Summary

  • Vodafone's H1 FY27 announcement prints the six months to 30 September 2026 twice: statutory revenue of €21,853m against €19,609m, up 11.4%, and the same revenue up 4.7% on the basis the company calls organic; service revenue is 12.8% one way and 5.8% the other.
  • Vodafone's own four-column bridge — reported growth, then M&A and Other points, then foreign exchange points, then organic growth — is printed in full in the 2026 Annual Report for the year ended 31 March 2026, where one complete row closes exactly. The matching half-year cells were not in any document read on 21 November 2026, so the 6.7-point group revenue gap can be measured but not split.

Two columns, and one table that carries only the first

The group financial summary is a three-column table: H1 FY27, H1 FY26, Reported change %. Revenue €21,853m against €19,609m, 11.4. Service revenue €18,409m against €16,327m, 12.8. Other revenue €3,444m against €3,282m, with no growth column of its own. Adjusted EBITDAaL €6,451m against €5,728m, 12.6. Adjusted EBITDAaL margin 29.5% against 29.2%, and no growth column there either. Every one of those cells is €m unless the row says per cent, and the whole table is reproduced under those three headings (RNS 4341C re-publication, London Stock Exchange).

What that table does not carry is any organic column at all. A reader who stopped there would conclude that Vodafone grew service revenue 12.8% and never learn that the company also publishes 5.8% for the same line.

The segment tables are a different instrument. They have four columns — H1 FY27, H1 FY26, Reported growth %, Organic growth % — and they are the place where the two perimeters are set beside each other on the same row, for the same six months, in the same currency (H1 FY27 results, National Storage Mechanism). The release also states the group organic service revenue rate directly: "Service revenue: Grew 12.8% to €18.4 billion (H1 FY26: €16.3 billion), and increased 5.8% on an organic basis, with growth across all segments" (Form 6-K, 10 November 2026).

The presentation goes further and prints organic rates for the other headline lines: total revenue 4.7%, and Adjusted EBITDAaL 6.4%, against reported rates of 11.4% and 12.6% (H1 FY27 results presentation). So the company's own material, in three places, gives a reader both numbers. This is not a hidden measure. It is a second column that happens to be absent from the first table a reader meets.

What 'organic' excludes, in the company's words

The definition is worth quoting rather than paraphrasing, because it is an instrument with a stated exclusion list rather than a synonym for 'real'. In the 2026 Annual Report, page 229: "Organic growth presents performance on a comparable basis, excluding the impact of foreign exchange rates, mergers and acquisitions, the hyperinflation adjustment in Türkiye and other adjustments to improve the comparability of results between periods" (2026 Annual Report, page 229).

Four exclusions, then. Currency. Acquisitions and disposals. The Türkiye hyperinflation adjustment. And a residual category for other comparability adjustments. The page also prints three statements that belong beside the quote, because they cap what the measure can be used for: organic growth is not intended to be a substitute for reported growth, nor superior to it; 'organic' is not a defined term under GAAP and may not be comparable with similarly titled measures reported by other companies; and no comparative is provided for organic growth rates because they describe change between the beginning and the end of the current period.

That last sentence is the one that most often goes missing in the retelling. The organic rate is not a rate on a base that can be reconstructed from a prior period's published organic figure, because the company does not publish one for comparison.

The wording has moved too, and the movement is small but real. The 2025 Annual Report, page 216, prints the same sentence with "the hyperinflationary adjustments in Türkiye" in the plural (2025 Annual Report, page 216). The 2026 Annual Report prints the singular. Two annual reports, one publisher, one letter of difference in a defined term. It is a wording difference, not a disagreement, and it does not change a single number. But the same page 216 is also where the company lists the measures for which organic growth is calculated: revenue; service revenue; mobile service revenue; fixed service revenue; Vodafone Business service revenue; financial services revenue in South Africa; M-Pesa revenue; Adjusted EBITDAaL; and Adjusted EBITDAaL margin.

