Between 1 and 28 September 2026, across the 20 dates this desk collected Spotify’s full 75-market chart panel, the lead-artist-only measure this desk has been publishing captured as little as 7.562 per cent of African-credited chart streams earned outside Africa’s four charted countries, and never more than 10.032 per cent. On every one of the 20 dates, it missed at least nine-tenths of what a full count, lead plus featured credits, actually found.
Every one of 20 collection dates sits between 7.6 and 10.0 per cent. There is no date where the lead-only measure comes close to complete.
Show the numbers
| Date | Lead-only share of combined, % |
|---|---|
| 2026-09-02 | 10.03 |
| 2026-09-12 | 9.69 |
| 2026-09-04 | 9.66 |
| 2026-09-24 | 9.55 |
| 2026-09-19 | 9.26 |
| 2026-09-13 | 9.24 |
| 2026-09-26 | 9.09 |
| 2026-09-16 | 9.01 |
| 2026-09-25 | 8.91 |
| 2026-09-20 | 8.77 |
| 2026-09-15 | 8.73 |
| 2026-09-17 | 8.72 |
| 2026-09-22 | 8.47 |
| 2026-09-01 | 8.44 |
| 2026-09-07 | 8.42 |
| 2026-09-21 | 8.12 |
| 2026-09-27 | 8.03 |
| 2026-09-28 | 7.80 |
| 2026-09-14 | 7.61 |
| 2026-09-06 | 7.56 |
Source: Computed from this desk's registered african-chart-credit-lead-vs-feature-01-to-28-september-2026 dataset, retrieved 1 Oct 2026 · download the data
The gap is not one day. It is every day this desk has checked.#
This desk disclosed the single-day version of this problem on 28 September: lead-artist credit alone put African music’s Spotify export footprint at 276,645 streams that day, while counting featured credits on the same chart rows took it to 3,546,656, a 12.82-times difference. The question that piece left open was whether 28 September was typical or a fluke. It was typical. Across all 20 collection dates between 1 and 28 September, combined credited export streams totalled 70,030,224, split 6,112,680 to lead credits and 63,917,544 to featured credits. Measured date by date rather than summed, the lead-only share of that combined total averaged 8.755 per cent across the 20 dates, never once rising above 10.032 per cent or falling below 7.562 per cent; the featured-credit share averaged the complementary 91.245 per cent across the same 20 dates.
| # | Date | Lead-artist export streams | Featured-credit export streams | Combined credited export streams | Lead-only share of combined, % |
|---|---|---|---|---|---|
| 1 | 2026-09-01 | 348,844 | 3.79m | 4.14m | 8.44 |
| 2 | 2026-09-02 | 385,566 | 3.46m | 3.84m | 10.03 |
| 3 | 2026-09-04 | 370,990 | 3.47m | 3.84m | 9.66 |
| 4 | 2026-09-06 | 298,445 | 3.65m | 3.95m | 7.56 |
| 5 | 2026-09-07 | 378,153 | 4.11m | 4.49m | 8.42 |
| 6 | 2026-09-12 | 329,559 | 3.07m | 3.4m | 9.69 |
| 7 | 2026-09-13 | 329,424 | 3.24m | 3.57m | 9.24 |
| 8 | 2026-09-14 | 306,996 | 3.73m | 4.03m | 7.61 |
| 9 | 2026-09-15 | 272,369 | 2.85m | 3.12m | 8.73 |
| 10 | 2026-09-16 | 278,959 | 2.82m | 3.1m | 9.01 |
| 11 | 2026-09-17 | 278,032 | 2.91m | 3.19m | 8.72 |
| 12 | 2026-09-19 | 294,899 | 2.89m | 3.18m | 9.26 |
| 13 | 2026-09-20 | 295,870 | 3.08m | 3.37m | 8.77 |
| 14 | 2026-09-21 | 292,148 | 3.3m | 3.6m | 8.12 |
| 15 | 2026-09-22 | 260,313 | 2.82m | 3.08m | 8.47 |
| 16 | 2026-09-24 | 289,023 | 2.74m | 3.03m | 9.55 |
| 17 | 2026-09-25 | 274,060 | 2.8m | 3.08m | 8.91 |
| 18 | 2026-09-26 | 272,736 | 2.73m | 3m | 9.09 |
| 19 | 2026-09-27 | 279,649 | 3.2m | 3.48m | 8.03 |
| 20 | 2026-09-28 | 276,645 | 3.27m | 3.55m | 7.80 |
Source: Computed from this desk's registered african-chart-credit-lead-vs-feature-01-to-28-september-2026 dataset, retrieved 1 Oct 2026 · download the data
The mechanism is the one already identified in the 28 September piece, and it recurs because it is structural, not incidental: Burna Boy is credited as a featured artist on Shakira’s “Dai Dai,” Wizkid on Drake’s “One Dance,” Tems on Dave’s “Raindance.” Records like these chart on dozens of markets this desk’s four published African country charts never touch, under a lead artist this desk’s roster correctly does not classify as African. A lead-artist-only count scores every one of those streams as zero African export, on every date, not just the date this desk happened to check first.
