What a gigabyte actually costs in the Maldives

The question this paper answers, and the one it cannot

Seventy-two per cent of Maldivian mobile subscriptions are prepaid, and that share has been flat since 2022 — see the market paper for why that matters. Prepaid economics is therefore not a corner of this market; it is most of it.

This paper takes every product the two operators publish — 204 of them — and puts them on a common basis: price, allowance, validity, price per day, price per gigabyte. It then asks whether that last metric means anything, and concludes that it usually does not.

What it cannot do is say which products sell. Neither operator publishes subscription counts by product, so there is no revenue mix here, no cannibalisation measurement and no estimated elasticity. The portfolio can be described precisely; its performance cannot be described at all. That boundary is held throughout.

Forty-six-fold, inside one duration band

Take only the fourteen published prepaid data packs lasting between 21 and 31 days — the closest thing this market has to a like-for-like comparison. Effective price per gigabyte across them runs from MVR 4.43 to MVR 202.75 [OWN-CALC from published catalogues].

Some of that spread is ordinary volume discounting, and it is steep: buying 168GB rather than 1.5GB reduces the unit price by 95%. But the ladder is not monotone, and two comparisons are worth pausing on.

MVR 99 buys 500MB for 28 days. MVR 125 buys 1.5GB for the same 28 days. Twenty-six rufiyaa more buys three times the data, so anyone on the smaller pack is paying 2.4× the unit price for the privilege of spending 26 rufiyaa less. That is a real customer outcome, and it is the ladder’s design rather than an accident.

MVR 500 buys 35GB pooled over 28 days; MVR 495 buys 56GB on the Salhi plan. Five rufiyaa less for 21GB more — but here the comparison is not clean, and saying so matters more than the finding. Salhi refreshes 2GB a day and does not pool: a customer who needs 10GB on one day cannot get it from the 56GB pack, and can from the 35GB one. The headline gigabyte figures make these look like the same product with different numbers. They are different products.

The metric that breaks

What a gigabyte costs depends entirely on how you count it

Median effective price per GB by pack duration · headline allowance to consumption-capped · MVR

daily (≤1 day)OWN-CALC4.6–51.2 MVR per GB

weekly (2–7 days)OWN-CALC10.4–13.9 MVR per GB

multi-week (8–21 days)OWN-CALC15.6–15.6 MVR per GB

monthly (22–31 days)OWN-CALC14.6–14.6 MVR per GB

quarterly (32–100 days)OWN-CALC5.8–5.8 MVR per GB

long duration (>100 days)OWN-CALC4.2–4.7 MVR per GB

4.2 MVR per GB51.2 MVR per GB

Now the problem. Group the same prepaid data packs by how long they last and take the median price per gigabyte two ways: on the advertised allowance, and on the allowance capped at three gigabytes a day of plausible consumption.

On the headline measure, daily packs are the cheapest data in the country at a median of MVR 4.58 per GB, undercutting monthly packs at MVR 14.64 by more than three to one. On the capped measure, daily packs cost a median of MVR 51.20 per GB — three and a half times the monthly packs they appeared to beat. The ordering does not narrow. It inverts.

The extreme case makes the mechanism obvious. Dhiraagu’s FLEXIHOUR 20 advertises 100GB plus 25GB for MVR 20, valid for one hour. On the headline that is MVR 0.20 per gigabyte — by a factor of twenty the cheapest data sold in the Maldives, and a number that would top any price-per-GB league table in the region. To actually consume 125GB in sixty minutes requires a sustained 280 megabits per second for the full hour, on a phone. The product is an unlimited hour, priced as access to a moment: a football match, a video call home, a download before a flight. Its true unit cost depends entirely on what the customer does with the hour, and for a plausible three gigabytes it is MVR 160 per gigabyte — the most expensive data in this catalogue.

Forty-five of the 123 products with a measurable allowance are in this category: they advertise more data than their validity allows anyone to use [OWN-CALC]. For more than a third of the portfolio, the advertised gigabyte figure is a fair-use ceiling doing marketing work.