Nine measures. In the H1 FY27 material read on 21 November 2026, printed organic rates were obtained for four of them: revenue, service revenue, Adjusted EBITDAaL and Adjusted EBITDAaL margin. Rates for mobile service revenue, fixed service revenue, Vodafone Business service revenue, financial services revenue in South Africa and M-Pesa revenue were not obtained in any document read that day. That is a boundary of the material, not a claim that the company does not publish them.

What the perimeter actually contains

The half-year report explains its own shape more thoroughly than most. Page 35 of the presentation: "Following the completion of the sales of Vodafone Spain and Vodafone Italy, we have updated our financial reporting to recognise that Vodafone Spain and Vodafone Italy are now discontinued operations in accordance with International Financial Reporting Standards (IFRS). Accordingly, except where otherwise noted, the Group's results exclude Vodafone Spain and Vodafone Italy. Discontinued operations are also excluded from the Group's segment reporting" (H1 FY27 results presentation, page 35).

The 2025 Annual Report supplies the dates those two businesses left, which is a different fact from the date they stopped being reported: "The Group completed the disposals of Vodafone Spain and Vodafone Italy on 31 May 2024 and 31 December 2024, respectively. These entities were classified as discontinued operations in the prior financial year" (2025 Annual Report, page 90).

So the two sold businesses are outside both the group totals and the segment tables. They do not appear as a segment with a negative or a zero. They are simply not there, which is why the group total needs no reconciliation line to reach the segments.

On the other side of the ledger, the release names what came in: "Total revenue: Increased 11.4% to €21.9 billion (H1 FY26: €19.6 billion), driven by strong service revenue growth and the consolidation of Safaricom, VodafoneThree, and Telekom Romania. This was partially offset by adverse foreign exchange movements" (Form 6-K, 10 November 2026).

And then a limitation that matters more than its brevity suggests: "Unless otherwise stated, H1 FY27 results include three months of Safaricom" (H1 FY27 results presentation). Three months, inside a six-month reporting period, for one of the three businesses named as a driver of an 11.4% increase.

Three roles, then, for three sets of names. Vodafone Spain and Vodafone Italy are sold and excluded from both the group results and the segment reporting. Safaricom, VodafoneThree and Telekom Romania are named among the consolidations behind the reported increase, with one of them contributing three months of the half-year. And the five reporting segments — Germany, the UK, Other Europe, Türkiye and Africa — are the lines on which both growth columns are printed for both periods. A reader who merges any two of those roles will misread the tables.

Nothing in the material read states what the H1 FY26 comparative contained for Safaricom, VodafoneThree or Telekom Romania. The presentation's limitation is written about the current half-year only. The comparative is what it is, printed without that caveat, and a reader cannot tell from these documents how much of the base already carried those businesses.

The segment pair, line by line

Here is where the two perimeters become measurable, because every segment row prints both columns. All figures €m unless stated; parentheses are the company's own, and they mean negative (H1 FY27 results, National Storage Mechanism).

Germany: total revenue 6,053 against 5,996, up 1.0%; service revenue 5,483 against 5,425, up 1.1% reported and 0.4% organic; other revenue 570 against 571; Adjusted EBITDAaL 2,139 against 2,191, (2.4)% reported and (3.2)% organic; margin 35.3% against 36.5%.

UK: total revenue 4,858 against 4,409, up 10.2%; service revenue 4,040 against 3,664, up 10.3% reported and 0.9% organic; Adjusted EBITDAaL 984 against 884, up 11.3% reported and 5.9% organic; margin 20.3% against 20.0%.

Other Europe: total revenue 2,912 against 2,804, up 3.9%; service revenue 2,520 against 2,415, up 4.3% reported and 1.4% organic; other revenue 392 against 389; Adjusted EBITDAaL 772 against 835, (7.5)% reported and (4.6)% organic; margin 26.5% against 29.8%.