The other pattern holding steady across the month is where the African lead-credit footprint sits. The count of charts carrying an African lead artist never fell below 8 and never rose above 10 across the 20 dates, holding at the bottom of that range for most of the second half of the month. The count of charts carrying any African credit, lead or featured, ranged from 44 to 54 over the same stretch, consistently well above the lead-only range on every date. Both measures move within a narrow band of their own; neither is a single-day artefact, and the gap between them is the same structural miss restated as market count instead of streams.
What would have to be true for this to be wrong#
The headline rests on the daily lead_share_of_export_pct column, and it is not held up by one unusual date. Removing 2 September, the single highest day in the series, changes the 20-day average by less than a point, and removing the five highest days still leaves the average under 8.5 per cent.
| # | Scenario | Rows | Total | Mean | Change in mean, % |
|---|---|---|---|---|---|
| 1 | All rows | 20 | 175.10 | 8.76 | 0 |
| 2 | Excluding the top 1 (2026-09-02) | 19 | 165.07 | 8.69 | -0.77 |
| 3 | Excluding the top 3 (2026-09-02, 2026-09-12, 2026-09-04) | 17 | 145.72 | 8.57 | -2.09 |
| 4 | Excluding the top 5 (2026-09-02, 2026-09-12, 2026-09-04, 2026-09-24, 2026-09-19) | 15 | 126.91 | 8.46 | -3.37 |
Source: Recomputed from african-export-credit-lead-share-01-to-28-september-2026, retrieved 1 Oct 2026 · download the data
The total streams this gap represents are similarly spread, not concentrated in one record’s single big day. Dropping the single largest day of lead-credited export streams takes the 20-day lead total down 6.31 per cent; dropping the five largest days takes it down 29.66 per cent, leaving more than two-thirds of the total intact.
| # | Scenario | Rows | Total | Mean | Change in total, % |
|---|---|---|---|---|---|
| 1 | All rows | 20 | 6.11m | 305,634 | 0 |
| 2 | Excluding the top 1 (2026-09-02) | 19 | 5.73m | 301,427 | -6.31 |
| 3 | Excluding the top 3 (2026-09-02, 2026-09-07, 2026-09-04) | 17 | 4.98m | 292,822 | -18.56 |
| 4 | Excluding the top 5 (2026-09-02, 2026-09-07, 2026-09-04, 2026-09-01, 2026-09-12) | 15 | 4.3m | 286,638 | -29.66 |
Source: Recomputed from african-chart-credit-lead-vs-feature-01-to-28-september-2026, retrieved 1 Oct 2026 · download the data
Three limits sit underneath all of this, and they do not cancel each other out. First, the featured-credit count that produces the 91 per cent figure is itself a floor, not a ceiling: it matches credited featured-artist names, normalised and lower-cased, against a limited set of roster names that resolve to exactly one African country, and the underlying dataset’s own method note records that roster names written wholly in Arabic, Cyrillic, Hebrew or CJK script collapse to the same blank key under that matching rule and are excluded from the match entirely, bar one. If anything, that means the true featured-credit total understates the miss further, not less. Second, this is chart presence, not consumption: Spotify’s daily country charts hold a fixed, limited number of ranked positions, so a catalogue streaming steadily without a single standout day is invisible to this measure on either side of the lead or feature count. Third, the panel covers 20 collection dates spread across four weeks, not all 28 calendar days in September, so this describes the dates this desk actually measured, not every day of the month.
What this means if you are making the music#
Three things follow, and they cut against how an export strategy usually gets framed.
A featured credit on someone else’s international record is not a consolation prize next to a lead single charting abroad. On this desk’s own 20-day count, featured credits supplied the large majority of every day’s measured African export presence, while lead-artist exports stayed small and narrow enough to be dominated by the same handful of names this desk has already identified as its persistent exporters. If the goal is a Spotify chart position outside Africa, a placement as a credited feature on a record with existing international chart momentum has outperformed leading your own export push on every single day this desk has checked.
Second, any chart-based claim about “African music’s global reach,” including the ones this desk has published since August under the lead-only measure, has been citing the smaller of two real numbers. That does not make the lead-only figures wrong as a description of African-led exports specifically. It means a reader using them as a proxy for African artists’ total chart presence abroad has been working with roughly a tenth of the credited total, every time.
Third, the correction itself has a floor artists and labels should know about before leaning on it: an artist whose name is written outside Latin script is structurally harder for this method, or any name-matching method built the same way, to credit correctly as a featured artist. That is not a reason to distrust the 91 per cent figure. It is a reason to expect the true gap for artists in that position specifically to be wider than this count can currently show.