The three-gigabytes-a-day cap is an analyst assumption, not an operator disclosure, and it is stated as such on every chart that uses it. A reader who prefers five or one can move it on the product explorer; the inversion survives any value inside an order of magnitude.

The methodological warning, stated plainly

Price per gigabyte is an imperfect proxy for customer value. Network quality, speed, validity, included calls, app-specific allowances, daily-refresh versus pooled allowances, family gifting and fair-use ceilings all differ between these products, and none of them appears on a per-gigabyte axis. Two products at the same MVR/GB can be entirely different propositions, and one of them can be worth twice the other to the same customer.

This programme uses the metric anyway, because it is the only common denominator available — but never as the sole basis for a competitive claim, and always alongside the capped version.

What the portfolio is actually built to sell

Read the whole catalogue rather than any one product and a clear commercial design shows through.

Access, not volume. Seventy of the 204 products carry an unlimited or “limitless” element, and 49 carry app-restricted allowances — data that only works on named social, streaming, gaming or professional apps [OWN-CALC]. The portfolio’s centre of gravity is not gigabytes; it is what the gigabytes are for.

Duration as the real price axis. The same nominal allowance is sold at MVR 75 for 24 hours, MVR 100 for 3 days, MVR 225 for 7 days, MVR 400 for 15 days and MVR 750 for 28 days. The customer is buying time, and the operator is monetising it far more steeply than volume.

Long-duration lock-in. Both a 180-day and a 365-day plan are priced at exactly MVR 2,026 — a year of connectivity paid up front, at the lowest unit price in the catalogue. For an operator with no published churn data, an annual prepaid plan is the cleanest retention instrument available.

The 2025 annual reports name the launches behind this design: all-in-one prepaid combo packs, enhanced international-calling packs, a prepaid plan built specifically for expatriate workers, entertainment and gaming bundles, and USSD-delivered personalised offers [OFFICIAL — Ooredoo Maldives Annual Report 2025]. The direction is consistent: away from selling data, toward selling occasions and identities.

Why the two catalogues cannot be counted against each other

Dhiraagu publishes 124 products, including its complete prepaid pack ladder with SMS activation keywords. Ooredoo publishes 80, and its prepaid add-ons are largely sold through its app, USSD (*929#) and self-care rather than its website.

That difference is disclosure practice, not portfolio breadth. Ooredoo’s own annual report describes a prepaid pack range that its public pages do not list. Any competitive analysis that counts published products and concludes one operator has a wider range is measuring two websites.

Where both do publish — postpaid rate plans, visitor packs, fixed broadband — comparison is possible and is made on the products page. Everywhere else, this programme compares what is comparable and says so.

What follows commercially

Three things follow from the catalogue alone, without needing volumes.

The short-duration band is where the competition is. It has the densest clustering of products, the most promotional flagging, and the most inverted per-gigabyte economics. That is where a price move by either operator would be felt first — and, in a market of two, where a matched response does most damage to both.

The ladder does not reliably reward buying more. A customer stepping from MVR 99 to MVR 125 nearly triples their data; a customer stepping from MVR 350 to MVR 450 doubles theirs; a customer stepping from MVR 450 to MVR 500 gains 17%. The rungs are uneven, which means the ladder is not steering anyone reliably upward. Re-laddering is the cheapest available intervention on revenue per subscription — and revenue per subscription is the number that is falling.

Simplification has a real prize. Multiple packs sit at near-identical price and validity, each carrying build, billing, support and marketing cost, and each adding a choice the customer has to make. Retiring the duplicates is the one initiative on this programme’s opportunity ranking that is entirely within an operator’s control and needs no new capability.

What none of this can establish is how much revenue any of it carries. That needs product-level subscription counts, which is an operator dataset — and it is the first thing a commissioned version of this analysis would ask for.

---

Sources

• Ooredoo Maldives — published plan catalogue and public rate-plan endpoint, observed 18 August 2026
• Dhiraagu — published prepaid and postpaid plan pages, observed 18 August 2026
• Communications Authority of Maldives — monthly telecom statistics, for the subscription base
• Ooredoo Maldives Plc and Dhiraagu — annual reports 2025, for the product-launch record