Account-wide opportunity map
MTN Pulse began as a SIM-registration tool. But its two engines — a nationwide operations dashboard and a plane that drives real phones — are general-purpose. This map walks every major MTN function and shows, concretely, where the platform plugs into that team's workflow and what it changes.
Operations intelligence
Live geospatial monitoring, root-cause analytics, anomaly & fraud flagging, and a plain-English assistant — over any operation with sites, agents, devices, or transactions spread across the country.
Real-device automation
Remotely drives real iOS/Android phones over a secure network — by AI instruction or fixed replay — to run, test, benchmark, or measure any app, USSD, or transaction flow. Hard stop at every PIN, OTP, and payment screen.
Figures throughout are drawn from MTN filings, NCC publications, and Nigerian tech press as of mid-2026, and are cited by name in each section. Where MTN does not disclose a number, the point is marked as indicative.
Department 01 · A+B
Owns route-to-market: trade marketing, field sales, and the physical + digital distribution of SIMs, airtime and data through a tiered dealer, sub-dealer and retailer network.
Airtime is going digital — retailers now buy stock through the Direct-to-Retail (D2R) app and USSD *456# (Virtual Top-Up), explicitly to cut banks out of the airtime chain. But visibility of who is actually stocked and selling in the informal kiosk and sub-dealer edge is weak, and trade-channel fraud is real. There is no automated, at-scale check that the retail apps and stock flows work across the field.
MTN rolled out the D2R app nationwide to digitise stock purchase at the retailer edge — a device-driven surface that is already the front line of trade execution.
Numeric distribution, POS/signage visibility, and active-vs-dormant retailer coverage plotted by state and territory, so trade managers see reach and gaps in real time instead of monthly.
Real phones drive the D2R app and *456# to confirm stock purchase, downstream top-up, and Paystack payment complete correctly across device types and regions — catching a broken retail money-path before retailers do.
Synthetic run-throughs of retailer registration and first activation confirm the onboarding journey works end-to-end on the handsets agents actually use.
Commission-gaming, ghost retailers, and abnormal recharge/float patterns surfaced automatically — the trade-side analogue of the registration fraud engine.
Health and management posture of field POS and registration devices — enrolment, kiosk lock, compliance — surfaced beside their sales activity.
Department 02 · A+B
Compliant subscriber acquisition — capturing and validating registrations against NCC/NIMC rules, and governing the field agents and dealers who do it.
This is the platform's origin story. Registration runs through a field-agent network on biometric kits and agent apps, with overwhelmed agents and glitches as the stated bottleneck. The compliance stakes are severe: MTN barred 8.6M lines over NIN issues, and the NCC penalises non-compliant registration at ₦200,000 per SIM — a competitor was fined ₦104M in May 2025 for exactly the anomalies this platform detects.
The NCC fine on Airtel (May 2025) for night-time registrations, one-NIN-many-SIMs, and out-of-shop devices — every category auto-detected here.
Every site across all 36 states + FCT on one live map, coloured by health, with an 8-stage funnel that pinpoints which step is slow or failing.
Auto-surfaces the exact anomalies the NCC penalises — midnight–6am registrations, duplicate-NIN clusters, out-of-shop devices, abnormal agent velocity.
Real handsets periodically run the registration flow to confirm the compliance gates — NIN validation, biometrics, image match — are genuinely enforced in the field.
Every operator scored against their own site's peers, with automatic outlier and velocity flags for ghost registrations and incentive-gaming.
Track NIN-linkage completion and re-registration throughput as barred users return — the exact flow behind the 8.6M-line squeeze.
Department 03 · A+B · lead opportunity
Owns the NOC, RAN operations, and service-quality assurance — meeting the NCC's regulated QoS thresholds and closing the gap between network counters and real customer experience.
Quality is measured by drive tests, surveys and NOC counters — but counters aren't experience, and drive tests are expensive, episodic and road-bound (no indoor, rural or 24/7 coverage). The regulator has moved: the NCC × Ookla alliance now publishes crowdsourced, device-side QoE maps quarterly. Meanwhile QoS breaches cost ₦5m + ₦500k per day, per parameter, and a March 2026 NCC directive ordered operators to compensate subscribers with airtime in failing areas.
The regulator itself now measures on-device experience. Continuous real-phone measurement (B) feeding a dashboard (A) is now a board-level need, not a nice-to-have.
Real phones across regions measure download, upload and latency against the NCC's 4G thresholds (DL ≥20 Mbps, UL ≥5 Mbps, 50 ms) — indoors, rural, and around the clock, where drive tests can't reach.
Every measurement mapped against its regulated threshold, with live fine-exposure (₦5m + ₦500k/day per breached parameter) by cell and region.
