Summary
- Vodacom Group printed revenue growth of 25.6 per cent and normalised growth of 12.5 per cent for the six months to 30 September 2026, and service revenue at 29.4, 13.4 and 13.3 per cent — three bases for one line in one period.
- Across four reporting periods in the material read, the reported-to-normalised revenue gap runs 0.7, 2.1, 5.5 and 13.1 percentage points, and the only printed percentage-point reconciliation belongs to the 0.7-point quarter.
Two indices, one period
The full-year booklet for the year ended 31 March 2026 sets the pattern. Group revenue of R167,652m against R152,227m is printed as 10.1 per cent growth, and in the next column as 12.2 per cent. Service revenue of R133,561m against R120,734m reads 10.6 and 12.9. EBITDA of R62,626m against R55,511m reads 12.8 and 14.2. The asterisk carries the definition: normalised growth adjusts for trading foreign exchange and for foreign-currency translation on a constant-currency basis using the current year as the base, for hyperinflation accounting, and excludes merger, acquisition and disposal activity at constant currency where applicable (https://www.vodacom.com/pdf/investor/annual-results/2026/results-booklet.pdf).
Both figures are exact at their own definition, and neither is the other. The reported number answers what the group booked in rand. The normalised number answers what the operating businesses would have booked had the currency and the perimeter stood still. A reader who quotes one without the other is quoting half the instrument.
The regulated JSE announcement repeats the same pairs and fixes the scope of the comparison: every growth rate compares the year ended 31 March 2026 with the year ended 31 March 2025 (https://vodacom.com/pdf/sens/2026/vodacom-group-reviewed-annual-results-and-cash-dividend-distribution-for-the-year-ended-31-march-2026.pdf). The integrated report restates the convention and the headline: revenue of R167.7 billion up 10.1 per cent from R152.2 billion, with normalised Group service revenue growth of 12.9 per cent (https://vodacom-reports.co.za/integrated-reports/ir-2026/documents/vodacom-group-limited-integrated-report-2026.pdf).
The gap is not a constant
Four periods in the material read, four different distances between the two indices. For the full year, 2.1 percentage points on revenue. For the quarter ended 31 December 2025, 0.7. For the quarter ended 30 June 2026, Group revenue of R42,404m against R40,028m printed 5.9 per cent reported and 11.4 per cent normalised, a gap of 5.5 points, with Group service revenue at 6.3 and 12.6 per cent; the company attributes reported growth to translation effects from the stronger rand (https://vodacom.com/pdf/sens/2026/trading-update-for-the-quarter-ended-30june-2026-the-quarter-short-form-announcement.pdf).
Then the half-year, where the instrument gets crowded. Revenue of R102,501m against R81,585m prints 25.6 per cent reported and 12.5 per cent normalised — a 13.1-point gap. Service revenue of R85,132m against R65,806m prints 29.4 per cent, 13.4 per cent and 13.3 per cent. Three rates for one line in one period. The third is pro-forma, which assumes Safaricom was part of the group for the full comparative period from 1 April 2025 and adjusts for constant-currency translation; the same statement prints pro-forma service revenue of R99,140m against R87,484m, up 13.3 per cent (https://vodacom.com/pdf/investor/interim-results/2026/booklet.pdf).
Normalised and pro-forma are not rivals here; they answer slightly different questions about the same consolidation. Normalised excludes Safaricom. Pro-forma rebuilds the comparison as though Safaricom had been inside the group all along. The 0.1-point distance between 13.4 and 13.3 is the size of the difference between those two constructions — and a reader who does not know which basis a colleague is quoting will not see it.
Where the arithmetic is printed
Only one document in the material read shows the bridge in full. The trading update for the quarter ended 31 December 2025 prints a reconciliation table: Group revenue 11.0 per cent reported, translation FX 0.7 percentage points, 11.7 per cent normalised (https://www.vodacom.com/pdf/investor/quarterly-results/2025/vodacom-trading-update-3q26.pdf). That table is re-performable with one addition. It is also the only period in the set where a single adjustment accounts for the whole distance, because the same update records the Maziv acquisition as implemented on 1 December 2025 and accounted for as an associate, which leaves no perimeter line to add.
