Openserve's Q1 FY2027 Results, Read for What They Mean for Your Fibre

Telkom released its Q1 FY2027 trading update on 3 August 2026, covering the quarter ended 30 June - and it's the first set of numbers investors and the industry have seen since Openserve launched its own direct-to-consumer ISP in July, a move that put South Africa's largest open-access fibre network in direct competition with more than 200 wholesale partners that resell its infrastructure.

The trading update itself doesn't mention Openserve ISP's subscriber count - Telkom reports Openserve as a wholesale network division, not the retail arm separately. What it does give is the underlying network's growth: how many homes Openserve's fibre now reaches, how many of those homes actually pay for a connection, and how the money moved. Read next to Openserve's public defence of the ISP launch three weeks ago - built around a 21%-of-homes-passed market-share figure - these results are the first independent, audited-adjacent data point against which that defence can start to be checked.

Here's the full set, straight from the trading update, for the quarter ended 30 June 2026:

The external-revenue figure is worth sitting with for a moment: it's the money Openserve earned specifically from the wholesale ISPs that resell its network - Afrihost, Webafrica, RSAweb, Cool Ideas and the rest - as distinct from internal Telkom Group traffic. That grew 18.2%, more than three times the 5.6% growth in Openserve's total revenue. Whatever is happening with Openserve's own retail push, the wholesale channel that every reseller depends on is still the faster-growing piece of the business by a wide margin.

Homes passed and homes connected are two different questions. Passing a home means Openserve's cable reaches the street; connecting it means a household actually signed up. The gap between the two - what we call the connectivity rate - is the number that tells you whether a network is winning customers in areas it already covers, or just running more cable without closing the sale.

Prior-year figures are back-calculated from Telkom's published year-on-year growth rates (+10.7% homes passed, +16.6% homes connected), not independently reported by Telkom this quarter - shown as approximate for that reason.

A roughly 2.7-percentage-point improvement in one year is a real shift, not noise - Openserve's connectivity rate had been broadly flat for several years before this. Whether that's the ISP-launch marketing push moving the needle, general market growth, or both isn't something the trading update can settle on its own; it's a number worth watching again next quarter once there's more than one data point to compare against.

Growth figures are only half the picture if the network doesn't stay up. Telkom also reported availability across three layers of the Openserve network for the quarter:

These are Openserve's own figures, not an independent measurement - Ispa's twice-yearly ISP survey and Ookla's crowdsourced Speed Score, both on our benchmarks page, are the closer thing to a third-party check on network quality. Still, 99.9%-plus availability across all three layers is consistent with Openserve's reputation among resellers for reliability, and gives useful context if you're comparing it against a newer or smaller network where uptime data is harder to find.

None of these figures directly settle the fairness dispute that's been running since mid-July - the trading update doesn't disclose Openserve ISP's own subscriber count, so there's no way to check Openserve's "roughly 21% of homes passed" claim against this release specifically. What the update does add is timing and scale context for a dispute that's otherwise been argued in press statements:

The practical read: if you're following the fairness question, this release is a "watch the next one" data point rather than an answer. Ispa said in July it was canvassing members on a possible Competition Commission complaint; whether that goes anywhere will likely matter more to how this plays out than any single quarter's homes-passed number.

Openserve is the largest network by revenue, but not the largest by homes passed - Vumatel still leads on that measure, on the last disclosed figures for each network:

Source: Telkom Q1 FY2027 trading update (Openserve, 3 Aug 2026); Remgro FY2025 results (Vumatel); ITWeb and TeleGeography (others) - full sourcing and the interactive chart are on our benchmarks page, updated with these same figures.

Vumatel's homes-passed figure is from March 2025 rather than this quarter, so it's not a clean apples-to-apples comparison - but on the numbers as disclosed, Openserve is closing the gap: its homes-passed count has grown roughly 10.7% in the past year alone, while Vumatel's most recent disclosure predates its Herotel acquisition being fully reflected.

These results don't change what's on offer today - Openserve ISP's packages, and every reseller's pricing on the same network, are unchanged by a trading update. What they do give you is context for three practical decisions:

• If you're waiting for Openserve to reach your address, the pace of homes-passed growth (+10.7% in a year) is a reasonable proxy for how quickly the network is expanding generally - check current coverage on our fibre coverage map rather than assuming last year's boundary still applies.
• If you're comparing Openserve ISP against an established reseller, this release doesn't tell you anything new about pricing or fairness - see our breakdown of that dispute and compare live pricing on our Openserve fibre deals page before deciding.
• If network reliability matters to you as much as price, this quarter's 99.9%-plus availability figures are a reasonable data point in Openserve's favour, though they're self-reported - cross-check against Ispa's independent ISP survey and Ookla's Speed Score on our benchmarks page before treating them as the full picture.

We'll revisit this when Openserve or Telkom next discloses Openserve ISP's subscriber numbers specifically, or when Q2 FY2027 results give a second data point on whether the connectivity-rate jump holds.