Newly deployed rural sites (RuralCow, Project Zero) validated by real handsets on the ground — confirming a live site actually delivers usable experience.
A real phone buys a data bundle, confirms it provisions, then measures the achieved speed — tying network quality to the actual revenue transaction.
Generate MTN's own independent, timestamped device-side QoE dataset — the same shape as the NCC×Ookla maps — to get ahead of the public reports.
Department 04 · A+B
Contact-centre operations, care across channels, complaint resolution, and NPS/CSAT for ~87M subscribers — with a hard mandate to deflect care into self-service.
Self-service runs on Zigi, an OpenAI/Microsoft/Superbo-built AI assistant, across WhatsApp, Telegram, Facebook, web and inside myMTN — plus IVR (180) and USSD. It's a huge, fragile surface: when a Zigi purchase path or NIN check breaks silently on one channel, it doesn't just fail — it generates care volume. Nobody is continuously testing these journeys end-to-end across real devices.
MTN claims the top net-promoter score — defended by a 5-channel AI care surface that has no automated end-to-end testing behind it.
Real phones drive Zigi, the IVR (180) and USSD self-service through common journeys — balance, buy/borrow, NIN status — catching silent breakages on every channel before customers hit them.
Reproduce a reported problem on a matching handset and OS on demand, so care and engineering see the actual failure instead of guessing.
A live view of which self-service journeys are healthy, where deflection is leaking back to human agents, and how NPS/FCR track by channel and region.
Department 05 · B
Owns the myMTN app, the USSD service menus, and web self-service — the front door of MTN's platform strategy.
myMTN has 10M+ installs and ships rapid point releases; USSD was consolidated onto *123# (with *310# balance, *303# borrow, *904# airtime) after NCC harmonisation. That's an enormous release surface over a fragmented, Android-dominant, low-to-mid-end handset base — with no visible automated multi-device regression testing. Journeys break silently between the app, the USSD gateway and billing.
A release surface this big, shipping this fast, with device-driven regression testing on real handsets is the single most natural fit for the automation plane.
Every release, the app's key journeys are re-run on a spread of real device/OS combinations — catching what a single test phone or an emulator would miss.
*123#, *310#, *303# and *904# menus driven on real handsets to confirm the money-journeys (buy, borrow, balance) still complete after each change.
The revenue-critical flows verified end-to-end across app and USSD — the ones that silently break between front-end and billing.
A fast, deterministic smoke test on physical handsets gates every build before it reaches 10M+ users.
Department 06 · A+B · high-value
Operates MTN's CBN-licensed Payment Service Bank — wallets, payments, an agent network for cash-in/cash-out, and the USSD/app rails that carry it all.
MoMo's own audited FY2024 numbers tell the story: active wallets fell ~47% to 2.8M and agents ~79% to 68k as MTN deliberately purged dormant, commission-gaming agents. Distinguishing genuinely active, liquid agents from padded rosters is the stated #1 operational problem. The rails — USSD *671# and the MoMo app — are the daily revenue surface, and the 2022 ₦22.3bn unauthorised-transaction incident is a permanent reminder of the fraud stakes.
Agents fell to 68k and wallets −47% to 2.8M — MTN has publicly admitted agent-activity and wallet integrity is its top fintech problem.
A geospatial view separating genuinely active, liquid agents from padded rosters — with dormant-wallet clusters and float anomalies by region.
Real phones drive cash-in/out, P2P, merchant-pay, bill-pay and onboarding to confirm they complete and to time them — stopping hard at every PIN and OTP.
Synthetic wallet onboarding verifies the KYC gates actually hold and the sign-up journey completes across devices.
Transaction-velocity and dormant-wallet analytics surface abuse patterns and re-engagement opportunities.
Department 07 · B
Consumer marketing, segmentation, tariff and promo design, and keeping watch on what Airtel, Glo and T2 are doing in the market.
MTN holds >50% of 188M subscribers and pushed ARPU to $3.60 (from $2.17), so promo integrity and competitor moves are high-stakes. Yet two things are still done by hand: verifying that MTN's own promos (the 700% bonus, XtraValue) actually provision across the device base, and benchmarking competitors' prices, offers, apps and USSD. Rivals change constantly — and Airtel already beat MTN on Opensignal's Video Experience.
Competitor price/offer/app benchmarking is manual mystery-shopping. A real-phone fleet can do it continuously, at scale, across My Airtel, Glo Cafe and MyT2.
Real phones drive competitor apps (My Airtel, Glo Cafe, MyT2), their USSD menus, prices and speeds — turning ad-hoc mystery-shopping into a continuous, structured feed.
Verify MTN's own promos actually provision the right value on real handsets across device/OS/SIM combinations — before a campaign goes wide.
