2degrees: The Economics of a Challenger Telecom Operator in NZ

Main finding: a challenger now heavy on infrastructure

The directory entry "TWO-DEGREES-AS-AP" does not correspond to a simple hosting wrapper, but to the network resource footprint of 2degrees, New Zealand's third integrated telecommunications operator. The clearest technical correspondence is AS9790, whose APNIC/BGP record gives as-name: TWO-DEGREES-AS-AP, description "Two Degrees Networks Limited", country New Zealand, and APNIC organisation ORG-TDML1-AP. BGP.tools lists AS9790 as "Two Degrees Networks Limited", website 2degrees.nz, active with APNIC, registered on 16 December 1999 and classified as an "Eyeball" network with tags residential ISP, business broadband, and mobile data.

The public brand is 2degrees. The public website presents 2degrees as a "full telecommunications operator" and states 1,520 employees in New Zealand, 4,600 km of underground optical fibre, and 2,225 mobile towers in its network. The footer of the same public site uses the legal name Two Degrees Mobile Limited. This is important: the canonical public brand is simple, but the legal and network resource perimeter is not. "Two Degrees Networks Limited" appears in routing records and peering materials, while "Two Degrees Mobile Limited" appears in APNIC organisation data, the website footer, spectrum records, public sector awards, and Commerce Commission filings. The careful wording is therefore: Two Degrees Mobile Limited / Two Degrees Networks Limited, trading as 2degrees, within the 2degrees Group.

The company is now a converged telecommunications group combining mobile, broadband, energy, enterprise, wholesale, and the public sector. It is not simply the third mobile network operator. The 2022 merger with Vocus New Zealand brought it scale in broadband, fixed-line network assets, the legacy of Orcon/Slingshot/CallPlus/FX Networks, and a larger enterprise and government portfolio. Its FY2025 update states that 2degrees completed a three‑year integration of the technology platforms of the Vocus Group and 2degrees, migrating more than two million customer connections across mobile, broadband, and energy to a single in‑house platform named Tahi.

The investment case is therefore mixed. 2degrees has real infrastructure, nationwide brand recognition, a public‑sector anchor through Network for Learning, and sufficient scale to matter in mobile and broadband. But it also faces the economics of a capital‑intensive challenger: 5G rollout, spectrum, reliance on tower lease contracts after the monetisation of passive towers, dependence on wholesale fibre in residential broadband, competition from Spark and One NZ, and pressure on service quality after the integration. Its own FY2025 figures show total revenue of NZ$1.38 billion, mobile revenue of NZ$581.5 million, broadband revenue of NZ$432.3 million, operating EBITDA of around NZ$395 million, capex excluding spectrum of NZ$187 million, and a statutory pre‑tax loss of NZ$22.6 million.

Identity is clear, but the perimeter is stratified

The directory label "TWO-DEGREES-AS-AP" corresponds most directly to AS9790. AS9790 is not the only visible ASN of 2degrees: AS23655 is also associated with 2degrees, but AS23655 uses the APNIC name TWODEGREES-NZ-AS, while AS9790 uses TWO-DEGREES-AS-AP. APNIC data for AS23655 shows descr: 2degrees Networks Limited, organisation ORG-TDML1-AP, and contacts under 2degrees, while the BGP view of AS9790 gives the exact name TWO-DEGREES-AS-AP.

A useful operational clue is PeeringDB. The PeeringDB record for AS23655 states that 2degrees is migrating all public peering to AS9790 and asks networks not to request peering directly with AS23655. AS23655 is still part of the historical network perimeter, but AS9790 is the best focal point for current public interconnection analysis.

This stratified perimeter likely reflects the merger history and operational separation rather than corporate ambiguity. The public brand "2degrees" sits over several legal and network entities. APNIC role data refers to a "Two Degrees Networks Limited administrator", while the APNIC organisation entity is "Two Degrees Mobile Limited". The official peering policy states: "Two Degrees Mobile Ltd (2degrees) is a New Zealand provider of network services to wholesale and enterprise customers in the Asia‑Pacific region".

The canonical research answer is therefore not a single name but a hierarchy. At the top sits the trading brand: 2degrees. For website and consumer‑facing identity, the source is 2degrees.nz. For mobile spectrum, retail mobile, and many official filings, the name is Two Degrees Mobile Limited. For routing and network resource identity, the relevant names are Two Degrees Networks Limited, Two Degrees Mobile Limited, AS9790 TWO-DEGREES-AS-AP, and AS23655 TWODEGREES-NZ-AS. The best shorthand for an intelligence directory is: 2degrees, legally centred on Two Degrees Mobile Limited and Two Degrees Networks Limited, operating AS9790 / TWO-DEGREES-AS-AP as a major telecommunications network in New Zealand.

