An MVNO earns on the spread between wholesale capacity and retail bundles, then widens it with add-ons, device and content bundles, and the value the parent business gets back. Airtime margin on its own rarely covers acquisition and care. Capitec Connect reported net income of R442 million for its 2026 financial year on about 1.5 million active subscribers, and Tesco Mobile ended 2024 with 5.7 million customers and £1.11 billion in revenue. Neither owns a tower or spectrum.
Capitec Connect, the mobile arm of a South African bank, reported net income of R442 million for its 2026 financial year, up 129 per cent, on about 1.5 million three-month active subscribers. Tesco Mobile in the UK ended 2024 with 5.7 million customers and revenue of £1.11 billion. Neither owns a tower or a spectrum licence. Both buy capacity from a host operator, sell it under their own brand and keep the difference, plus whatever the mobile line does for the rest of the business. That is the MVNO business model in one sentence. The two companies make their money in different places, though, and that difference is what a business plan has to get right.
The market an MVNO business plan sits in
The category is large and still growing. Juniper Research estimates 333 million MVNO subscribers globally in 2026, rising to 438 million by 2030, with subscriber revenue climbing from $47 billion to more than $54 billion over the same period. GSMA Intelligence counted 2,138 MVNOs worldwide in August 2025, of which 265 were sub-brands of network operators and 96 were Full MVNOs with their own core elements. The great majority run on a host’s or an enabler’s infrastructure with their own brand and commercial logic on top.
Who is entering tells you where the money is. Omdia’s Matthew Reed put it plainly in February 2025: “Banks, supermarkets, and broadband and TV service providers continue to be key players.” In Mexico, regulator data reported by TeleSemana in August 2025 gave MVNOs 15.9 per cent of the mobile market, ahead of Movistar, with Walmart’s Bait alone at around 20 million lines. Mature markets show the same pattern at a slower pace: CCS Insight, cited by TM Forum in September 2025, put MVNOs at 19.7 per cent of UK connections at the end of 2024, up from 14.9 per cent in 2021, with MVNO customer numbers growing 9.8 per cent in a year when the whole market grew 0.7 per cent. The lesson is not that MVNOs win on price. Organisations with an existing customer base and a reason to talk to it every month are taking share from operators that have neither.
MVNO revenue streams
The airtime margin is the base layer. An MVNO buys minutes, messages and gigabytes wholesale, or a per-subscriber allowance, and sells bundles at retail. The spread is gross margin, and it is thin for a plain bundle because the customer can compare it with the host’s own price in thirty seconds. Every other revenue stream exists because airtime margin alone rarely pays for acquisition and care.
Bundles and add-ons widen the spread. Data boosters, international calling packs, content inclusions, device instalments and insurance carry higher margin than the base plan, and a well-designed catalogue moves customers up a tier without a price war. Roaming and travel connectivity have become a category of their own: Juniper Research put travel eSIM revenue at $1.8 billion in 2025, up 85 per cent in a year, and Airalo reported passing 30 million users in June 2026. For a conventional MVNO, sponsored roaming and multi-IMSI make it possible to sell travel bundles at a margin instead of passing through the host’s rates.
B2B and M2M lines pay differently. A corporate account brings many lines under one contract, lower churn, and billing features the consumer market does not need: cost centres, limits by employee, pooled allowances for devices. Sponsored billing, where a company covers an agreed share of an employee’s usage and the individual pays the rest, is a revenue model in itself, because the employer becomes a payer with a budget rather than a channel. Brilliantel in South Africa runs this model for B2B and B2G customers. On the machine side, Kaleido Intelligence counted 290 million IoT connections managed by MVNOs in 2025 and expects 600 million by 2031: low ARPU per SIM, long lifetimes, almost no care cost.
Partner and loyalty economics are where banks and retailers earn most of their return, often not on the telecom P&L at all. A data reward for card spend, a plan included in a premium account tier, or a gigabyte for a receipt scanned in the app costs the sponsor the wholesale price of the data and buys retention, transaction volume and a daily reason to open the app. Nubank’s NuCel reached one million customers in Brazil by June 2026. Rakuten’s mobile arm passed ten million subscribers in five years and eight months from its April 2020 launch, inside a group that sells everything from books to insurance. In these models the mobile line can run at a modest margin and still be one of the group’s best-performing products.
