ARPU does not grow from tariffs: bundles, devices, subscriptions, partner offers

ARPU structure as an asset

ARPU — Average Revenue Per User — usually appears as a single number in management reports. That simplification hides the structure that matters for management.

Real ARPU is made up of components that behave differently. Connectivity ARPU (tariff, minutes, gigabytes, SMS) stagnates or declines as the market saturates. Roaming ARPU depends on travel patterns, dropped sharply post-pandemic and recovers slowly. Device ARPU comes from smartphone sales and instalment plans. Bundle ARPU is extra services sold with the tariff. Partner-offer ARPU is the commission on third-party services sold through the operator’s channel. B2B ARPU is corporate subscriptions.

At most operators connectivity ARPU is 70-85% of the total. Which means that for ARPU to grow, what has to grow is not connectivity (where saturation has already arrived) but the other components. And that growth requires a different operating model for each.

What does not work

Raising tariff prices. In a mature market with MNP, raising tariffs triggers churn faster than it brings revenue. In Uzbekistan operators compete actively, MNP works (mnp.uz), and free porting windows intensify competition (beeline.uz). Pricing discipline matters more than price increases here.

Adding functionality to the tariff at no extra price. “Give the customer more for the same money” looks right in the moment, in practice it accelerates commoditisation. The customer gets used to more for the same money, competitors copy, ARPU does not grow.

A pure VAS strategy (selling extras through mass SMS campaigns). Conversion on VAS offers via mass campaigns is low, retention of those services is low, and there is no long-term ARPU effect.

Premium tariffs without a soft migration. Creating an expensive tariff and asking customers to switch. Most do not switch voluntarily. Forced migration drives churn.

Four components that can grow

Bundles

A bundle is a package of several services at one price below the sum of individual prices. The classic telecom bundle is connectivity plus internet plus content or streaming.

What works. Bundles with a clear value proposition (“save 20%”). Bundles with a partnership (a tariff plus a streaming subscription at a discount). Family bundles with flexible distribution between members.

What does not work. Bundles with artificial composition (“connectivity plus insurance plus antivirus”). Bundles with complicated pricing where the customer cannot see the saving.

Unit economics. A bundle lifts ARPU by 15-30% as adoption grows by 20-40%. Net effect — ARPU growth alongside retention improvement (a bundle is harder to leave).

Device financing

Smartphone and device instalment plans. One of the most effective ways to lift ARPU and reduce churn at the same time — the financing contract anchors the customer for 12-24 months.

What works. Financing on mid-to-premium tier devices. Partnership with retailers for a wide channel. Linking financing to the service contract: if the customer leaves, the financing accelerates.

What does not work. Financing on entry-level devices — margin does not cover operational cost. Financing without a credit check — defaults rise.

Unit economics. Device financing adds 50-100K UZS per month to ARPU for 12-24 months. At scale, a meaningful uplift.

Partner subscriptions

Streaming, music, gaming, education content — each is growing in Uzbekistan. The operator can sell subscriptions through its channel (billed alongside the tariff) and receive commission or revenue share.

What works. Local services with strong popularity. International streaming with local pricing adaptation.

What does not work. Generic “video content” without specific partners. Subscriptions on top of the tariff with no discount versus the standalone price.

Unit economics. A subscription adds 10-30K UZS per month at 5-15% conversion of the base. Material at scale.

Partner offers and marketplace

Do not try to sell everything through one channel. Instead — a partnership platform where partners sell through the operator’s customer base.

What works. Insurance (especially device insurance), travel, retail offers, B2B services. A structured partner programme with clear revenue-share rules.

What does not work. Random partnerships without strategic logic. A marketplace without logistics — the buyer expects delivery, the operator has no infrastructure.

Unit economics. A partner offer adds 5-15K UZS per month at 10-30% conversion of the target audience. Less per user, but scaling potential is large.

What ties the four together

Each of the four components needs two preconditions to work.

The first — app frequency. As discussed in the app monetisation article, low frequency kills any commercial channel. If a customer opens the app 2-4 times a month, selling a bundle, financing or subscription through the app is hard. Frequency first, sales second.

The second — an operating model. Understanding which customer is ready for which product, which channel works best for which type of sale, how experimentation runs. Without an operating model, all four components produce a marginal result.

Focusing on one component without these two conditions yields a marginal lift. With both conditions, all four components can be developed in parallel.

A realistic 24-month roadmap

Months 1-6. Foundation. A frequency project (bill payments hub, family management or similar). Operating model for commercial decisions.

Months 7-12. Pilot of the first two components. Usually — bundle redesign plus device financing. Both have relatively easy integration and produce results fast.

Months 13-18. Expansion. Partner subscriptions (one-two partners). Operating model deepening with experimentation.

Months 19-24. Marketplace. Partner offer engine. Multi-channel orchestration.

Two years in, ARPU should be up by 15-30% with retention improvement. Not “magic” — operational discipline and investment in the right directions.

When ARPU expansion is not a priority

If the operator is fighting for market share (survival), ARPU expansion is not the priority. Market position first, monetisation second.

If the technology base (BSS/OSS, billing) does not support new products — every bundle launch takes months — the priority should be around-core architecture, not product expansion.

If the commercial organisation cannot work with complex products (bundles, subscriptions, partnerships) and only with basic tariffs, ARPU expansion delivers a marginal result until capability is built.

If the board is focused on market share or volume rather than margin, ARPU-expansion investments will look off-priority.

Discussion points for the committee

What is the current ARPU structure? Not one number, the breakdown by component.

Where is margin best — connectivity, devices, partnerships? That determines where growth investment delivers more impact.

Which component has the largest gap today (high market potential, low operator share)? That is the first priority.

What needs to be in the operating capability to capture that gap? That is the roadmap.

How SamaraliSoft can help

ARPU Growth Architecture — analysis of the operator’s current ARPU structure, identification of 2-3 gaps where growth is realistically possible in 12-18 months, design of an operating model for capture, and a 90-180 day pilot plan. Not “we will improve marketing” — a concrete expansion programme with unit economics for each product.

Related reading

• /en/insights/telecom-app-monetization/ — frequency as foundation
• /en/insights/telecom-growth-after-connectivity/ — five layers of growth
• /en/insights/telecom-subscriber-intelligence-operating-model/ — operating model
• /en/insights/telecom-wallet-trap/ — when wallet is a trap

Sources

• Digital Government Portal — digital infrastructure UZ
• UzDaily — telecom market 24,873 billion UZS
• Ministry of Digital Technologies — 98% telecom coverage
• MNP.UZ — number portability
• Beeline UZ — porting conditions