Ownership: challenger origins, private infrastructure holding, tower monetisation

The current ownership context of 2degrees is best understood through the 2022 merger. The New Zealand Commerce Commission granted clearance in March 2022 for Voyage Digital (NZ) Limited to acquire all the shares of Orcon Holdings Limited, formerly Vocus NZ Holdings, and all the shares of Two Degrees Group from Trilogy International New Zealand LLC and Tesbrit B.V. The Commission treated the transaction as a merger of the Vocus Group and 2degrees and concluded that it was unlikely to substantially lessen competition, partly because Spark and Vodafone, now One NZ, would remain significant competitors.

This clearance is the public turning point. Before it, 2degrees was mainly known as the third mobile challenger, while Vocus New Zealand was a large fixed‑broadband, enterprise, and wholesale group. After, the combined company became a more credible integrated competitor: mobile network, residential broadband, business connectivity, wholesale network services, data centre/cloud services, energy bundles, and public‑sector connectivity. Today, 2degrees is therefore an integrated telecom challenger, not a pure mobile play.

The parent and investor context is one of private infrastructure‑style holding. Macquarie Asset Management lists 2degrees as a full‑service telecommunications provider based in New Zealand in its digital portfolio. Public‑market speculation around the 2021‑2022 deal included discussion of shelved IPO plans and a merger with Orcon/Vocus; those reports were market noise at the time, but the subsequent Commerce Commission clearance corroborated the core transaction.

A second capital event matters as much as the merger: the tower monetisation. In 2023, the Commerce Commission authorised Connexa Limited to acquire certain passive mobile telecommunications infrastructure assets from Two Degrees Networks Limited and Two Degrees Mobile Limited. Accessible transaction reports on that deal described Connexa and the Ontario Teachers' Pension Plan agreeing to acquire 100% of 2degrees' passive mobile tower assets for NZ$1.076 billion from 2degrees Mobile, an entity owned by funds managed by Macquarie Asset Management and Aware Super. The business consequence is simple: 2degrees gained capital flexibility but swapped fully‑owned passive infrastructure for a long‑term lease and service dependency.

This is a classic telecom capital transaction. Selling towers can reduce debt or fund network investment, but it also creates a fixed‑cost relationship with a tower company. For a challenger with a smaller mobile market share than Spark and One NZ, tower monetisation is rational; it recycles capital from passive steel into active network, customer systems, spectrum, and growth. The risk is that future densification, 5G expansion, rural coverage obligations, and site upgrades could be mediated by a third‑party passive infrastructure owner. This does not make the network weak, but it changes the dependency map.

Public excerpts from S&P Global Ratings add a useful analyst signal. S&P affirmed a BB‑ rating for the 2degrees Group in July 2024 and subsequently reported an upgrade to BB in December 2025 following improved results. The public excerpts also indicate that S&P expected Macquarie Asset Management and Aware Super to prioritise deleveraging over shareholder distributions. This is of medium confidence because the accessible S&P pages show only limited text, but it aligns with a private infrastructure owner template: stabilise earnings, integrate assets, manage leverage, and selectively fund capex.

The merger with Vocus changed the cost base before transforming the market

The merger thesis was not simply "mobile plus broadband". It was cost‑system consolidation. 2degrees' FY2025 update states that the company completed the final stage of its three‑year technology integration programme, merging the technology platforms of the Vocus Group and 2degrees and migrating more than two million customer connections to a single in‑house platform, Tahi. The company pitches this as one of New Zealand's largest telecom digital transformations, spanning consumers, SMEs, enterprise, and the public sector.

This matters because telecom competition in New Zealand plays out on thin margins. A third player can only pressure the incumbents if it can keep unit costs low. If 2degrees had kept separate systems forever for Orcon, Slingshot, Vocus, CallPlus, mobile, and energy, it would have inherited complexity without harvesting merger economies. Tahi is therefore not just an IT project; it is a margin project, a retention project, and a cross‑sell project.

The FY2025 financial results show a company with revenue growth and improving operational metrics, but not yet a net statutory profit. Total revenue rose from NZ$1.34 billion to NZ$1.38 billion, mobile revenue grew 4.8% to NZ$581.5 million, broadband rose 3.9% to NZ$432.3 million, and operating EBITDA increased 11.5% to NZ$395.3 million. But the company reported a statutory pre‑tax loss of NZ$22.6 million, which it attributed largely to non‑cash items and one‑off expenses, particularly related to integration.

This leads a sceptical analyst to two simultaneous conclusions. First, the operating business is substantial and cash‑generative: net operating cash flow was reported at NZ$237.3 million and capex excluding spectrum at NZ$187 million. Second, 2degrees is not immune to capital pressures. The growth story depends on maintaining investment discipline while funding 5G, fixed‑network expansion, customer‑platform execution, public‑sector transitions, security services, and product competition.

The company's own strategy acknowledges this competitive posture. It states it will remain a "value‑led operator", will offer better value than the competition, and will lean on a lean, efficient cost base. This is not a premium‑price strategy. It is challenger economics: the company must be cheap enough to win, good enough to retain, and efficient enough not to destroy margins.