Wholesale itself is the revenue stream for MVNEs and MVNAs. An enabler charges tenants a set-up fee plus a per-subscriber or per-transaction platform fee. An aggregator buys capacity in bulk under one host agreement and re-wholesales it to smaller brands at tiered rates, keeping the spread. Both need multi-tenant billing and per-tenant settlement; the roles are set out in MVNO vs MVNE vs MVNA.
The cost side: what eats the margin
Wholesale rates are the largest line once the base passes a few tens of thousands of subscribers, and they are set by negotiation, volume commitments and how much the host wants your segment. A per-gigabyte rate with a minimum commitment exposes you to under-use; a per-subscriber allowance exposes you to heavy users. The wholesale model you sign decides which retail plans you can sell profitably, so the person who will design the catalogue belongs in the room before the term sheet is agreed.
The BSS is the second line, and its shape depends on the operating model: a per-subscriber fee inside the enabler’s price, or a licence or subscription plus implementation and a small operations team on your own platform. The breakdown by model is in What it costs to launch an MVNO.
Acquisition cost is what kills most consumer MVNOs that fail. A greenfield brand pays for every customer through advertising and channel commissions, then waits months to recover it from a thin airtime margin. A bank or retailer with an app that customers already open, and a loyalty programme that already has a budget, acquires for a fraction of that. This is the structural reason the entrants named by Omdia are the ones that last.
Care and collections are the fourth line and the easiest to underestimate. Every failed port, unexplained charge or disputed roaming bill becomes a contact, and contacts cost money. Prepaid and app-first models reduce collections risk and contact volume; postpaid B2B increases both but is paid for by higher ARPU and lower churn.
Seven MVNO business-model archetypes
Most MVNOs fit one of the archetypes below, and each has a different primary money-maker. Deciding which one you are is the first line of the business plan.
| Archetype | Example | Primary money-maker | What the billing has to do |
|---|---|---|---|
| Flanker / sub-brand of an MNO | giffgaff (Virgin Media O2, online-only since 2009); 265 sub-brands per GSMA Intelligence | Segment defence: keep price-sensitive customers on the parent’s network at lower cost to serve | Low-touch self-service, simple plan ladder, cheap care |
| Retail loyalty MVNO | Tesco Mobile (5.7M customers, £1.11bn revenue); Walmart’s Bait in Mexico (~20M lines) | Airtime margin at scale plus basket uplift and loyalty retention | Receipt or loyalty events converted into data or credit under capped rules |
| Bank / fintech MVNO | Capitec Connect (R442m net income FY2026); Nubank’s NuCel (1M customers) | Retention and engagement in the core business, lower SMS cost, a profitable telecom line as a bonus | Balance-based and account-linked rewards, real-time integration with the banking app |
| Enterprise / ICT provider MVNO | Brilliantel (South Africa, B2B and B2G) | Converged contracts, sponsored billing, lower churn than consumer | Cost centres, employee limits, sponsored splits, postpaid with adjustments |
| IoT / M2M MVNO | A segment of 290M connections in 2025 (Kaleido Intelligence) | Volume of low-ARPU SIMs with very long lifetimes | Pooled allowances, lifecycle states, multi-IMSI, threshold-based spend controls |
| Travel eSIM brand | Airalo (30M users) | Margin on short-term data plans sold to travellers, no legacy base to protect | Instant eSIM issuance, multi-country rating, app-only payments |
| MVNE / MVNA | Enablers and aggregators hosting the brands above | Platform fees per tenant, or the wholesale spread on re-sold capacity | Multi-tenant catalogues and charging, per-tenant settlement, tiered wholesale rating |
Two observations. Only the flanker and the travel eSIM brand make most of their money on the telecom P&L alone; the others earn a large share of the return somewhere else, and the plan has to measure it there. And the right-hand column is what separates archetypes in practice. A retailer and a bank can buy the same wholesale capacity from the same host; what makes them different businesses is the rules that turn a purchase or a balance into a mobile benefit, and where those rules live.
