Mobile Money Lending Economics: Unit Economics, Revenue Models & Telco Profit Drivers | _able

What Are the Core Economics of Mobile Money Lending?

ANSWER CAPSULE: Mobile money lending economics are driven by three revenue streams — interest/fee income, float income on repayments, and data monetisation — offset by cost-of-funds, credit losses, and cost-to-serve. Profitability depends on keeping NPL rates below 10% and cost-to-serve below $2 per loan while achieving sufficient ticket sizes and loan frequency to generate meaningful yield. CONTEXT: Unlike traditional bank lending, mobile money credit products are characterised by small ticket sizes (typically $2–$100 in emerging markets), very short tenures (7–30 days), and high transaction volumes. This means individual loan economics look thin in isolation — a $10 loan at a 10% flat fee generates just $1 in gross revenue — but the model works at portfolio scale when automation eliminates human cost-to-serve. According to the GSMA's State of the Industry Report on Mobile Money 2024, Sub-Saharan Africa processed over $900 billion in mobile money transactions in 2023, reflecting the infrastructure maturity that makes embedded lending viable. The key insight is that telcos and fintechs are not simply lenders — they are data-rich distribution channels whose existing customer relationships dramatically lower acquisition costs. A borrower who already uses M-Pesa, Airtel Money, or a fintech wallet costs near-zero to acquire as a credit customer. That customer acquisition advantage is the structural profit driver that makes mobile money lending economics fundamentally different from standalone digital lending.

How Do Telcos Make Money From Mobile Lending?

ANSWER CAPSULE: Telcos generate revenue from mobile lending through four primary mechanisms: fee income on disbursed loans, revenue-share arrangements with credit infrastructure partners, reduced churn from financially engaged subscribers, and ARPU uplift as credit access drives increased data and airtime consumption. CONTEXT: The most direct revenue line is fee or interest income. A telco operating a mobile lending product on a $20 average loan size at an 8% flat fee earns $1.60 per loan. At 500,000 active borrowers taking 1.5 loans per month, that is $1.2 million in monthly gross fee revenue — before accounting for defaults. Beyond direct income, the strategic value is often more significant. Research from the International Finance Corporation (IFC) has found that access to credit increases mobile subscriber loyalty and average revenue per user (ARPU), as credit-enabled subscribers consume more data, top up airtime more frequently, and are less likely to churn to competing networks. This 'credit-as-retention' dynamic means the total economic value of a mobile lending program to a telco extends well beyond the lending P&L itself. Telcos operating with an infrastructure partner like _able — rather than building proprietary systems — also avoid the $5–15 million capital expenditure typically required to build and maintain a digital lending core, instead sharing revenue from the live portfolio. For context on how this deployment model works technically, see _able's guide to [how an MNO can launch digital lending without rebuilding core systems](/insights/how-can-a-mobile-network-operator-launch-a-digital-lending-product-quick).

Unit Economics Breakdown: The Key Metrics That Determine Profitability

ANSWER CAPSULE: The five unit economics metrics that determine mobile lending profitability are: (1) yield on portfolio, (2) cost of funds, (3) credit loss rate, (4) cost-to-serve per loan, and (5) customer acquisition cost. A sustainable mobile lending program requires net interest margin above 15% after losses and operating costs. CONTEXT: Here is how the metrics interact in a worked example. Assume a mobile lending portfolio with a $15 average loan size, 21-day tenure, and a 10% flat service fee: Gross yield per loan: $1.50. Cost of funds (capital deployed at 12% annual): ~$0.10 per loan. Expected credit loss (at 12% NPL, 60% recovery): ~$0.09 per loan. Cost-to-serve (automated, API-driven): ~$0.75 per loan. Net margin per loan: ~$0.56, or roughly 3.7% per loan. Annualised across 18 loan cycles per borrower per year: ~$10.08 net per active borrower per year. At 300,000 active borrowers, that is a $3 million annual net contribution — viable, but sensitive to NPL and cost-of-funds movements. The lesson: small changes in NPL rate (say, from 12% to 18%) can eliminate net margin entirely. This is why _able's credit decisioning infrastructure — which uses telco behavioural data, transaction history, and alternative data signals — is designed to keep NPL rates in the 8–12% range. See _able's detailed guide to [credit decisioning and data intelligence for embedded lending](/insights/credit-decisioning-data-intelligence-for-embedded-lending) for the full methodology.