Türkiye: total revenue 1,836 against 1,601, up 14.7%; service revenue 1,523 against 1,327, up 14.8% reported and 32.8% organic; Adjusted EBITDAaL 531 against 485, up 9.5% reported and 32.7% organic; margin 28.9% against 30.3%.

Africa: total revenue 5,427 against 3,950, up 37.4%; service revenue 4,504 against 3,183, up 41.5% reported and 13.4% organic; Adjusted EBITDAaL 2,058 against 1,347, up 52.8% reported and 18.0% organic; margin 37.9% against 34.1%.

The same segment tables, with the same margins, appear in the later 6-K filed on 18 November 2026 (Form 6-K, 18 November 2026).

The spreads between the two printed rates are my own subtractions, and they are the reason this piece exists. Africa service revenue: 41.5% reported less 13.4% organic is 28.1 percentage points of difference between two figures describing the same six months in the same table (A1). Türkiye runs the other way: 32.8% organic against 14.8% reported, an 18.0-point gap in the opposite direction (A2). The UK: 10.3% against 0.9%, a 9.4-point spread (A3). Germany: 1.1% against 0.4%, 0.7 of a point (A5).

Other Europe is the one to sit with. Service revenue is 4.3% reported against 1.4% organic, a 2.9-point spread in one direction. Adjusted EBITDAaL is (7.5)% reported against (4.6)% organic, a 2.9-point spread in the other. In one segment, on one page, on the same line item and the same period, the reported column is better than the organic column for revenue and worse than it for profit. The two columns are not simply a high reading and a low reading of one business. They are two perimeters that can pull apart in opposite directions at once (A4).

The quarterly path sharpens the point. Group organic service revenue was 5.2% in the first quarter and 6.4% in the second, which is how a 5.8% half-year happens. Germany printed 1.2% then (0.5)% inside a 0.4% half-year (A5), and the UK printed 0.6% then 1.2% inside 0.9%. A half-year rate is an average of two quarters that can differ from each other by more than either differs from zero.

Where the segment table prints both components of revenue, the components add up: Germany 5,483 plus 570 equals 6,053, and 5,425 plus 571 equals 5,996; Other Europe 2,520 plus 392 equals 2,912, and 2,415 plus 389 equals 2,804 (A11). The segment tables are internally exact. Nothing about the two-column structure is a loose end.

Group level: three spreads and a margin with two printed values

At group level the same subtraction gives 12.8% less 5.8% for service revenue: 7.0 points (A6). Total revenue: 11.4% less 4.7% is 6.7 points (A7). Adjusted EBITDAaL: 12.6% less 6.4% is 6.2 points (A8). Each pair is printed for the same measure, the same period and the same currency unit.

The margin is the most interesting of the four, because the two printed values are of different kinds. The reported margin moved from 29.2% to 29.5%. Subtract one printed percentage from the other and you get 0.3 percentage points — that is what the table gives a reader who does nothing but take the difference between two adjacent cells. The company separately states that the margin "was 29.5% versus 29.2%, and improved 0.5 percentage points on an organic basis". One movement, two printed values: 0.3 points as the difference between two printed percentages, 0.5 points as the company's reported organic change (A9).

Neither is a misprint. They measure the same margin over two different perimeters, and the 0.2-point difference between them is the part of the movement that comes from the exclusions rather than from the continuing business.

Germany's margin shows the same structure with a negative sign. 35.3% less 36.5% is minus 1.2 points as a subtraction. The table prints (2.4)% reported and (3.2)% organic for the EBITDAaL line of the same segment, so there are three numbers describing one movement and the reader has to choose which instrument answers the question (A10).

Four columns, one printed row, the wrong year

The 2026 Annual Report, page 230, prints the mechanism in full. The columns are Reported growth %, M&A and Other pps, Foreign exchange pps, Organic growth % (2026 Annual Report, page 230). One row is complete in the material: Group Adjusted EBITDAaL, €11,351m against €10,932m, reported growth 3.8%, M&A and Other (2.4) pps, foreign exchange 3.1 pps, organic growth 4.5%.