Mobile: national mobile network operator, but spectrum and investment continue to constrain the challenger

2degrees is a genuine mobile network operator. Its FY2025 update states that its services are delivered from more than 2,000 cell sites nationally and that it had over 550 active 5G sites. The public "About" page gives a slightly different, broader figure, stating there are 2,225 mobile towers in the network. The difference is not necessarily contradictory, because "cell sites" and "mobile towers" are not always counted identically, but the point is clear: the network is national and physical.

The spectrum history shows why 2degrees remains a challenger rather than an equal incumbent. In the 700 MHz band auction, 2degrees acquired 2x10 MHz for NZ$44 million, while Telecom and Vodafone each acquired 2x15 MHz for NZ$66 million. The government statement indicated that 2x10 MHz was sufficient for a viable 4G network, but the asymmetry matters. Low‑band spectrum favours coverage and in‑building propagation; holding less can increase pressure on site density and capacity management.

For 5G, 2degrees has a more balanced position. Government 5G readiness material recorded an early 3.5 GHz allocation of 60 MHz to 2degrees in 2020 after the COVID‑related auction cancellation, and 2degrees later announced an 80 MHz 3.5 GHz allocation in 2023, describing the outcome as equal allocation among mobile network operators and as investment certainty for the nationwide rollout.

Vendor and technology choices suggest a multi‑vendor network but not a vendor‑neutral one. Ericsson announced the launch of 2degrees' 5G in February 2022 in parts of Auckland, Wellington, and Christchurch, and later disclosed a five‑year deal under which 2degrees would deploy Ericsson radio access network technology across up to 1,200 sites nationally. Ericsson also described work to replace a large part of 2degrees' microwave network with MINI‑LINK 6000 equipment, positioning microwave as a cost‑effective backhaul tool for mobile coverage where fibre is not available or economic.

Nokia reported a separate win in 2024 for 5G core software. It stated that 2degrees had selected Nokia 5G Core Registers and Shared Data Layer software on Red Hat OpenShift, to manage data more cost‑effectively and improve reliability and serviceability for around 1.6 million subscribers. Nokia also noted that 2degrees offered broadband and mobile on 3G, 4G, and 5G networks covering 98.5% of the places where New Zealanders live and work, with a national fibre‑optic network.

The sceptical view is that 2degrees invests like an integrated operator but must do so without the same historical scale as Spark or One NZ. A challenger with less historical low‑band spectrum and a smaller mobile subscriber base must either accept differentiated coverage/performance, spend more per customer, share more infrastructure, rely on smarter radio/backhaul engineering, or price aggressively enough to offset perceived network gaps. The company's 2025 claim that its network is "up there with anything in the market" is a marketing statement; the tangible evidence is more nuanced: the network is extensive, 5G is rolling out, but investment, tower leases, spectrum depth, and rural economics remain real constraints.

Fixed broadband: scale, but last‑mile dependency persists

2degrees' fixed‑broadband business is now substantial. FY2025 broadband revenue was NZ$432.3 million, up 3.9% year on year, and the public website offers fibre plans, wireless plans, Hyperfibre, and rural plans. The company's "About" page claims 4,600 km of underground optical fibre, and the FY2025 update refers to a national fibre‑optic network.

But the economics of fixed broadband in New Zealand is not about owning every last‑mile fibre access line. The ultra‑fast broadband structure in New Zealand separates wholesale fibre access from retail service provision across much of the country. Chorus and the local fibre companies remain the main access providers; retail service providers such as 2degrees, Spark, One NZ, Mercury, and smaller ISPs compete on service, support, price, and bundles. A rural connectivity report summarising the New Zealand broadband market identifies Chorus as the largest wholesale fibre provider and also names local fibre companies such as Enable, Tuatahi, and Northpower, noting that these access providers do not offer retail broadband services.

This structure creates both opportunity and dependency. It allows 2degrees to compete nationally as a retail fibre provider without replicating the full last‑mile fibre build. But it also means 2degrees depends on wholesale access pricing, installation processes, fault coordination, and product roadmaps from Chorus and the local fibre companies. A 2degrees broadband customer may blame 2degrees for a failed install, an ONT issue, or a fibre outage, even though the underlying ground‑level dependency rests with a wholesale access provider. This customer‑experience risk is visible in industry complaint data.

Fixed‑wireless access is a partial escape from wholesale fibre dependency, but not a free one. Fixed wireless uses mobile spectrum and radio capacity, meaning it competes with mobile data traffic for network resources. The economics of rural broadband are even tougher. The rural connectivity report notes that fibre is most cost‑effective in dense areas, while fixed wireless can be more economic in less dense rural areas and satellite suits the most remote premises. It also notes that fibre already covers around 87% of the population and that extending coverage to 95% was estimated by Chorus at about NZ$2‑2.5 billion, with the cost per premise passed rising steeply.

