

Dual-stack involves running IPv4 and IPv6 concurrently across all network interfaces, switches, and firewalls. While dual-stack provides maximum native compatibility with legacy external services, the UK National Cyber Security Centre (NCSC) highlights that dual-running expands the operational and attack surface. Routing tables, firewall access control lists, monitoring systems, and address management must be maintained in duplicate, which increases operational complexity and risks configuration drift if security policy parity is not enforced.

IPv6-only with NAT64/DNS64 is one option for selected greenfield environments; assess application compatibility, provider support and operational requirements before recommending it. In this architecture, internal endpoints receive only IPv6 addresses. When an application queries an external host that possesses only an IPv4 A record, the DNS64 synthesiser generates an IPv6 AAAA record within a defined translation prefix. The outbound traffic traverses a NAT64 stateful gateway that translates the packets into IPv4. While this design eliminates internal IPv4 address exhaustion, common compatibility checks must be completed prior to deployment, including auditing systems for hard-coded IPv4 literals, verifying upstream ISP and cloud provider translation support, and ensuring operational tools can inspect synthetic prefixes.

For external-facing web services, an intermediate edge-proxy model allows organisations to present native IPv6 AAAA records to internet users via content delivery networks or application delivery controllers while maintaining legacy IPv4 connections to back-end origin servers. While edge-proxying accelerates external accessibility, it does not resolve internal address scarcity or eliminate shadow IPv6 risks on client networks.

Enterprise Address Planning: Slicing a /48 Without IPv4 Conservation Habits

Transitioning to IPv6 requires unlearning decades of IPv4 subnet conservation habits. In IPv4, network administrators routinely sliced address blocks into /28, /29, or /30 subnets to conserve scarce host space. For ordinary SLAAC-enabled LANs, /64 is the conventional subnet size; document exceptions for point-to-point, infrastructure and specialised networks. A /64 contains 2^64 address values—about 18.4 quintillion—but not all are necessarily assignable or usable under a particular addressing policy.

A /48 is a common enterprise planning example, but the assigned prefix depends on the RIR, ISP and documented address plan. A single /48 allocation contains 65,536 individual /64 subnets, allowing structured, hierarchical network engineering. Network teams can model their allocation using an IPv6 subnet calculator to partition the 16 available subnetting bits between routing domains.

Consider a mid-market UK enterprise or public sector entity aligning with Central Digital and Data Office (CDDO) IPv6 standards, allocated 2001:db8:a000::/48 via Janet or commercial transit. The 16 bits between /48 and /64 (represented by four hexadecimal characters) can be allocated systematically across sites and functional zones. The first hex nibble can denote geographical site (e.g., 2001:db8:a000:1000::/52 for London Headquarters, 2001:db8:a000:2000::/52 for Manchester branch operations). Within the London site, the subsequent bits define network tiers: 2001:db8:a000:1010::/64 for corporate users, 2001:db8:a000:1020::/64 for server management, and 2001:db8:a000:1030::/64 for guest wireless, fulfilling NCSC principles of network segregation while rendering address exhaustion concerns obsolete.

Methodology

This data study compiles publicly available network telemetry and secondary address market pricing data published across 2026. Telemetry for the United Kingdom was drawn from APNIC Labs' measurement platform, which analyses client connection behaviour across millions of web-based automated tests. The UK dataset represents a seven-day rolling average covering 26 October to 1 November 2026, comprising 49,400,931 verified samples evaluating dual-stack capability and protocol preference. These metrics reflect measurement-system telemetry rather than adoption percentages across UK organisations, websites, or broadband lines.

Commercial IPv4 address market statistics were compiled from verified broker transaction reports and market indexes published between January and November 2026. Spot medians and block-size comparisons were drawn from the DCXV monthly ip4.market report for October 2026, deal-flow analysis from CircleID's September 2026 report, IPv4Center's 2026 market review, and published market summaries from IPbnb and WBOC. Historical context regarding 2025 transaction volumes and depreciation trends was verified against secondary market reports.

All secondary market transaction figures are reported in US dollars (USD), reflecting the international standard for RIR address trading; no inferred or unverified GBP conversions have been applied. Figures evaluate transfer market purchase medians and published lease ranges, which represent address acquisition and operational leasing costs rather than universal operating costs applied to static legacy networks.

Strategic Verdict: The IPv6 Implementation Roadmap for 2026

In APNIC’s late-October 2026 UK measurement, IPv6 capability and preference each exceeded 50%, while secondary IPv4 block purchases maintained median prices between $15 and $26 per IP depending on allocation size. To navigate this landscape, UK IT infrastructure teams require a practical decision framework covering six core operational areas:

1. ISP support: Audit corporate broadband lines, leased lines, and cloud transit to confirm native IPv6 routing support before deploying prefixes.

2. Firewall and monitoring parity: Audit edge appliances, IDS/IPS, and log collectors to ensure policies and inspection capabilities match across both IPv4 and IPv6.

3. DNS and AAAA validation: Deploy dual-stack DNS resolvers and validate external AAAA records on ingress proxies and web tiers.

4. Application compatibility: Audit core enterprise applications and middleware to identify dependencies on hard-coded IPv4 literals before architectural changes.

5. VPN behaviour: Verify whether remote-access and split-tunnel VPN clients correctly inspect or encapsulate IPv6 traffic to prevent unmonitored external leaks.

6. Staged testing: Pilot IPv6 in isolated test subnets or guest networks before expanding to campus LANs, datacentre fabrics, or considering selective IPv6-only designs.

Sources

Every figure in this article traces to the sources below.

• •APNIC — UK IPv6 capability and preference measurement (26 Oct–1 Nov 2026)
• •DCXV — Monthly IPv4 Market Report October 2026 (ip4.market transfer index)
• •DCXV — H1 2026 Deal-flow Data and Block Size Recovery Analysis
• •IPv4Center — 2026 IPv4 Market Report and Block Size Pricing Table
• •CircleID — September 2026 IPv4 Market Report on Block Recovery
• •IPbnb — November 2026 IPv4 Purchase Pricing Ranges
• •WBOC — 2026 IPv4 Leasing Pricing, Quality, and Market Trends