New offers and journeys tested across the fragmented, low-to-mid-end handsets MTN's customers actually carry.
Competitor prices, offers, app changes and measured speeds tracked over time against MTN's own — a live scoreboard for the marketing team.
Department 08 · A+B
B2B — enterprise connectivity, IoT/M2M, fixed and fibre broadband (FibreX, FWA), managed SD-WAN, cloud and data-centre services, all under contractual SLAs.
Enterprise grew +22.7% (ex a USSD one-off) and home broadband reached ~4.2M on the way to an 8M-homes-by-2028 target across 14 states. That fast-scaling base needs two things this platform provides: independent, on-device proof that the marketed SLAs (up to 99.99%, especially for banks on SD-WAN) actually hold at the customer premise, and validation that circuits, SIMs and IoT fleets are correctly provisioned.
The Own-The-Home target on FibreX/FWA across 14 states — a scaling base that needs provisioning validation and independent SLA proof.
Real devices at (or standing in for) the customer premise independently verify throughput, latency and availability against the contracted SLA — proof for demanding enterprise clients like banks.
Validate SIM provisioning, connectivity and rules across large IoT/M2M device fleets managed in the Control Center.
A geospatial view of the fixed footprint, active premises and SLA-compliance status across the 14 live states.
New circuits, SIMs and SD-WAN policies exercised on real devices to confirm they're configured correctly at activation.
Department 09 · A+B
Regulatory affairs and statutory compliance — managing the NCC and NDPC relationships, keeping the licence in good standing, and avoiding fines.
The reporting burden is heavy and evidence-hungry: quarterly NCC QoS filings across 2G–5G KPIs, SIM-registration audits, and NDPA obligations including 72-hour breach notification. The downside is existential — QoS breaches at ₦5m + ₦500k/day per parameter, NDPA penalties at the greater of ₦10m or 2% of annual gross revenue, and a history that includes the $5.2bn SIM fine. Most of the evidence is still assembled by hand.
The NDPA penalty ceiling — on top of quarterly QoS filings and ₦5m/parameter fines. A heavy, evidence-hungry burden the platform can automate.
Aggregate quality and coverage measurements into NCC-format filings and audit trails — the quarterly report assembled automatically.
Real phones generate timestamped, independent QoS and USSD records — proving quality the way the regulator now experiences it via NCC×Ookla.
The registration self-audit output formatted as a regulator-facing compliance report against the exact penalised categories.
Verify consent and data-handling steps in customer-facing journeys are present and working, supporting NDPA audit-return filings.
Department 10 · B · light A
Protect telecom and fintech revenue from leakage and fraud — SIM-box bypass, subscription fraud, promo/bonus abuse, and USSD/MoMo fraud.
SIM-box detection depends on test-call generation — placing controlled international calls and inspecting the number that lands on-net — which fundamentally needs real SIMs placing on-net and off-net calls. The NCC pegs cumulative bypass losses near $3bn across ~750,000 fraud numbers, and in March 2026 launched TIRMS, a real-time telecom-identity platform governing SIM recycling and re-verification. That's a regulator-mandated, real-SIM problem — inherently a device-fleet task.
A brand-new, regulator-backed real-SIM verification mandate — plus the ~$3bn SIM-box loss — that the device-automation plane directly serves.
A fleet of real SIMs places on-net and off-net test calls and inspects the terminating CLI to expose masked, local-appearing numbers — then feeds the SIM ranges for blocking.
Drive real devices through promo mechanics to confirm bonuses can't be gamed or double-claimed.
Prove rating, billing and USSD/MoMo correctness from the actual on-device flow — the errors CDRs alone can't reveal.
Detections plotted by route and region to reveal bypass hotspots and quantify leakage geographically.
Department 11 · B · A-adjacent
Internal IT, BSS/OSS, service management, and the automation and AI agenda — including the group-wide Genova programme.
MTN has already validated this category: 13 computer-vision bots replaced ~200 SIM-verification workers, and the Genova AI programme targets R30bn of value. The open problems are legacy sprawl, slow and costly manual UAT/regression, and the brittleness of traditional GUI-scraping RPA — bots that break the moment an app's UI changes. AI-agent, computer-use automation is the emerging answer, and MTN is funding it.
MTN already built and funded exactly this class of automation internally. This is the scalable, less-brittle version of a bet MTN has already placed.
Computer-use automation that drives real apps by intent, surviving UI changes that break traditional screen-scraping RPA — replacing slow, costly manual UAT.
Continuously exercise charging and billing journeys end-to-end on real devices to catch faults in the Ericsson-era stack before customers do.
The same app/USSD automation, re-pointed across MTN operating companies — each new market or app is a playbook, not new plumbing.
A dashboard of what's automated, what's passing, and where coverage gaps remain across the estate.