Published: 9 July 2026 | By AOLC
South African businesses have become remarkably resilient. Over the past few years, most organisations have found ways to keep operating during load shedding — generators, inverters, solar panels, and adjusted working hours have become part of the normal business toolkit. But resilience at the human level and resilience at the IT level are two very different things.
While your staff may carry on with their work during an outage, your IT infrastructure may be quietly accumulating damage with every power cycle. Servers, storage systems, network equipment, and desktop computers are all affected by the repeated interruptions that load shedding causes — and the consequences range from nuisance (slow boots) to catastrophic (permanent data loss).
This article breaks down exactly what load shedding does to your IT infrastructure, and what steps you can take right now to reduce the risk.
South Africa experienced over 6,000 hours of load shedding in 2023. Each outage is a potential IT event — and most businesses are still relying on inadequate protection.
Most business owners assume that if their computers turn off during load shedding and turn back on when the power returns, no harm has been done. This assumption is wrong — and it can be expensive.
Here is what is actually happening inside your equipment during a typical load shedding cycle:
faster hardware degradation in a daily load shedding environment compared to stable-power environments — a significant factor in total cost of ownership.
Hardware damage is painful but replaceable. Data loss is often neither. Load shedding creates several specific data risks that businesses overlook:
Tip
Schedule your backup jobs to run at 10:00 PM rather than midnight or 2:00 AM — Eskom's Stage 2 to Stage 4 schedules often hit the early morning hours hardest. A completed 10 PM backup is far better than an interrupted 2 AM job.
If your business runs on-premise servers — a physical server room or a rack in your office — you are directly exposed to every load shedding event. Your UPS provides a bridge, but it is rarely long enough to carry a server room through a 2.5-hour Stage 4 outage, let alone the extended Stage 6 cuts of recent years.
Businesses running on cloud infrastructure are largely insulated from local outages — Microsoft Azure, Google Cloud, and AWS data centres have multi-redundant power systems that your local load shedding schedule cannot touch. Your staff lose internet connectivity during an outage (if your router is down), but your data and applications remain intact.
| Risk | On-Premise Server | Cloud / Microsoft 365 |
|---|---|---|
| Data during outage | At risk (UPS dependent) | Safe — not affected |
| Hardware wear | High (power cycling) | None (your hardware is minimal) |
| Staff productivity | Stops completely (if no generator) | Limited by internet/device battery |
| Backup integrity | Interrupted if outage hits backup window | Continuous — cloud handles it |
| Recovery after outage | Manual restart and integrity checks needed | Instant — reconnect and continue |
This comparison does not mean every business should abandon on-premise infrastructure immediately. Many SA businesses have specific compliance, latency, or connectivity reasons to keep servers local. But it does highlight why managed IT providers strongly recommend a hybrid approach — keeping critical applications in the cloud while managing local hardware with proper power protection.
Not all UPS units are created equal, and most businesses buy the wrong kind. Here is what you actually need in a South African load shedding environment:
maximum lifespan of a UPS battery in a daily load shedding environment — far shorter than the 5-year rating on the label, which assumes stable grid power.
South Africa's Protection of Personal Information Act (POPIA) requires responsible parties to implement reasonable technical measures to protect personal information. A business that loses customer data because it had no UPS, no offsite backup, and no disaster recovery plan is not just facing a data loss — it may be facing a POPIA compliance investigation.
The Information Regulator has made clear that "reasonable measures" include protecting data from foreseeable risks. In South Africa in 2026, load shedding is not an unforeseeable risk — it is a scheduled, published, predictable event. Failing to plan for it is hard to defend.
Under POPIA, load shedding is a foreseeable risk. Organisations that lose personal data due to power-related IT failures without documented mitigation controls may face regulatory scrutiny from the Information Regulator.
Documenting your power protection measures — UPS specifications, backup schedules, tested recovery procedures — is not just good IT practice. It is part of your POPIA compliance posture. AOLC's managed IT service includes documentation of all infrastructure controls as part of your monthly reporting.
Use this checklist to assess your current exposure and identify gaps:
Tip
Ask your IT provider to run a simulated power failure test. Pulling the plug on your server room (with proper preparation) is the only way to know if your UPS, shutdown software, and recovery procedures actually work. Don't discover a gap during a real Stage 6 event.
Not sure if your IT infrastructure can handle South Africa's power environment? We'll assess your UPS coverage, backup integrity, and recovery readiness — and give you a clear action plan.
Get a Free IT Assessment