For 2degrees, this implies a segmentation problem. Urban fibre customers are contestable but have margins constrained by wholesale access and retail competition. Rural customers may pay more but are expensive to serve and increasingly contested by Starlink, wireless ISPs, and fixed‑wireless mobile alternatives. Enterprise fibre and managed networks can offer better margins but demand service assurance, security integration, and account‑management capabilities. The Vocus legacy helps here, but it does not remove the need for disciplined capital allocation.

Enterprise and wholesale: the post‑merger strategic centre

The main prize of the Vocus merger is not consumer broadband. It is enterprise, public sector, and wholesale. 2degrees' business website markets mobile, fibre networks, support services, cloud, DDoS protection, Mobile as a Service, Hyperfibre, data centres, and Starlink, to small business, large enterprise, public sector, and secure‑access and collaboration segments.

The public‑sector anchor is Network for Learning. In October 2024, N4L announced that Palo Alto Networks and 2degrees Mobile had been awarded long‑term contracts for the next iteration of its managed network service for schools and kura across New Zealand. N4L stated that 2degrees would provide the internet services and enable N4L to operate as a virtual ISP, while Palo Alto would provide firewall, cybersecurity, and web‑filtering services. N4L also noted that the procurement process started in September 2023 and included an independent probity auditor and a procurement board with Ministry of Education representation.

GETS records support the procurement context. N4L's advance notice stated that existing contracts were expiring in 2026 and that N4L intended to go to tender for nationwide internet services and/or network equipment, with replacement requirements to be confirmed by mid‑2024 to ensure continuity of school internet connections. This is a corroborated procurement signal, not just a vendor press release.

In August 2025, N4L stated that over 750 schools had been upgraded and that all schools were to complete the upgrade by mid‑2026. It further noted that Palo Alto and 2degrees had won long‑term contracts after a competitive procurement process focused on best value for the Crown.

2degrees' own FY2025 update underlines the scale: N4L provides managed internet, Wi‑Fi, cybersecurity, and web filtering to around 2,500 schools and kura, reaching 905,000 principals, teachers, and students, and accounts for nearly a quarter of New Zealand's daytime business internet traffic. The report calls N4L one of New Zealand's largest customers and states that 2degrees was confirmed in October 2024 to deliver the internet connectivity for the upgrade.

This is both a validation and a concentration risk. Winning N4L from an incumbent, reported by Reseller News as displacing Spark as the network services provider, is strong evidence that 2degrees can compete for nationwide public‑sector connectivity. But public‑sector accounts impose reputational and operational obligations. A school network outage is not a normal corporate SLA event; it can become a political and media issue.

Wholesale is also a strategic lever. The company's peering policy describes 2degrees as a provider of network services to wholesale and enterprise customers in the Asia‑Pacific region, and its FY2025 update states that the N4L partnership strengthens the depth and competitiveness of its wholesale business. This aligns with a company using the Vocus/FX Networks legacy to monetise backbone, fibre, and managed network services beyond retail subscribers.

Routing and peering footprint: visible national scale, with AS9790 now central

AS9790 provides the strongest technical proof for the resource record. BGP.tools lists it as an active eyeball network allocated by APNIC, with 62 announced IPv4 prefixes and 5 IPv6 prefixes, upstreams including Hurricane Electric, Vocus Connect International Backbone, NextHop, Lumen, and Solarix, and ranking second in New Zealand for AS‑cone and estimated eyeballs in its model. It labels the AS as residential ISP, business broadband, and mobile data/carrier. These are third‑party routing observations, not audited company disclosures, but they align with the commercial footprint.

IPIP's AS9790 directory gives a compatible but not identical view: AS9790, AS name TWO-DEGREES-AS-AP, organisation name Two Degrees Networks Limited, 62 IPv4 prefixes, 8 IPv6 prefixes, and 681,728 IPv4 addresses. The divergence with BGP.tools on the IPv6 prefix count is normal for dynamic BGP views and different visibility methodologies. The analytical conclusion does not depend on an exact prefix count. It depends on scale and role: AS9790 is a large national access/backbone AS, not a small enterprise stub.

The AS23655 record remains relevant. APNIC lists AS23655 as TWODEGREES-NZ-AS, description 2degrees Networks Limited, and BGP.tools shows it as another active 2degrees network. PeeringDB's instruction not to request direct public peering with AS23655 because public peering is migrating to AS9790 strongly suggests consolidation or an operational migration of public interconnection.

2degrees' official peering policy is selective but route‑server‑friendly. It states that 2degrees openly peers with IX route servers, does not establish settlement‑free peering with current or recent IP transit customers, and requires 24/7 NOC availability, abuse contacts, DDoS security measures, and redundant networks for bilateral IX peering. This is a commercially disciplined peering stance: open enough to reduce domestic exchange costs, selective enough not to cannibalise transit revenue.

