Summary

  • TCRA’s 8 October notice starts a scratch-card phase-out pilot on 1 January 2027 in all Dar es Salaam districts and named districts in four other regions. It lists USSD, mobile applications, mobile-money agents, bank ATMs and agents, and MNO or authorised-partner platforms; this is not an app-only or nationwide cut-off.
  • An unused voucher held in a pilot area after the start date requires the subscriber to contact the MNO for assistance and activation. The notice does not specify the timing, fees, validation rules, rejection route or measured results. The quality of that support handoff is therefore a central implementation question, not an established failure.

The date is narrower than the headline

“Cessation of airtime scratch cards” sounds like a national switch. TCRA’s notice is more limited. Public-awareness activity began on 1 October 2026; the pilot is scheduled to begin on 1 January 2027. It covers all districts of Dar es Salaam, plus named districts in Dodoma, Arusha, Mwanza and Mbeya. TCRA says it will announce a nationwide cut-off later. No date is given for customers outside the pilot.

That boundary matters for subscribers and for companies planning a transition. A customer who lives, works or travels in a pilot area is asked to prepare for digital recharge. The measure should not be described as a countrywide ban already in force, or as evidence that every operator has switched every retail outlet. The notice describes the intended scope and timetable; it does not provide a district-by-district channel inventory or implementation report. TCRA’s public notice is the governing public statement reviewed here.

Digital does not mean app-only

The replacement menu has several routes. A subscriber may buy airtime from a mobile-money account through USSD or an app. The notice also names mobile-money agents, bank ATMs, bank agents, and other digital channels or platforms supplied by MNOs and authorised partners.

That mix is important. USSD can work on a basic handset; an agent or bank counter can connect a cash customer to an electronic transaction. The notice’s “cashless” rationale does not mean each customer must own a smartphone or hold a mobile-money balance personally. At the same time, listing a channel is not proof that a customer can reach it, that an agent has liquidity, that a USSD session completes, or that a bank ATM sells airtime in the relevant district. Those conditions are not documented in the notice.

The national figures show why a large digital ecosystem is not enough to settle the local question. TCRA’s report for the quarter ending June 2026 records 87.05 million mobile-money subscriptions and 31.34 million smartphones. Those are national subscription and device counts, not unique people, named-pilot-district availability or successful airtime purchases. The TCRA report says nothing about the proportion of pilot-area customers who can use each listed channel at the time and place they need it.

The Bank of Tanzania’s 2025 National Payment Systems Annual Report says the mobile-money agent distribution network reached 1.98 million, up 33.78% from 1.48 million a year earlier. That scale is relevant context for distribution. It still cannot establish agent density, opening hours, float, network uptime or recharge support in the districts TCRA selected. National reach and local readiness answer different questions. The Bank’s report describes the broader network; it is not an audit of this pilot.

The voucher moves from a shelf to a support process

The most consequential detail may be what the notice says about old stock in customers’ hands. Subscribers in pilot areas who still hold unused scratch vouchers after 1 January are told to contact their respective MNO through designated customer-care channels or authorised service points for assistance and activation. The notice encourages customers to use existing vouchers beforehand where possible.

It does not say that an unused voucher expires automatically, nor does it promise instant or unconditional activation. It does not define what proof a subscriber must show, how a damaged or unreadable voucher is handled, whether a customer may visit any authorised point, how long an answer may take, or how an unsuccessful request can be challenged. These are open operating terms, not evidence that a voucher will be rejected.

The shift therefore changes where the last mile is tested. Under the old model, the subscriber could buy a card from a shop and enter its code. Under the pilot, an unredeemed card after the start date becomes a support case initiated by the subscriber and handled through the MNO’s designated route. The practical outcome depends on the availability and clarity of that route, the evidence accepted, the time to activation and how customers learn what to do. The notice names the responsible operator but does not report these performance measures.

There is also a retailer question, but the available evidence is thin. TCRA’s stated purpose includes reducing the production, distribution and disposal of physical cards. The notice does not quantify card sales, retailer margins, unsold stock or lost income. It would be premature to claim that shops will be harmed or that operators will save a specific amount. What can be said is that the channel mix changes, while the commercial terms for retailers and authorised agents remain outside the public notice reviewed here.

A pilot can produce a decision record

The planned geography creates an opportunity for a more useful assessment than a national adoption statistic. Results could be reported separately for the participating districts: which channels are available, how often top-ups succeed, how many customers need help with old vouchers, how long activation takes, and how unresolved cases are handled. Operators could distinguish failed transactions from customers who did not attempt a digital route; TCRA could distinguish the pilot boundary from areas awaiting a later national decision.

None of those results is in the notice. The existence of 87.05 million mobile-money subscriptions does not demonstrate that customers have an account, balance, nearby agent or reliable service at a specific moment. Nor does a nationwide agent count show whether a particular pilot district has enough agents or whether they can perform the transaction. For the same reason, an unreported service gap should be treated as a question to measure rather than a failure already proved.

The national cut-off is still open. That gives TCRA and the operators a chance to publish what the pilot tests before the next policy decision: the territorial reach of each channel, the treatment of remaining vouchers and the support burden created by the transition. A clear operating record would make expansion easier to assess and exceptions easier to resolve. Without one, the headline may travel farther than the evidence.

Sources