In November 2024 we stood in a Kumasi market and bought 15 used PCs. Eighteen months later Christian Life Academy is running a fully equipped computer lab — 40+ machines, teacher computer, projector, AC, and Starlink internet. This guide is the summary of what we learned, sorted into the phases every comparable project goes through.
Phase 1 — Needs assessment (before any order)
Before buying a single computer, you need to clarify with the school:
- Who covers running costs? Electricity, internet, maintenance — the school, not you. If that's unclear, the project will quietly die after 12 months.
- Which room is available? An air-conditionable, lockable room with enough power outlets. Ideally built or renovated by the school itself — that's the strongest commitment signal.
- Who will be the lab steward? A specific teacher whose schedule has time for upkeep and supervision. Without someone in charge the room is unusable in 6 months.
- How many students will actually use it? A school of 200 doesn't need 40 PCs day one. Start small, document success, then scale.
We didn't do this systematically in November 2024 — and had to retrofit some pieces in Phase 3. Today we'd budget two weeks for the needs assessment, with an on-site visit and a written commitment from the school.
Phase 2 — Procurement: local vs. container
No black-and-white answer; it's a pragmatic mix.
Local purchase (Kumasi market)
When it makes sense: small batches, fast start, first 5–15 devices to test.
Pros:
- Operational from day one, no customs delay
- Local warranty + repair chain
- Invoices with Ghanaian VAT, clean for German bookkeeping
- Money stays in the region
Cons:
- Unit prices ~30 % above container imports
- Specs vary by stall — what's available today might not be tomorrow
- Requires trusted local contacts
Container import (ex-USA, used)
When it makes sense: scaling to 30+ devices, planned expansion.
Pros:
- Lower unit prices (no wholesale margin)
- Uniform specifications
- More modern hardware (US ex-lease devices are typically 1–3 years old)
Cons:
- Customs in Ghana is complicated and expensive (~15 % of value + fees)
- 4–8 week wait
- Warranty repairs require shipping back → practically not feasible
Our mix: first 15 local (November 2024), then container import of 25 PCs (January 2025). In hindsight we'd have timed the second tranche slightly later — the school needed time to physically prepare the room.
Phase 3 — Infrastructure: power, climate, network
The hardware is only one third of the investment. Often forgotten:
Power
Ghanaian schools often have unstable grid power. Three building blocks:
- Surge-protected power strips at every workstation
- Online UPS for the central server/switch area
- Solar backup with inverter and battery as a medium-term goal — coming as our next expansion phase
Climate control
Tropical climate + 40 PCs in one room = slow-motion hardware death. Air conditioning isn't a comfort, it's a capex-protection measure. An 18,000 BTU inverter unit costs ~€1,000-1,200 in Ghana and keeps the room at 24 °C with 40 active PCs.
Network
Structured cabling with RJ45 instead of Wi-Fi. Wi-Fi sounds simpler but with 40 simultaneous clients it's a complete failure. A 4U network cabinet with a 48-port switch and patch panels costs ~€600 and lasts 10 years.
Phase 4 — Internet
Where fiber is missing, Starlink is currently the only practical option. Three points:
- Official reseller: Starlink has several official resellers in Kumasi with invoices, installation, and support. Don't import via third countries — it's tariff-expensive and excludes support.
- Tariff: "Residential Lite" handles 40 PCs comfortably (multiple Mbit/s per device for normal browsing/e-learning). Cost: ~€50/month.
- Ownership: antenna and hardware stay with the gUG, the school covers the running tariff. Documented in the equipment-use agreement.
Phase 5 — Operation: what must be in the model
This is where most well-meaning projects fail. Three pieces:
Ownership stays with the gUG
Don't gift it to the school, lend it. With a written equipment-use contract. That protects against privatization in conflicts, against resale, and gives the gUG the legal position to recover the equipment if misused. Specifically: the school is the user, not the owner. See also asset binding and Nutzungsüberlassungsvertrag.
Operating costs with the recipient
Electricity, internet, maintenance — the school. This isn't harshness — it's the mechanism that makes the project sustainable. If the school can't afford these running costs, it's not yet ready for the donation. Help with a smaller first step instead.
Designated person + onboarding calls
One teacher as the lab steward, with monthly calls during the first 6 months. Topics: small hardware issues, software updates, classroom management, new lesson content. After 6 months the person is usually self-sufficient.
Phase 6 — Documentation and follow-through
What we document continuously — for accounting and for donors:
- Photo + date of every major purchase (computer, AC, Starlink kit)
- Invoices with Ghanaian VAT for German tax filing
- Equipment-use contract with the school, countersigned
- Half-yearly status report with usage figures and challenges
- Publication of relevant lessons on this site
What we'd do differently today
- More time for Phase 1: needs assessment 4 weeks rather than 4 days.
- Smaller first tranche: 10 instead of 15 PCs locally, then container — the room was initially too small for 40+.
- Plan solar earlier: don't wait for grid failures, build solar backup with one initial panel from the start.
- Recorded training videos instead of live calls: with time zone differences, short documented training videos are more efficient than scheduled calls.
Planning your own?
If you're planning a similar project with an African school, write to us at info@empowered-africa.org. We'll share templates for equipment-use contracts, needs-assessment question lists, and our supplier contacts in Kumasi — for free. Two good projects in parallel beats one project repeating avoidable mistakes.