FinOps in Africa: Controlling
Cloud Costs When the Rand Is Volatile

Cloud cost management is a mature discipline. The FinOps Foundation has published frameworks, tooling vendors have built dashboards, and most hyperscalers now offer native cost management capabilities that are genuinely useful. The frameworks work well for organisations operating primarily in the currency their cloud bills are denominated in.

South African organisations largely do not. Azure, AWS, and Google Cloud bill in US dollars. Revenue is generated in rand. When the rand weakens by 15% against the dollar in a twelve-month period, which has happened more than once, cloud spend in rand terms rises by 15% without a single additional resource being provisioned. This creates budget pressure that no amount of right-sizing or reserved instance purchasing can fully address.

Most FinOps programmes are not designed to account for this. They optimise consumption. They reduce waste. They improve unit economics. They cannot hedge currency exposure. Understanding the distinction between these two problems is the starting point for managing cloud costs effectively in South Africa.

What FinOps Can and Cannot Do

The legitimate value of a FinOps programme is significant. Most organisations that have not actively managed cloud costs have material waste: idle resources, oversized instances, storage that is never accessed, services that were provisioned for a project that ended two years ago. A proper cost optimisation engagement typically finds savings of 20 to 35% in cloud spend without any reduction in capability.

That saving is real and worth capturing. But it is a one-time correction, not a permanent control. Once the waste has been eliminated, ongoing savings come from discipline: tagging resources so you know who owns what, building cost visibility into the deployment pipeline so engineers see the cost implications of what they provision, establishing governance that requires approval for spend above defined thresholds.

What FinOps cannot do is protect you from currency movement. A rand that depreciates 20% against the dollar in a year increases your cloud bill by 20% in rand terms even after every optimisation has been applied. That exposure is a treasury problem, not an engineering problem, and the solutions are treasury solutions.

Managing Currency Exposure Practically

Organisations with large cloud spend and rand-denominated revenue have several practical options. The first is forward contracts. If your cloud spend is predictable in dollar terms, a treasury function can hedge a portion of the exposure using forward contracts, locking in a rand-dollar exchange rate for future cloud payments. This is common practice for organisations with any material foreign currency obligations and there is no reason cloud spend should be treated differently.

The second is budget construction. Cloud budgets should be set in dollars, not in rand, with the rand equivalent calculated at a conservative exchange rate assumption. Using last year's average rate as a budget assumption means any rand depreciation produces a budget overrun that looks like overspend but is actually currency movement. Separating those two things in your reporting makes the real picture visible.

The third is consumption control. Reserved instances and savings plans lock in a committed spend level in dollar terms, which at least makes the exposure predictable even if it does not reduce it. Predictable exposure is easier to manage than variable exposure.

Building a Functional FinOps Practice

A FinOps programme that works in the South African context needs to address both consumption and currency. On the consumption side, the fundamentals are tagging discipline, cost allocation by business unit, anomaly alerting, and a regular review cadence that includes engineers alongside finance. Without engineering involvement, cost decisions get made by people who do not understand what they are cutting.

On the currency side, the cloud team needs to work with treasury, not around it. Cloud spend needs to be part of the organisation's foreign currency exposure reporting. The Chief Financial Officer needs visibility into dollar-denominated cloud commitments alongside every other dollar obligation the organisation carries.

This is a governance and reporting change as much as a technical one. It requires breaking down the organisational boundary between technology spend and treasury management. That boundary is common in large organisations and it produces predictable blind spots.

The Governance Structure That Works

The most effective cloud cost governance we have seen uses a Cloud Economics function that sits at the intersection of technology and finance. It is not owned by either exclusively. It has representatives from engineering who understand what is being provisioned, from finance who understand the budget and treasury implications, and from architecture who understand whether the current spending pattern is producing the right outcomes.

That function meets monthly at minimum, has authority to challenge provisioning decisions above certain thresholds, and reports to both the CTO and CFO. In organisations where cloud spend has crossed the threshold of materiality for the income statement, this governance is not optional. It is how you maintain control of a cost line that, left unmanaged, grows faster than almost any other operating cost.

Cloud cost management in South Africa is the same discipline as anywhere else, with one additional variable that requires an additional response. The organisations that manage it well are the ones that treat the currency exposure as a first-class problem rather than an accounting footnote.