The market that stopped growing in 2017

What this paper is

Two companies sell telecommunications in the Maldives. Both are listed, both file quarterly, and a regulator has published monthly subscription statistics since 2012. That is an unusually good public record for a market this small — and it is still not enough to answer most of the questions a commercial team would ask.

This paper sets out what the record does support: the size and shape of the market, how it has moved over fifteen years, where the two operators’ revenue comes from, and the one number that matters most and is falling. It is deliberately explicit about the boundary between what is measured and what is inferred, because in a market with no published churn, no product volumes and no usage data, that boundary is where most telecom analysis quietly goes wrong.

The interactive version of this work — the dashboards, the product explorer and the commercial models — sits at the telecom programme.

The market has not recovered its peak

The Maldives mobile market grew steadily from 2012 to a peak of 907,763 subscriptions in September 2017 [OFFICIAL]. It then fell for four years, to 692,853 by July 2022 — a 24% contraction — and has been recovering since. At May 2026 it stands at 839,615, still 7.5% below the 2017 peak [OWN-CALC].

Two things drove the fall, and they are separable in the monthly data.

The 2018–19 decline was prepaid substitution, not an event. Prepaid subscriptions fell in almost every month while postpaid rose in almost every month; the total drifted down by roughly half a percent a month with no single step. That is the signature of customers moving between products, not of a base clean-up. A regulatory explanation is sometimes offered for this period, but the SIM registration regulation that caps numbers per person came into force in August 2023 [OFFICIAL — CAM], five years later, and no step change is visible in the series at that date either.

The 2020 fall was the border closure. It is abrupt, it coincides with the collapse in arrivals, and it removed both visitor SIMs and a large part of the migrant-worker base that carries them.

Neither is a data error, and neither is silently fitted over in the forecasting work attached to this programme.

Prepaid stopped falling, and that matters

Prepaid’s share of the mobile base fell from 88% in 2016 to 71% in 2022 and has been flat since — 72.0% at May 2026 [OFFICIAL]. Postpaid has more than tripled since 2012, from 71,463 to 235,067 subscriptions, and is still growing. But prepaid is growing too, at a similar rate, which is why the share has stopped moving.

The commercial reading is that the Maldives is not a market where prepaid is a legacy segment being run down. Seven in ten subscriptions are prepaid and will remain so. Any commercial strategy for this market is, first and mostly, a prepaid strategy — which is why the product and pricing work in this programme is built around prepaid economics rather than treating it as a tail.

One fixed product is dying; another is replacing it

Fixed telephone lines have fallen from 25,699 in 2012 to 10,992 — a 57% decline, running steadily through the whole period [OFFICIAL]. Over the same fifteen years fixed broadband connections rose from 16,304 to 106,611, a 6.5-fold increase.

This is the clearest substitution in the Maldivian record: the copper voice line is being replaced by a household internet connection, and the household is paying more for the replacement than it did for the original.

One definitional trap sits inside these series and is worth naming, because it changes a headline by 4%. CAM’s “Broadband (Mobile)” figure is not mobile-only: across all 101 months where the components are published, it equals data-with-voice plus data-only plus LTE fixed broadband — a household router product — exactly, with no residual [OWN-CALC]. At December 2025 that LTE component was 13,284 of 342,800. Reading the headline as mobile data users overstates them by that amount.

Two operators of similar size, built differently

Dhiraagu is the larger business by revenue in every quarter of the observed period — MVR 2,838m against MVR 2,216m in 2025 — and the gap is entirely in fixed, broadband and enterprise. On mobile the ranking reverses and the gap narrows: MVR 1,645m for Ooredoo against MVR 1,515m for Dhiraagu [OWN-CALC from filed statements].

That produces two businesses serving the same country with different centres of gravity. Roughly three-quarters of Ooredoo’s revenue is mobile (74.2%); for Dhiraagu it is just over half (53.4%). Both are drifting toward fixed and enterprise, which is the only part of either business growing consistently — Ooredoo’s fixed, broadband and enterprise revenue rose from 23.6% to 25.7% of the total in a single year.

The flat market-level mobile line also hides a divergence. Ooredoo’s mobile revenue fell 2.1% in 2025, from MVR 1,681m to MVR 1,645m, while Dhiraagu’s rose 2.7%, from MVR 1,475m to MVR 1,515m. Whatever is eroding mobile monetisation is not falling on the two operators equally.