BGP.tools' visible relationship list for AS9790 includes a mix of networks such as Network for Learning, Chorus, Auckland Council, Auckland Transport, Health New Zealand, Transpower, Mercury, Fortinet, and others. Routing visibility is not contractual proof. It does, however, reinforce that AS9790 sits in the middle of domestic enterprise, public‑sector, and wholesale traffic flows.

The routing footprint also exposes dependencies. The upstreams include Vocus Connect International Backbone and global/trans‑Tasman networks. This is not surprising for a New Zealand operator: international capacity, submarine cable paths, Australian peering, and global transit all matter. Forum discussions about routing to Australian services, while weak evidence, show how quickly end users perceive international path choices as an ISP quality indicator. A 2021 Geekzone discussion attributed some latency to return‑path routing and peering beyond 2degrees' direct control, a useful reminder that "the ISP" does not control every path on the public internet.

Customer dependency: scale, bundle stickiness, and service complexity

2degrees' customer base is large but must be read in terms of connections, not unique people. The FY2025 update states that more than two million customer connections across mobile, broadband, and energy are now on Tahi. The Nokia release in 2024 referenced around 1.6 million subscribers. These figures are not necessarily inconsistent because a household may represent several service connections, and because mobile, broadband, electricity, and business services are counted differently.

The company's stickiness therefore rests on multi‑product economics for households and SMEs. Mobile‑only customers have lower switching costs, especially when number portability and SIM/eSIM replacement are straightforward. Broadband customers face moderate switching costs tied to installation timing, router setup, email, or bundled billing, but the wholesale fibre structure in New Zealand makes provider switching possible in many areas. Business and public‑sector customers have high switching costs because WAN design, security, IP addressing, service‑level processes, managed Wi‑Fi, firewall, procurement, and migration risk are all material.

2degrees deliberately raises switching costs through bundling and platform integration. The merger created a base spanning mobile, broadband, and energy, and the company's FY2025 update repeatedly highlights a single view of customer connections and operations. The economics are plain: a mobile‑only customer can leave cheaply; a household with broadband, two mobile plans, and electricity is harder to dislodge; a business using WAN, SIP, DDoS protection, a mobile fleet, and managed security is harder still.

The downside is service complexity. 2degrees says it handles more than 10 million customer interactions each year across online, app, email, phone, and stores. The company claims Tahi has reduced in‑store transaction times by an average of 15 minutes and simplified customer operations. These are company claims, but they identify the real operational issue: integration complexity was a customer‑service risk, not just an IT risk.

Price pressure: challenger promise versus sector concentration

2degrees' founding economic role was to discipline a duopoly. Its FY2025 update explicitly invokes the "2degrees effect" and claims the company helped end the era of the 20‑cent text message. The company still positions itself as a value champion and states it will offer better value than the competition through a lean cost base.

But New Zealand's mobile market remains concentrated. Bill Bennett's July 2025 analysis of the Commerce Commission's 2024 telecommunications monitoring report notes that Spark, One NZ, and 2degrees collectively controlled 97.5% of the mobile market and that the report described mobile as a stable three‑player oligopoly with low MVNO penetration relative to OECD benchmarks. In urban broadband, the same three leaders held a reported 73% share, although broadband is seeing more disruption from wholesale fibre and energy‑provider bundles. This is an analyst summary of a regulator report, not a primary market‑share filing, but it points in a direction consistent with market structure.

The price question is therefore not whether 2degrees competes. It does. The question is whether the market exerts enough pressure below the top three. MVNOs remain small in mobile. Skinny is Spark's sub‑brand, Kogan uses One NZ, and Warehouse Mobile appeared in TDR data under 2degrees' parent reporting group. Fixed broadband is more open because Chorus and local fibre companies are wholesale only, but that also means retail broadband providers have less scope to differentiate the underlying access product.

This creates an uncomfortable equilibrium. 2degrees must keep prices low enough to preserve challenger credibility, but high enough to fund 5G, customer systems, enterprise SLAs, and debt service. Spark and One NZ can retaliate with bundles, sub‑brands, device financing, fixed wireless, and loyalty offers. Chorus and local fibre companies can influence the retail cost structure through regulated wholesale products and pricing. Starlink and wireless ISPs add rural pressure. MVNOs and energy providers exert niche pressure. The result is not a free‑for‑all; it is a concentrated market with selective price wars.

Competition: Spark, One NZ, Chorus, local fibre companies, and MVNOs each pressure a different margin

Against Spark, 2degrees competes with a national incumbent that has brand strength, scale in both mobile and broadband, enterprise relationships, Skinny as a fighting brand, and a large fixed‑wireless customer base. Spark can pressure 2degrees on mobile price, SME bundles, large‑enterprise procurement, and customer perception of network quality.