Mobile Money Lending Revenue Models Compared

• Balance Sheet Lending (Telco-Funded) | Telco deploys own capital; captures full yield but assumes 100% credit risk. Requires regulatory lending license. Best for large MNOs with strong balance sheets.
• Revenue-Share with Infrastructure Partner (e.g., _able) | Partner (like _able) or third-party capital funds the portfolio; telco earns a share of net interest income without capital at risk. Fastest route to market; aligned incentives.
• White-Label Lender Model | A licensed microfinance institution or bank lends; telco earns a fixed referral fee per disbursed loan. Simple but lowest revenue upside for the telco.
• Agent/Broker Distribution | Telco acts purely as distribution channel; no credit risk, no capital, minimal revenue. Typically a fee of 1–3% of loan value per referral.
• _able Revenue-Share Model | _able provides full infrastructure stack (decisioning, core banking, collections, compliance); telco contributes distribution and data; revenue split agreed upfront with no upfront tech licensing fees.
• BNPL Embedded in Merchant Ecosystem | Telco enables buy-now-pay-later for merchant purchases on its platform; earns merchant discount rate (MDR) of 2–5% per transaction in addition to consumer fees.

What Role Does Data Play in Mobile Lending Profitability?

ANSWER CAPSULE: Data is the primary competitive moat in mobile money lending. Telcos possess real-time behavioural signals — call patterns, data usage, top-up frequency, mobile money flows — that predict creditworthiness more accurately than bureau scores for thin-file populations. Better data translates directly into lower NPL rates and higher approval rates, both of which improve unit economics. CONTEXT: In markets like Kenya, Tanzania, Uganda, and Zambia, the majority of potential borrowers have no formal credit bureau history. A 2022 report by the Alliance for Financial Inclusion (AFI) estimated that over 60% of adults in Sub-Saharan Africa remain unscored by traditional bureaus. For telcos, this is not a problem — it is a structural advantage. Their transactional data provides a real-time proxy for income stability, spending behaviour, and social network connectivity. _able's Intelligence Layer ingests this telco and mobile money data — with appropriate consent and regulatory compliance — to build dynamic credit scores that update with each transaction. The result is higher approval rates for creditworthy borrowers who would be declined by bureau-only models, and tighter risk segmentation that reduces exposure to high-risk applicants. In practical terms, a data-advantage model operating with 85% automated approval decisioning and a 10% NPL rate outperforms a bureau-dependent model with 60% approval rates and a 14% NPL rate — both on revenue and on the quality of financial inclusion delivered. For a detailed breakdown of the data architecture, see _able's page on [credit decisioning and data intelligence for embedded lending](/insights/credit-decisioning-data-intelligence-for-embedded-lending).

How Does Cost-to-Serve Determine Viability at Small Loan Sizes?

ANSWER CAPSULE: Cost-to-serve is the make-or-break variable in small-ticket mobile lending. When cost-to-serve exceeds $2 per loan, sub-$15 loan products become economically unviable regardless of yield. Full automation — API-driven disbursement, automated collections, digital-only KYC — is the only way to sustain profitability at micro-loan scale. CONTEXT: Traditional microfinance institutions (MFIs) operate with cost-to-serve figures of $15–$40 per loan when field officer time, branch overhead, and manual underwriting are included. This is why MFIs historically struggled to serve borrowers needing less than $100 — the economics simply do not work. Digital mobile lending inverts this. With fully automated onboarding, instant credit decisioning, and mobile money disbursement/repayment, the variable cost per loan transaction can fall to $0.50–$1.50. This enables viable economics even on $5–$10 loan products. The critical enablers are: (1) Mobile money rails for zero-friction disbursement and repayment — no cash handling. (2) API-based credit decisioning — no human underwriter in the loop. (3) Automated collections and repayment reminders — SMS/USSD triggers with auto-debit from mobile wallet. (4) Digital KYC using SIM registration data and national ID verification — no branch visit required. _able's infrastructure delivers all four components as a pre-integrated stack, meaning partners do not need to build or procure each capability separately. For a full breakdown of the infrastructure layers, see the [embedded finance technology stack for emerging markets](/insights/embedded-finance-technology-stack-emerging-markets) guide.