That row closes exactly: 3.8 plus 3.1 less 2.4 equals 4.5 (A12). Note what the arithmetic says about direction. In that year, the scope column reduced reported growth by 2.4 points while currency added 3.1, so reported growth of 3.8% was lower than organic growth of 4.5%. The scope column was a drag, not a lift (A13).

The period is the trap, and the document sets it plainly. That row covers the year ended 31 March 2026. It does not cover the six months to 30 September 2026. It proves that the four-column instrument exists, that it closes, and what each column means. It cannot be used to split the half-year gap, and the H1 FY27 values in those two middle columns were not obtained in any document read on 21 November 2026.

The direction point still matters, because the half-year release describes its reported increase in the opposite way. The consolidations of Safaricom, VodafoneThree and Telekom Romania are named among the drivers of growth above the organic rate, partially offset by adverse currency movements. In the annual row, scope subtracted 2.4 points. In the half-year narrative, scope is part of what pushed reported growth above organic. The scope column changes sign between the two periods the material covers, and no single rule carries across (A13).

What a reader can rebuild, and where the arithmetic stops

Take group revenue. A reader has 11.4% reported and 4.7% organic for the same six months, both printed. The difference is 6.7 percentage points. That much is measurable from the documents without any estimate.

What a reader cannot do is say how much of that 6.7 came from acquisitions and disposals and how much came from currency. The residual is not the scope column. It is what is left after two printed rates are subtracted, and it can be split only by the two numbers the half-year material did not supply. The same holds for the 7.0-point service revenue gap and the 6.2-point EBITDAaL gap, and for the margin items, where the two printed values differ by 0.2 points.

The four things a reader should be able to state about each headline measure, and can state for these, are: which perimeter it belongs to; which period it covers; which currency column it is stated in; and which printed percentage change attaches to it. On those four, the documents are complete. The fifth — the decomposition of the difference into scope and currency — is printed annually and was not obtained for the half-year. That is a fact about what the material read on 21 November 2026 contained, not a finding about what the company publishes elsewhere.

Why a company keeps two of them

The exclusion list explains itself. Currency moves a euro reporter's revenue without any change in what customers bought. A disposal removes revenue permanently and would otherwise make a flat business look shrinking. Adding three months of one company and six of another makes the total jump for reasons no operator controls. Strip all of that out and the resulting rate answers a narrow, useful question: on the businesses held throughout, in their own currencies, did the operation sell more than it sold a year earlier? To that question, 5.8% is the honest answer for group service revenue, and 13.4% for Africa.

The reported column answers a different question, and it is not a lesser one. It is what the group actually collected in euros, on the perimeter it actually owned, for the period it actually reports. That is the number that pays for capital, funds spectrum and settles dividends. 12.8% is the honest answer to that question.

The cost of confusing them is arithmetic. A reader comparing Vodafone's reported 12.8% service revenue growth with a peer's organic rate would be comparing a perimeter that gained three months of Safaricom with one that did not. The error would be 7.0 points at group level and 28.1 in Africa. The company itself says neither measure is superior. The documents do not say the two are interchangeable.

What would close the gap

Two observable conditions would settle the half-year decomposition without any inference from me. One is a readable copy of the interim non-GAAP reconciliation carrying the M&A and Other pps and Foreign exchange pps columns for the six months ended 30 September 2026. The other is the 2027 Annual Report, whose page 230 equivalent would cover the year ending 31 March 2027 and would therefore contain the second half-year inside a full-year bridge.

Until one arrives, the honest statement is the one printed above: two rates, both the company's, one period, and a difference that can be measured to the tenth of a point but not attributed.

Vodafone's growth rate is not one number. It is two, both printed, and the document that would show a reader how they connect covers the year that ended eight months before the half-year it is used to explain.

Elias Ward writes on company strategy and telecom economics. Full coverage of the group, including the disposals and the UK consolidation, sits in the Vodafone Group directory entry.