Against One NZ, 2degrees faces another national mobile network operator with Vodafone legacy, spectrum, retail stores, enterprise relationships, and satellite‑to‑mobile marketing. The One NZ rebranding created some opening for customer churn, but it remains a sizeable competitor. The May 2026 outage report affecting both One NZ and 2degrees in parts of both islands is also a reminder that mobile networks can share infrastructure dependencies or common external failure modes, even when they compete.

Against Chorus and the local fibre companies, the relationship is not simple competition. Chorus, Enable, Tuatahi, Northpower, and other access owners provide wholesale infrastructure on which retail service providers depend. They are suppliers, bottlenecks, and ecosystem partners. They do not generally compete as retail ISPs, but they influence product quality, fault resolution, and costs.

Against MVNOs, 2degrees has a dual role. It is a mobile network operator that could sell wholesale capacity, but it also competes with low‑cost mobile offerings. The Commerce Commission's 2022 merger analysis specifically examined the vertical integration of 2degrees' mobile network with Vocus, then described as the largest MVNO, and concluded the transaction was unlikely to materially change incentives to grant MVNO access, given competition from Spark and Vodafone.

Against rural alternatives, the competitive landscape is shifting. Fixed wireless, satellite, and wireless ISPs compete where fibre is not available, expensive, or slow to install. The rural connectivity report notes that fibre is nationally dominant but that fixed wireless and satellite have advanced, with Starlink being the leading current LEO fixed‑satellite broadband provider at the time of the report. This limits the ability of 2degrees to treat rural broadband as a protected high‑margin niche.

Reliability and complaint record: not catastrophic, but service pressure is visible

The public evidence does not support a claim that 2degrees has particularly poor reliability across its services. It does support a more cautious statement: 2degrees experiences visible pressure on service quality, particularly around broadband complaints and integration‑era customer operations.

The TDR half‑yearly report for July–December 2023 shows that 2degrees' mobile complaints per 10,000 connections were 0.67 in Q1 and 0.52 in Q2, higher than Spark and One NZ in those periods. For broadband, 2degrees' complaint rate was much higher: 7.21 per 10,000 connections in Q1 and 5.57 in Q2, versus 2.47 and 1.48 for One NZ, 1.95 and 1.43 for Spark, and 1.13 and 0.80 for Mercury. TDR warns that complaint volume alone does not represent performance and that media issues, service migrations, or promotion of the dispute‑resolution process can affect complaint volumes. This caveat is important because the Vocus/2degrees integration was underway.

Officially, 2degrees maintains a mobile and broadband network status page and notes that networks can be affected by planned maintenance or unexpected outages. The page details that managed‑service business customers receive advance email notice of scheduled maintenance. These are normal telecom hygiene practices, not proof of superior reliability.

There are also recent outage signals. Newstalk ZB reported on 1 May 2026 that One NZ mobile and broadband outages were affecting parts of the lower North Island, the South Island, and Stewart Island, and that 2degrees' mobile and wireless broadband customers were also affected by coverage issues across multiple South Island and lower North Island sites, according to 2degrees' network status page. The report also noted that some customers were having top‑up and purchase issues. Because One NZ was also affected, the episode looks more like a shared regional failure mode or external dependency than a 2degrees‑specific network outage.

Weak customer signals are negative but noisy. Geekzone users discussed a voice outage in March 2026 and app/billing visibility issues, suggesting some perceived core‑system impact. Reddit and Trustpilot contain complaints about routing, customer service, broadband downtime, and billing and service frustrations, but these are self‑selected reviews. Trustpilot itself states it does not fact‑check and that the company has not solicited reviews, so the sample is more useful for identifying friction themes than for measuring service quality.

The harder regulatory issue is marketing accuracy. In May 2024, the Commerce Commission filed eight charges under the Fair Trading Act against Two Degrees Mobile Limited over allegedly misleading "free business roaming in Australia" claims. In April 2025, Reseller News reported that 2degrees was fined $325,000 after admitting five Fair Trading Act breaches. The underlying issue was that the "free" roaming was capped at 90 days per year, after which daily charges applied; 2degrees removed the cap, refunded affected customers, and updated promotional material. This is not proof of network reliability, but it is relevant to trust and business‑customer risk.

Unofficial signals: what is useful and what is just noise

Corroborated signal: Bill Bennett's 2021 reporting that 2degrees and Orcon/Vocus were in merger talks and that IPO plans were shelved was market noise at the time, but the subsequent Commerce Commission clearance confirms the transaction direction. The same source correctly framed the merger as shifting 2degrees from a mobile challenger to a full telecom operator with broadband, enterprise, government, cloud/data centre, and bundle capabilities.

Medium‑confidence analyst signal: the S&P rating excerpts indicate that the company's credit profile improved after integration and earnings growth, but also that deleveraging remains central. The accessible pages provide limited detail, so the signal is useful for direction rather than for a full independent reconstruction of debt capacity.