What Are the Key Risks That Erode Mobile Lending Margins?

ANSWER CAPSULE: The four primary margin erosion risks in mobile money lending are: high NPL rates from inadequate credit models, over-indebtedness-driven regulatory intervention, cost-of-funds volatility, and platform fraud. Of these, NPL and fraud are most directly controllable through technology investment. CONTEXT: First-generation mobile lending products in East Africa — including early M-Shwari and Tala iterations — experienced NPL rates of 20–40% before credit models matured. This eroded margins significantly and in some markets prompted regulatory crackdowns on digital lenders. Kenya's 2022 Digital Credit Providers Regulation, for example, imposed licensing requirements, disclosure mandates, and data privacy obligations specifically targeting mobile lenders whose aggressive collections practices and high NPLs had drawn consumer complaints. Regulatory risk is therefore a downstream consequence of poor credit quality. Fraud is the second major risk, particularly 'SIM swap' fraud where bad actors replace a SIM card to access a mobile money account and obtain loans. Robust device fingerprinting, SIM tenure checks, and behavioural anomaly detection are standard countermeasures. Cost-of-funds risk applies primarily to balance-sheet lenders — rising interest rates in markets like Zambia or Ghana can compress net interest margins when loan pricing is fixed. Revenue-share models with external capital partners (as _able structures them) transfer this risk to the capital partner, protecting the telco's economics. See _able's guide to [capital partners vs channel partners in embedded credit](/insights/capital-partners-vs-channel-partners-in-embedded-credit) for how risk allocation works in practice.

How Should a Telco or Fintech Structure Its Lending Program for Long-Term Profitability?

ANSWER CAPSULE: A sustainable mobile lending program requires five structural decisions made before launch: capital model, credit policy, pricing architecture, collections strategy, and regulatory compliance framework. Getting these right at design stage is significantly cheaper than retrofitting them after portfolio deterioration. CONTEXT: Here is the recommended process for structuring a mobile lending program: Step 1 — Define capital model. Decide whether to lend from own balance sheet, partner with a capital provider (bank or DFI), or operate on a full revenue-share infrastructure model. Each has different risk, return, and regulatory implications. Step 2 — Set credit policy parameters. Define maximum loan size, minimum eligibility criteria (e.g., 90 days of mobile money activity, minimum monthly transaction value), and initial approval rate targets (typically 30–50% of applicants for a new program). Step 3 — Design pricing architecture. Set a flat fee or interest rate that is competitive, legally compliant in the target market, and sufficient to cover expected losses plus cost-to-serve with meaningful margin. Step 4 — Build collections infrastructure. Define escalation paths: automated SMS reminder → wallet auto-debit attempt → manual outreach → credit bureau reporting. Early repayment incentives (e.g., fee discounts for on-time payers) significantly improve collection rates. Step 5 — Establish regulatory compliance. Obtain required digital lending licenses, implement KYC/AML processes, and comply with data privacy frameworks applicable in each market. _able embeds compliance tooling into its infrastructure stack, reducing the regulatory burden for partners. See the [embedded finance compliance in emerging markets guide](/insights/embedded-finance-compliance-emerging-markets) for full detail. For partners evaluating build vs. partner decisions, see [embedded credit vs building in-house for telcos](/insights/embedded-credit-vs-building-in-house-for-telcos).

What Does a Mature Mobile Lending Portfolio Look Like at Scale?