It also means the two are exposed differently to the same market event. A price war in prepaid data costs Ooredoo more of its revenue base than Dhiraagu’s — which is worth holding in mind when reading any competitive move in that segment.

What cannot be said

There is no market share in this paper, and there is no market share anywhere in this programme. Ooredoo states a customer base of 426,000 [OFFICIAL — Ooredoo Group FY2025]; Dhiraagu states over 452,000 [OFFICIAL — Annual Report 2025]. The regulator counts 839,615 mobile subscriptions and 106,611 fixed broadband connections. The operator figures are customers, the regulator’s are subscriptions, a customer may hold several subscriptions across mobile and fixed, and neither operator publishes the definition it uses.

Adding 426,000 and 452,000 gives 878,000, which is close enough to the mobile total to be tempting and means nothing: it ignores fixed entirely and double-counts multi-product customers. The two numbers are not comparable, and the discrepancy is not resolvable from outside. This is the single most common error in published analysis of this market.

The number that is actually moving

Market-level mobile revenue per subscription — both operators’ mobile service revenue divided by the average subscription count — was MVR 342.78 a month in 2023, MVR 342.26 in 2024 and MVR 325.61 in 2025 [OWN-CALC]. That is a 4.9% fall in a single year.

The arithmetic behind it is not in dispute. Mobile service revenue was MVR 3,156m in 2024 and MVR 3,160m in 2025, a rise of 0.1%. The average base was 768,422 and 808,699, a rise of 5.2%. Revenue divided by a base growing 5% while itself growing 0.1% has to fall by about 5%.

What the arithmetic does not say is why. Four explanations are consistent with it and cannot be separated from outside:

1. Low-value additions. Second SIMs, data-only devices and short-lived visitor SIMs all add to the count and add little to revenue.
2. Migrant composition. A base with more migrant-worker SIMs is a base with lower average spend and higher international calling.
3. Price erosion. Genuine competitive discounting on the packs that most customers buy.
4. Product mix. Customers moving down the ladder to shorter, cheaper packs.

All four have the same consequence for a commercial plan, which is why the distinction matters less than it appears: acquisition is not the constraint in this market. The denominator is already growing. It is the numerator that has stopped.

The visitor cycle inside the resident base

Prepaid subscriptions rise on average in December through March and fall in June and July. The pattern matches the arrivals season, and two million visitors a year passing through a market of 590,000 residents is more than enough to produce it.

This is stated as an association, not a measurement. CAM’s series counts subscriptions, not who holds them; it cannot separate a visitor SIM from a resident one. What can be said is that the shape is there, that both operators sell visitor products, and that nobody publishes how many. Sizing the visitor segment properly needs one dataset neither operator releases — activations by SIM type — which is why this programme frames the segment rather than estimating it.

What the public record cannot see

Of the metrics a telecom commercial function watches, roughly a third can be reconstructed from outside. The financial branch is well covered: both operators file quarterly income statements with a revenue split, and those statements reconcile — summed by year they reproduce both operators’ separately published annual revenue to the rounded million [OWN-CALC], which is the strongest available evidence that this programme’s extraction is right.

The retention branch is entirely dark. Neither operator publishes a churn figure in any filing, annual report or results release, and neither the regulator nor the exchange requires one. That is the largest hole in the public record of this market, and it is why every model in this programme that needs churn runs on a stated assumption rather than an observation.

The honest summary of this paper is therefore in two halves. The market’s size, shape, direction and revenue can be measured, and they say something specific: a market that has been adding subscriptions for four years without adding revenue. Everything below that — who the customers are, what they buy, why they leave, and what it would take to make them worth more — can be framed rigorously and cannot be measured from outside at all.

The frameworks for that second half, with their assumptions on the surface, are in the models section. The decisions the first half supports are in what should management do.

Sources

• Communications Authority of Maldives — monthly telecom statistics, 2012–2026
• Ooredoo Maldives Plc — quarterly reports and annual reports 2023–2026, via CMDA InformInvestor
• Dhivehi Raajjeyge Gulhun Plc (Dhiraagu) — quarterly reports and annual reports 2023–2026, via CMDA InformInvestor
• Ooredoo Group — FY2025 results release
• Maldives Bureau of Statistics — population projections, via Nyra People dossier P01
• Ministry of Tourism — monthly arrivals, via the Nyra tourism pipeline