Corroborated procurement signal: the N4L/GETS records and N4L's own update show a competitive procurement leading to 2degrees' role in school connectivity. Reseller News adds that Spark was displaced as the incumbent network services provider. This is stronger than ordinary vendor marketing, because procurement and implementation records exist.

Weak operational signal: routing complaints on Geekzone and Reddit identify user pain points around international routing, return paths, CGNAT, billing/app systems, and voice outage perceptions. These are leads for further testing rather than performance statistics.

Weak customer‑service signal: Trustpilot sentiment is markedly negative, but self‑selection bias is severe. The useful point is not the rating itself; it is the recurring theme: friction around customer service, communications, billing, and resolution. TDR data are stronger and point in a similar direction for broadband complaints in 2023.

Labour‑market signal: 2degrees' recruitment material mentions roles in data centre engineering, network engineering, managed network engineering, managed voice engineering, and a 24/7 security and network operations centre. This is a weak but relevant signal that the company runs a substantial in‑house technical estate rather than a purely outsourced retail brand. It proves neither hiring volume nor project timing.

Economic assessment: more credible, not invulnerable

2degrees has crossed the threshold from "mobile challenger" to "integrated national telecom operator". The AS9790/TWO-DEGREES-AS-AP record, the relationship with AS23655, the official peering policy, the BGP footprint, the mobile spectrum registration, the 5G vendor contracts, the fixed‑broadband revenue base, the public‑sector win with N4L, and the Vocus integration all point in the same direction.

The company's strategic opportunity is to be the only credible integrated third alternative to Spark and One NZ. This gives it a valuable procurement role. Enterprises and public agencies often prefer not to be locked into a two‑supplier market. 2degrees can exploit that preference if it can demonstrate price discipline, service reliability, technical capability, and national coverage.

The constraint is capital. Mobile networks need spectrum, radios, towers, backhaul, core software, and operations. Fixed broadband needs wholesale access, backbone, routers, customer premises equipment, provisioning systems, and support. Enterprise demands redundancy, security, account management, and SLA credibility. The public sector requires implementation discipline and political resilience. The FY2025 figures show scale, but also the burden: capex remains sizeable, statutory profitability is not yet clean, and private infrastructure ownership will likely prioritise deleveraging and disciplined investment.

2degrees' dependency map is therefore wider than the brand suggests. It depends on Chorus and local fibre companies for a large part of the residential fibre access layer. It depends on Connexa for passive mobile tower infrastructure after the tower sale. It depends on Ericsson and Nokia for significant mobile RAN/core domains. It depends on international and domestic peers and transit for user experience beyond New Zealand. It depends on Tahi working as a unified platform rather than becoming a single point of operational fragility. It depends on executing N4L well so it becomes a reference account, not a source of public‑sector embarrassment.

The competitive conclusion is sceptical but constructive. 2degrees is not a weak challenger. It is a sizeable, converged telecom operator with a genuine routing footprint, a national mobile network, a credible broadband base, and meaningful public‑sector traction. But its challenger economics are relentless. Its value promise needs lower unit costs; its enterprise ambition needs stronger service assurance; its 5G ambition requires capex; and its customer complaints show that the integration burden was not invisible to users.