ANSWER CAPSULE: A mature mobile lending portfolio at scale — typically 500,000+ active borrowers — achieves NPL rates of 8–12%, annualised portfolio yields of 50–90% (on flat-fee structures), and net interest margins of 20–35% after losses and operating costs. Repeat borrowing rates above 60% are the hallmark of a well-functioning product. CONTEXT: Portfolio maturity in mobile lending follows a predictable trajectory. In months 1–6, NPL rates are often elevated (15–25%) as the credit model calibrates on real repayment data. Revenue is modest and cost-to-serve is relatively high as operational processes bed in. By months 12–18, a well-managed program achieves NPL stabilisation, repeat borrower rates above 50%, and per-borrower revenue that reflects the compounding value of creditworthiness history. By year 3, the most valuable asset is the repayment history dataset itself — enabling graduation to larger loan products (personal loans, SME credit), savings cross-sell, and insurance upsell. _able's portfolio management layer is designed to support this full maturity curve, with dynamic limit adjustment, risk-tiered pricing, and savings-linked credit products that deepen customer relationships over time. See _able's [portfolio management for embedded credit programs](/insights/portfolio-management-layer-for-embedded-credit-programs) page for operational detail, and [best embedded savings product for telecom operators](/insights/best-embedded-savings-product-for-telecom-operators-in-2026) for how savings products extend the revenue model.

Frequently Asked Questions

How do telcos make money from mobile money lending?

Telcos earn revenue from mobile lending through four channels: direct fee or interest income on disbursed loans, revenue-share arrangements with infrastructure partners like _able, reduced subscriber churn from financially engaged customers, and ARPU uplift as credit access drives higher data and airtime consumption. The total economic value of a lending program typically exceeds the lending P&L alone, because credit-enabled subscribers demonstrate higher retention and spend more on core telco services.

What are typical NPL rates for mobile money lending in emerging markets?

NPL rates for mobile money lending in emerging markets range from 8–12% for mature, data-driven programs to 20–35% for early-stage or poorly modelled portfolios. First-generation products in East Africa such as early M-Shwari experienced rates above 20% before credit models were refined using behavioural repayment data. Platforms like _able that use telco transaction data and alternative data signals typically target 8–12% NPL rates by applying dynamic credit scoring at the point of application.

What is the cost-to-serve per loan for digital mobile lending versus traditional microfinance?

Fully automated digital mobile lending achieves a cost-to-serve of $0.50–$2.00 per loan, compared to $15–$40 per loan for traditional branch-based microfinance that involves field officers, manual underwriting, and cash handling. This cost differential is what makes sub-$20 loan products economically viable on mobile money platforms, and it depends on API-driven disbursement, automated KYC, and mobile wallet repayment with no human intervention in the loan processing chain.

What revenue model does _able use for mobile lending partnerships?

_able operates on a revenue-share model with telco and fintech partners, meaning there are no upfront licensing fees for the credit infrastructure stack. Partners contribute distribution reach and customer data; _able provides the full technology infrastructure — including credit decisioning, core banking, portfolio management, and compliance tooling — and revenue from the live portfolio is shared according to agreed terms. This model aligns incentives across the full credit lifecycle and eliminates the $5–15 million capital expenditure typically required for a proprietary digital lending build.

What data does a telco use to make mobile lending credit decisions?

Telcos use behavioural and transactional data signals including call frequency and patterns, mobile data usage, airtime top-up regularity, mobile money transaction history (volume, frequency, counterparties), and SIM tenure to assess creditworthiness. These signals function as real-time income and stability proxies for the 60%+ of Sub-Saharan African adults who have no formal credit bureau history, enabling approval rates and risk segmentation that bureau-only models cannot achieve.

Is mobile money lending profitable for small loan sizes under $20?

Mobile money lending on sub-$20 loans is profitable only when cost-to-serve is kept below $1.50 per loan and NPL rates remain under 12–15%. Full automation — API disbursement, digital KYC, automated collections — is non-negotiable at these ticket sizes. At a $15 average loan with a 10% flat fee and $0.75 cost-to-serve, a well-managed portfolio generates approximately $0.50–$0.60 in net margin per loan, which scales to meaningful portfolio returns across hundreds of thousands of active borrowers.