Business forecasting in volatile macro: a 2026 methodology for South African CFOs

Business forecasting is the discipline of producing a forward view of cash, profit and risk that holds up under realistic shocks.

In South Africa, the forecasting environment is harder than in most peer economies — the SARB rate cycle, rand-dollar volatility, electricity reliability and water security all introduce material variability that a static annual budget cannot absorb.

This piece addresses volatile as the core focus, drawing on Kettle's South African public-sector and corporate advisory practice.

The annual budget remains the default for most South African mid-market and listed-entity finance functions. Business forecasting in volatile macro: a 2026 methodology for South African CFOs is largely an artefact. By the second quarter of the year, the assumptions on inflation, prime, exchange rates and load-shedding stage have shifted enough that variance analysis is more archaeology than steering. This advisory note sets out the rolling-forecast methodology Kettle Consulting recommends for CFOs in 2026 — anchored in King IV governance expectations, ISO 31000 risk principles and the COSO ERM 2017 framework.

Why a static annual budget fails in the South African environment

Three macroeconomic features make the annual budget structurally unfit for purpose in 2026.

• The SARB MPC operates an inflation-targeting framework and has historically moved rates intra-year in response to inflation prints. A budget that assumed prime at 11% in January can be operating against prime at 12% by July or 10% by November.
• The rand has had multi-percentage-point monthly moves against the dollar in most years of the last decade. Companies with material foreign-denominated cost lines or revenue cannot operate on a static assumption.
• Electricity supply has had unpredictable stages of load-shedding, with operational impact on production, logistics and demand. Water restrictions in metros have introduced similar variability.

The methodological fix is moving from a single annual budget toward a 13- or 18-month rolling forecast, refreshed at least quarterly, with explicit scenario branching at the decision-relevant assumptions.

The four benefits of a properly run forecast

Forecasting, done well, produces four benefits that the annual budget does not.

1. Earlier signal on funding need. A rolling cash forecast surfaces a shortfall four to six months before an annual budget revision would.
2. Better capital allocation. Capex sequencing changes when the forecast incorporates updated cost-of-capital and demand assumptions.
3. Stronger risk conversation. The forecast becomes the language in which the audit committee and risk committee discuss exposure, rather than a parallel exercise.
4. Cleaner board reporting. Variance analysis against a rolling forecast is more actionable than variance against a budget locked in 12 months ago.

The methodology in five layers

Layer 1 — the financial backbone

The backbone is a fully integrated three-statement model — income statement, balance sheet, cash flow — driven by a small set of operating assumptions and macroeconomic inputs. The model rolls forward 13 or 18 months. The macro inputs reference published authority: SARB MPC outlook for prime and the repo rate, SARB Quarterly Bulletin and Stats SA CPI release for inflation, Treasury MTBPS for growth and fiscal context.

Layer 2 — operating drivers, not line items

Forecasts that work are driven by a small number of operating drivers — volume, price, conversion rate, customer churn, payroll headcount, capacity utilisation — not by individual general ledger lines. The driver tree should be explicit, with sensitivity bands set against the macro inputs.

Layer 3 — scenarios at the decision-relevant assumptions

Three scenarios — base, downside, severe downside — modelled at the assumptions that matter for the business in question. For an importer, the rand-dollar path. For a manufacturer, the electricity availability and unit cost path. For a financial services entity, the SARB rate path. For a public-sector contractor, the National Treasury budget cycle and Section 71 transfer reliability.

Layer 4 — risk registers tied to the forecast

ISO 31000 and COSO ERM both treat risk identification as part of strategic planning, not a separate exercise. The risk register should map directly onto the forecast assumptions — for each assumption, the risk that it moves materially against the company, with mitigation actions and trigger thresholds. King IV expects the audit committee to see this linkage.

Layer 5 — refresh cadence and governance

Monthly refresh of the operating drivers, quarterly refresh of the macro inputs and scenarios, board sign-off at each quarterly refresh, with a half-year deep review aligned to the interim reporting cycle. The forecast is owned by the CFO; the inputs are governed by a documented assumption-setting process.

The macro inputs you should reference, not improvise

A common error in mid-market forecasting is using internally invented macro assumptions rather than the published views of the institutions that move the variables.

_Source: SARB, Stats SA, National Treasury, NERSA, Eskom, SARS. Use the published source — not press summaries or competitor analyst notes — as the input._

Variable · Authoritative source · Refresh cycle

• Variable: Repo / prime rate — Authoritative source: SARB MPC Statement — Refresh cycle: Six-weekly
• Variable: Headline CPI — Authoritative source: Stats SA CPI release — Refresh cycle: Monthly
• Variable: GDP growth path — Authoritative source: SARB Quarterly Bulletin; Treasury MTBPS — Refresh cycle: Quarterly / annual
• Variable: Fiscal context — Authoritative source: National Treasury Budget and MTBPS — Refresh cycle: Annual
• Variable: Rand-dollar / rand-euro — Authoritative source: SARB and traded forward curves — Refresh cycle: Daily / monthly
• Variable: Electricity supply — Authoritative source: Eskom System Status reports; NERSA decisions — Refresh cycle: Weekly / monthly
• Variable: Unemployment — Authoritative source: Stats SA QLFS — Refresh cycle: Quarterly
• Variable: Trade balance — Authoritative source: SARS customs and SARB BoP statistics — Refresh cycle: Monthly

Where forecasts most often break

Across the engagements Kettle runs with mid-market and listed-entity finance teams, four failure modes recur.

• Optimism in the base case. The base case is set at what the team hopes will happen, with the downside set at what it fears. The base case should be the modal expectation.
• Working capital ignored. Net working capital movement is one of the largest swing items in the cash forecast. Treat it explicitly.
• Capex not sequenced. Capital expenditure is presented as an aggregate annual number with no monthly sequencing — making the cash impact invisible.
• Sensitivity treated as a tail. Single-variable sensitivities (rand at 18, rand at 20, rand at 22) are run but multi-variable scenarios are not. The real downside is a compound shock — rand weakens AND demand softens AND prime ticks up.

A budget is what the company said at the start of the year. A forecast is what the company says now. The board should be steering against the forecast, not the budget.

— Kettle Consulting risk advisory

Governance — King IV and ISO 31000 alignment

King IV Principle 11 sets out the board's responsibility for risk governance. The audit committee charter should explicitly require that the forecast and risk register are tied — and that the audit committee reviews both at each meeting. ISO 31000:2018 sets out the risk management process; the forecast assumption register is the operational expression of clauses 5.4 (risk assessment) and 5.5 (risk treatment). The COSO ERM 2017 framework provides the integration into strategy that audit committees in JSE-listed entities reference.

Practical implications for 2026 finance teams

Three steps a mid-market or listed-entity CFO should take in the next quarter. Replace the annual-budget-only cadence with a 13-month rolling forecast refreshed monthly on operating drivers and quarterly on macro. Tie the risk register to the forecast assumptions explicitly. Document an assumption-setting process that references the authoritative sources — SARB, Stats SA, Treasury, NERSA — rather than internally improvised numbers.