Connectivity · Operations Memo · November 2026
Typhoons do not read coverage maps. Through the season’s hardest weeks, Rimvo’s 500+ rural towers held availability above plan for 1.2 million connected residents — on solar-hybrid power, battery backup and district crews who pre-staged spares before the clouds closed roads. This is the storm memo behind the 1.2M case study.
- 500+ towers held; 1.2M connected stayed online; ticket latency within SLA.
- Power: solar-hybrid + storage skipped diesel on most sites; backhaul via edge caches.
- Method: weekly ops, decision latency −45% (see Active Governance).
- Finance: $30M blended facility; shared-tower economics (see Catalytic Capital).
1. What the storm hit: wind, flood, brownout
Wind stressed mounts and microwave alignment; flood cut road access to three clusters; brownouts tested batteries sized for 2–4 hour autonomy. Urban carriers measure such weeks in complaints; rural carriers measure them in access — school video-hours, clinic records, mobile-money uptime. Devex readers know the gap: coverage exists, capacity fails. Rimvo’s rural-first design (see build memo) exists precisely for these weeks.
2. What held: hybrid power, caches, crews
Solar-hybrid sites rode through brownouts on batteries; edge caches kept schooling and payments usable on throttled backhaul; congestion models (Nervyra-derived) prioritized emergency and education over bulk video at peak. District crews — hired locally, trained via Foundation apprenticeships — cleared alignment and refueled only where hybrid could not cover. One cluster ran constrained for a day on microwave reroute; cause code logged, spare pre-positioning adjusted. Availability, not ARPU, remains the KPI (see Patient Capital).
“We stopped asking what rural ARPU is. We ask what rural availability is. Availability created the ARPU.” — Rimvo, Network Lead
3. Exhibit: storm weeks vs. plan
| Metric | Actual vs. plan |
|---|---|
| Tower availability (rural) | 102 |
| Ticket latency (lower better) | 88 |
| Diesel liters vs. plan (lower better) | 58 |
| Data sessions vs. plan | 109 |
Management logs, typhoon window. Table canonical.
4. Tickets, not just towers: latency is the SLA
Failed rectifiers became day-stories, not quarter-stories — 45% faster decisions via weekly rooms (see cadence). Congestion shaping held schooling and payments first; bulk video queued. The same traces hardened the $12M European carrier deal at 99.99% — South-to-North transfer in one storm.
5. Densification after the storm: small cells, caches, corridors
Next: small cells on commercial corridors, edge caches at schools/clinics, enterprise corridors to energy sites and mills. Each addition reuses shared power and spares — incremental CapEx falling as usage compounds (2.1× year one). Maintenance transitions to district crews; consent compounds into the next 100 permits (see 10-City). Coverage connected 1.2M; densification keeps them — the November lesson for December builds.
Spares, access, rehearsal: unglamorous readiness
Pre-staged rectifiers, microwave spares and battery strings at district warehouses — not central depots behind flooded roads. Access rehearsals before the season: which track floods first, which microwave path degrades first, which school hosts the edge cache if backhaul throttles. Trouble-ticket drills with congestion shaping (schooling + payments first, bulk video queued) meant peak-week prioritization was muscle memory, not a meeting. The 45% faster decision room (see cadence) turns a failed mount from a quarter’s story into a day’s.
Power math deserves emphasis: solar-hybrid + 2–4 hr autonomy skipped diesel through most storm evenings (diesel −42% vs. plan), the same mesh logic as Meridian East. Shared towers split rent; enterprise anchors (energy sites, clinics) stabilize site revenue before consumer ramp compounds. Low-ARPU bankability is a stack, not a single stream — and storms prove stacks, not slogans.
Densification logic: corridors, caches, enterprise
Next: small cells on commercial corridors, edge caches at schools/clinics, enterprise corridors to mills and energy sites. Each addition reuses hybrid power and spares — incremental CapEx falling as usage compounds (sessions +9% vs. plan even in storm weeks). Maintenance transitions to district crews trained via Foundation apprenticeships; consent compounds into the next 100 permits. Coverage connected 1.2M; densification keeps them — video that loads, payments that clear, telehealth that holds.
ARPU follows availability: the stack that pays
Consumer data ramp (2.1× year one), mobile-money activation, enterprise anchors and edge-cache savings — four revenues stacking where one ARPU never pencils. Shared towers split rent; hybrid power cuts diesel; anchors stabilize before consumer compounds. Patient holds (5+ years) beat quarter models that strand assets before ramp. Availability created ARPU; storms just proved it again — then the European carrier bought the proof at 99.99%.
FAQ
Were any areas cut off?
One cluster ran rerouted microwave for a day; no tower lost power beyond battery autonomy. Full cause codes via data room.
How is diesel tracked?
Liters per site per week, reported with availability; hybrid sites near zero through storm evenings.
What is next?
Densification + enterprise corridors; see Rimvo case.
Sources: Rimvo network + ticket logs (management). External links for context only.