Evidence register

Routing identity AS9790 / TWO-DEGREES-AS-AP — high confidence. BGP.tools lists AS9790 as Two Degrees Networks Limited, website 2degrees.nz, active under APNIC, with as-name: TWO-DEGREES-AS-AP.
2degrees identity associated with AS23655 — high confidence. APNIC lists AS23655 as TWODEGREES-NZ-AS, description 2degrees Networks Limited, organisation ORG-TDML1-AP; it appears to be a related 2degrees ASN rather than the exact directory label.
Public brand and legal website identity — high confidence. The 2degrees website describes the company as a full telecom operator and uses "© Two Degrees Mobile Limited 2026" in its footer.
2degrees scale claims — company‑disclosed, medium‑high confidence. The official site claims 1,520 employees in New Zealand, 4,600 km of fibre, and 2,225 mobile towers.
Vocus/2degrees merger — high confidence. Commerce Commission clearance documents identify the acquisition of Orcon/Vocus and Two Degrees Group by Voyage Digital and the Commission's competitive assessment.
Tower sale to Connexa — high confidence for clearance, medium‑high for commercial value. The Commerce Commission authorised Connexa's acquisition of passive mobile infrastructure assets; transaction reports described consideration of NZ$1.076 billion.
FY2025 financial results and integration — company‑disclosed, medium‑high confidence. 2degrees reported total revenue of NZ$1.38 billion, operating EBITDA of NZ$395.3 million, capex excluding spectrum of NZ$187 million, and completion of the Tahi migration.
Mobile spectrum — high confidence. Government/RSM records show allocation asymmetry in the 700 MHz band and 5G allocations in the 3.5 GHz band.
5G vendor roadmap — high confidence for vendor announcements. Ericsson announced the 5G launch, RAN modernisation, and microwave upgrades; Nokia announced 5G core software.
Broadband market structure — medium‑high confidence. The rural connectivity report identifies the roles of Chorus and local fibre companies and explains the economics of fibre, fixed wireless, and satellite.
N4L public‑sector contract — high confidence. N4L, GETS, 2degrees, and trade press all corroborate a competitive procurement and 2degrees' role as the internet service provider enabling N4L to operate as a virtual ISP.
Peering policy and public peering migration — high confidence. 2degrees' peering policy sets selective rules; PeeringDB indicates that public peering is migrating to AS9790.
Market concentration — medium confidence. Bill Bennett summarises the Commerce Commission's 2024 monitoring findings: the top three telcos control 97.5% of mobile and 73% of urban broadband. Use as an analyst summary unless the primary report is reviewed separately.
Customer complaints — high confidence for TDR numbers, medium for interpretation. TDR shows elevated broadband complaints for 2degrees in July–December 2023, but warns that complaint volumes alone do not represent member performance.
Outage track record — medium‑high confidence for reported incidents. 2degrees' own network status page acknowledges planned and unplanned outages; Newstalk ZB reported a disruption in May 2026 affecting both 2degrees and One NZ customers across wide areas.
Marketing compliance issue — high confidence. The Commerce Commission laid charges over "free business roaming in Australia" claims; Reseller News reported the subsequent $325,000 fine and guilty pleas.
Forum/review‑site signals — low confidence. Geekzone, Reddit, and Trustpilot surface recurring complaints about routing, service, and outages, but samples are self‑selected and work best as leads rather than statistical evidence.
Analyst credit/ownership signal — medium confidence. S&P excerpts indicate rating affirmation/upgrade and owner priority on deleveraging, but the accessible text is limited.

Watch points for the 12‑36 month horizon

AS9790 consolidation. Watch whether public peering fully migrates from AS23655 to AS9790 and whether legacy Vocus/Orcon/Snap ASNs are consolidated or kept for segmentation purposes.
N4L execution through mid‑2026. The school network migration is the most visible public‑sector proof point. Successful completion builds 2degrees' enterprise and government credibility; delays or outages would damage the challenger narrative.
Capex vs. leverage. FY2025 capex excluding spectrum was NZ$187 million. The key question is whether owners will prioritise deleveraging over 5G density, enterprise resilience, or fibre/backbone upgrades.
5G deployment beyond 550 sites. Ericsson's RAN deal for up to 1,200 sites creates a measurable benchmark. Coverage, indoor performance, and fixed‑wireless capacity need tracking against Spark and One NZ.
Tower lease economics. After the Connexa sale, site adds, upgrades, and densification will reveal whether passive infrastructure reliance limits speed or economics.
Spectrum adequacy. 2degrees' 80 MHz of 3.5 GHz supports 5G competitiveness, but lower historical holdings in 700 MHz still matter for coverage and indoor performance. Watch any future spectrum swaps, renewals, or regulatory allocations.
Broadband complaint trend post‑Tahi. Post‑integration TDR data should show whether the high 2023 complaints were migration noise or a durable service problem.
Enterprise and public‑sector wins and losses. N4L is a reference account. Watch local government, education, health, transport, managed security, and WAN tenders to see whether 2degrees displaces Spark/One NZ beyond price‑led bids.
MVNO positioning. The Commerce Commission accepted that the merger would not materially change MVNO incentives. Watch whether 2degrees opens up more to mobile wholesale deals or keeps capacity for its own retail/bundling strategy.
Chorus/LFC wholesale changes. Regulated fibre pricing, Hyperfibre take‑up, and LFC roadmaps will shape 2degrees' fixed margins and ability to differentiate.
Rural substitution. Starlink, wireless ISPs, and fixed wireless will pressure rural broadband. 2degrees must decide where to use mobile capacity for fixed wireless and where to avoid low‑return rural competition.
Customer trust after the roaming case. The $325,000 fine is not financially material for a multibillion‑dollar telco, but it is reputationally relevant to business customers. Watch whether similar complaints about headline offers emerge.
International routing quality. AS9790's upstream mix and peering policy should be monitored for latency changes to Australia, the US, and Asia. Forum complaints are weak evidence, but repeated patterns can indicate cost‑oriented path choices.
Satellite strategy. The AST SpaceMobile deal is a counter to One NZ's satellite messaging/coverage positioning. The question is whether it becomes a differentiated service or remains a defensive marketing point.
Energy bundling. 2degrees' energy revenue growth gives it a household bundle weapon. Watch whether energy improves retention or distracts management from telecom execution.
Owner exit optionality. No credible sale or IPO rumours were confirmed in the sources reviewed. Nonetheless, the private infrastructure ownership model implies eventual exit optionality. Watch leverage, rating trajectory, EBITDA margin, and tower‑lease stability for signs of IPO or sale readiness.