VNO launch plans are written around data bundles, eSIM onboarding, and customer-acquisition cost. Messaging usually appears as a single line — 'SMS: included' — inherited from whatever the host-MNO agreement says. That treatment hides two facts. First, messaging is a launch dependency: the MVNO's own KYC and onboarding flows need OTP delivery working before the first subscriber activates, and the porting, billing, and service-notification flows all ride on it. Second, messaging is a revenue line: from the day the subscriber base exists, enterprises will pay to terminate A2P traffic to it, and that revenue either accrues to the MVNO or silently to someone else in the chain. The build-versus-partner decision deserves the same rigour as the core network one — and it mostly resolves differently.
What an MVNO actually needs on day one
- Outbound OTP and transactional messaging for its own flows — activation, SIM-swap confirmation, top-up receipts, eSIM install links. This must work internationally, because subscribers activate while travelling.
- P2P SMS interworking for subscribers — usually inherited from the host MNO on thin and light models, owned on full MVNO models.
- An SMSC, or access to one — full MVNOs terminating their own traffic need message-centre capability; lighter models consume the host's.
- Sender-ID and short-code capability for customer care and marketing, registered in every regulated market the MVNO addresses.
- A2P termination monetisation — the machinery to bill enterprises for reaching the subscriber base, plus the firewall to stop the traffic arriving unbilled.
- Fraud controls — inherited from the host on light models, but the commercial exposure (AIT, spam complaints, regulatory findings) lands on the brand either way.
What building it actually costs
A full owned messaging stack means an SMSC (licensed software or carrier-grade cloud deployment), SMPP interconnects to the host and to international partners, a firewall in front of the message centre, monitoring, and the regulatory registrations. The realistic budget conversation: a seven-figure first-year cost between licences, integration, and infrastructure; 12–18 months from contract to production-grade; and a permanent team of three to five engineers who understand SMPP bind lifecycles, SS7 or Diameter interworking, and the fraud landscape. That team is the expensive part — messaging-core engineers are scarce, and an under-operated SMSC fails in ways that take subscriber-visible services down with it. For an MVNO whose differentiation is brand, pricing, or a niche audience, none of this is differentiating spend.
What partnering looks like
The partner model consumes messaging as a service: a hosted SMSC gateway, OTP and transactional APIs for the MVNO's own flows, sender-ID registration handled per market, and — the part most MVNOs discover late — an A2P monetisation arrangement where the partner terminates enterprise traffic to the subscriber base and shares the revenue. Time to production is weeks rather than quarters; the cost model is per-message or per-subscriber opex with no standing engineering commitment. The trade-offs are real but bounded: a margin share on A2P revenue, a dependency on the partner's roadmap, and the need for contractual exit terms (number-range portability, traffic-history export, notice periods) negotiated before signature rather than after.
The decision, component by component
| Component | Build makes sense when | Partner makes sense when |
|---|---|---|
| SMSC / message centre | Full MVNO, multi-million subscriber ambition, in-house telco engineering | Thin or light MVNO, sub-500k subscribers, lean team |
| OTP for own flows | Almost never — undifferentiated and latency-critical | Nearly always; buy it as an API from day one |
| SMS firewall | Owned SMSC and appetite to operate detection models | Every other case — bundled with hosted SMSC |
| A2P monetisation | Base above ~1M subscribers and commercial team to run it | Smaller bases — revenue-share with a monetisation partner |
| RCS / rich messaging | Rarely at launch on any model | Defer, then partner when enterprise demand appears |
The thresholds that actually decide it
Across MVNO launches the pattern is consistent. Below roughly half a million subscribers, the owned stack never pays back: the A2P revenue base is too small to fund the team, and the engineering budget is better spent on the subscriber-facing product. Between half a million and two million subscribers, hybrid models appear — owned sender-ID strategy and commercial relationships, partnered infrastructure. Above two million, and especially where the MVNO is really an MVNE hosting other brands, owning the message centre starts to make sense because the volume amortises the team and the margin share saved exceeds the operating cost. The mistake to avoid is sequencing: MVNOs that build first and shrink later write off the investment; those that partner first and insource later carry no stranded cost.
The revenue line people forget to negotiate
Even a small subscriber base terminates A2P traffic — banks, delivery platforms, and government services will message those subscribers whoever runs the network. Industry-typical A2P wholesale yield works out to roughly $0.50–2.00 per subscriber per year depending on market and enforcement maturity. On a 300k-subscriber MVNO that is a meaningful contribution line, and on launch day it is leaking by default: without an explicit agreement, the termination revenue sits with the host MNO or disappears into grey routes. Put A2P revenue share on the term sheet — with the host, or with a messaging partner who brings the firewall and monetisation machinery to enforce it.
BroadNet provides hosted SMSC, OTP APIs, sender-ID registration, and A2P monetisation with revenue share for MVNOs and MVNEs — one agreement instead of five builds. Start at the operator solutions overview or the SMSC gateway page.