Why the operating model changes the P&L
The same archetype can run as a Light MVNO on an enabler or as a Medium MVNO with its own BSS, and the two produce different profit and loss statements. The ownership split is on the operating models page; this is what it does to the numbers.
Light on an enabler converts fixed cost into a per-subscriber fee. Nothing is cheaper at 20,000 subscribers, and for a standard bundle under a known brand it can stay the cheaper option indefinitely. The limit is the enabler’s shelf: the plans, rewards and integrations it offers are the ones you can sell, and a change to pricing logic waits for its roadmap. If the money-maker is airtime margin plus brand, that is an acceptable trade.
Own BSS turns the fee into a platform you pay for once and a team you pay for continuously, and gives you the catalogue, the charging engine and the subscriber data. That matters when the money-maker is a rule rather than a bundle: a gigabyte for on-time repayment, a plan inside a paid loyalty tier, a sponsored split by employee. Those rules are configuration in your own product catalogue and a real-time decision in your own charging engine. On an enabler they are a feature request. Direct wholesale terms with the host also lower unit cost as volumes grow, and that margin lands with you.
Light on an enabler is right for most consumer brands in year one. Own BSS pays off when pricing logic is the product, or when subscriber data has to sit inside your own systems for regulatory or strategic reasons. The break-even between the two depends on ARPU, wholesale rates and churn, and there is no universal number; the mechanics are worked through in MVNO unit economics and billing architecture.
What a credible MVNO business plan contains
A plan that survives contact with a host operator and a board has six parts: the archetype in one sentence and the money-maker it implies; the wholesale model you will ask for and the retail plans it supports at margin; acquisition cost by channel, with existing-base channels separated from paid ones; care and collections cost per subscriber by payment model; the operating model with its fixed and variable cost, and the trigger that would move you from Light to own BSS; and the non-telecom return, measured where it occurs. What you cannot know before launch is churn and heavy-user behaviour on your own base. The plan should say so and carry a range rather than a point.
Where Avante fits
Avante supplies the BSS an MVNO, MVNE or MVNA runs in the Medium and Full models: product catalogue, online charging, convergent billing, CRM, mediation, provisioning and partner management. It does not run a network or resell airtime. The lever it gives a business model is billing and charging logic: balance-based and account-linked rewards for banks, sponsored billing with cost centres and employee limits for enterprise providers, loyalty-linked data and receipt-to-reward rules for retailers, and multi-tenant settlement for enablers. Delivery is on-premises, in private cloud or as a managed service, with an OPEX-based option and year one as a managed service where no BSS team exists. Sector detail is on the pages for banks and for retail.
Frequently asked questions
How does an MVNO make money?
By buying capacity wholesale from a host operator and selling it at retail under its own brand, then adding higher-margin add-ons, roaming, B2B contracts and partner or loyalty economics on top. For banks and retailers, a large share of the return is retention and engagement in the core business rather than telecom margin. MVNEs and MVNAs earn platform fees and the wholesale spread instead.
Is an MVNO profitable?
It can be. Capitec Connect reported R442 million in net income for its 2026 financial year, up 129 per cent, and Tesco Mobile reported £1.11 billion in revenue for 2024. Profitability depends on acquisition cost more than on wholesale rates: organisations with an existing base and an app acquire cheaply, greenfield consumer brands do not. Juniper Research expects global MVNO subscriber revenue to exceed $54 billion by 2030.
How much do MVNOs pay the host operator?
Wholesale rates are negotiated and confidential, so no reliable public figure exists. The structure matters as much as the rate: a per-gigabyte price with a minimum commitment, a per-subscriber allowance and a revenue share behave differently as the base grows. Once past a few tens of thousands of subscribers, wholesale is usually the largest cost line.
Does owning the BSS change how much an MVNO earns?
It changes where the margin can come from. On an enabler, revenue is limited to the plans and rewards on the enabler’s shelf, and the platform cost is a per-subscriber fee. With an own BSS, pricing rules such as balance-based rewards, sponsored billing or loyalty-linked data become configuration, and direct wholesale terms lower unit cost as volumes rise. Neither is cheaper in every case; it depends on whether pricing logic is the product.