The pattern is worth stating carefully. The period with the smallest adjustment is the period where the arithmetic is complete. The period with the largest adjustment, thirteen points of it, prints the two endpoints and no schedule between them. That is an observation about where the arithmetic appears, not an allegation about why. A reconciliation of 0.7 points is cheap to print and demonstrates diligence. A reconciliation of 13.1 points requires separating a currency layer from a consolidation layer, and the material read does not do it.
The presentation's exchange-rate appendix shows why direction alone will not close the gap. Average rates for the year: ZAR/USD 17.35 against 18.24, ZAR/EGP 2.82 against 2.69, ZAR/KES 7.46 against 7.12 (https://www.vodacom.com/pdf/investor/annual-results/2026/presentation.pdf). The rand strengthened against the dollar and weakened against the Egyptian pound and the Kenyan shilling in the same year. A stronger rand cuts translated revenue; a weaker rand against two other currencies pushes the other way. Mixed rates of that kind are exactly the case in which a reader cannot infer the rand effect from the direction of one exchange rate, and the appendix quantifies rates, not rand.
What three subtractions show
Apply each printed normalised rate to the printed prior-year base and compare the result with the printed total. For FY26: R152,227m times 12.2 per cent lands at R170,799m against the printed R167,652m. For the quarter to 30 June 2026: R40,028m times 11.4 per cent lands at R44,591m against R42,404m. For the half-year: R81,585m times 12.5 per cent lands at R91,783m against R102,501m — this time below the printed total, because the reported half-year carries the Safaricom consolidation while the normalised rate does not.
Those are my subtractions on printed quantities. They are not reported currency effects, and Vodacom has not published them as such. What they establish is narrower and more useful: the two indices are not related by a fixed constant, so a reader cannot convert one into the other from the definitions alone.
The boundary of this reading
The material read on 20 November 2026 prints no rand bridge that separates a currency effect from an acquisition and disposal effect on revenue. For the full year the 2.1-point gap is the combined normalised adjustment, not a standalone foreign-exchange number. For the half-year the reconciliation labels pro-forma Safaricom inclusion and constant-currency adjustments as separate lines, but the extracted table is layout-sensitive and this briefing does not reconstruct a per-component rand allocation from it. For the June quarter the short-form update gives no rand translation amount.
Those are limits of what was obtained, and other tables in the results pack may carry more.
What the material does print is enough to fix the question a reader should ask before repeating any of these numbers: which base, which perimeter, which currency, and over what period.
Why it matters at the desk
A growth rate that moves thirteen points between two columns of the same statement is not only a statement about the business. It is a statement about measurement, and the choice of column changes the sentence a reader can honestly write. Quoting 25.6 per cent describes a group that has just consolidated a large operator. Quoting 12.5 per cent describes the same group's existing businesses at constant currency and constant perimeter. Both sentences are true. They are not interchangeable, and an investment case, a lending conversation or a regulatory submission that silently swaps one for the other has changed its subject.
This is where BTW's earlier Vodacom coverage stops and this briefing begins. We have written about the KwaZulu-Natal capex pledge and the record needed to test it, the Western Cape investment plan, the virtual wheeling rebate cycle, and a MIMO antenna trial that addressed site space rather than expansion economics. None of those read the results documents as an instrument. The delta here is the instrument itself.
Read the base first. The growth rate is the second question, and it is the one that travels. The group's own entry in our directory is https://btw.media/en/directory/vodacom-za.
Member Briefing
Read the full trend briefing
Members read the full briefing with the source notes behind every finding.
Strategic Circle
A daily briefing in your language, built from changes we have verified.
Join for USD 20/monthCancel anytime.
Members also get the full research library, a daily briefing on the networks they follow and the private forum.






