Liquid Kenya and the Enterprise Fibre Margin

Liquid Telecommunications Kenya Limited is economically interesting for what it is not. It is not a SIM-scale consumer telco, not a beachhead submarine-cable owner in Kenya, and not a volume-driven residential broadband leader. The public record instead points to a Kenyan operating company inside a pan-African carrier stack whose local advantage is built around enterprise fibre, wholesale transit, cloud adjacency, government-grade uptime, and cross-border route control. The core economic question is therefore not “How many subscribers does Liquid Kenya have?” but “How much gross margin can a Kenyan enterprise network capture when it owns or controls metro access, cross-border backhaul, fault management, and cloud on-ramps better than rivals?” Public evidence suggests the answer is: enough to matter strategically, but not because of local scale alone. Liquid Kenya appears to earn its place by being a high-value node in a larger continental network rather than by dominating the Kenyan access market in its own right.

The company’s current public identity is unusually clear on the basics and unusually opaque on local economics. Its current contract documentation identifies the legal entity as Liquid Telecommunications Kenya Limited, trading as Liquid Intelligent Technologies, with company registration number C.41705 and Kenyan governing law. Liquid’s Kenya office page describes it as a “full-service data communications carrier” offering IP, broadband transport, infrastructure services and co-location from Sameer Business Park in Nairobi, and names current public-facing local leaders including Neeraj Pradhan, Judy Njeru and Christopher Mwangi. At the group level, audited 2025 financial statements list Liquid Telecommunications Kenya Limited as an active telecommunications subsidiary with its country of incorporation and principal place of business in Kenya; the holdco says it has 100% economic interest in the Kenyan subsidiary even though the table shows a 79.99% holding structure. AFRINIC’s 2025 voter register further shows Liquid Telecommunications Kenya Limited with Mathew Chigwende listed as designated voter and Chief Technology Officer, which is a useful public-responsibility signal even if it is not a full directors register.

That identity matters because it frames the economic model correctly. The Kenyan company shows up in the Communications Authority of Kenya’s May 2026 Unified Licensing Framework register as an International Gateway Operator, Network Facilities Provider Tier 2, Application Service Provider, and Content Service Provider. It does not appear in the same register’s Submarine Cable Landing Rights Operators list, and it does not appear in the Network Facilities Provider Tier 1 list alongside the mobile incumbents and nationwide top-tier access owners. That is the signature of an enterprise-carrier subsidiary: globally connected, locally licensed, but economically optimized around managed routes and enterprise services rather than beach manholes, spectrum scale, or mass-market radio access.

The market-share data makes the same point more brutally. In Kenya’s April–June 2025 regulator data, Liquid Telecommunications Kenya had 16,366 fixed data subscriptions, equal to 0.8% market share. In the same period Kenya had 76.69 million active mobile SIM subscriptions and 58.58 million mobile data subscriptions. On a simple ratio, Liquid’s fixed lines equal roughly 0.02% of the country’s active SIM base. Safaricom alone had 735,749 fixed-data subscriptions, Jamii Telecommunications had 442,076, Wananchi 271,822, and Poa 268,554. This is why treating Liquid Kenya as a consumer broadband challenger is analytically wrong. It is a small access player sitting in a very different, higher-ARPU layer of the stack.

The trend line reinforces the thesis. In June 2020, Liquid Kenya had 9,444 fixed-data subscriptions and 1.5% market share. By late 2024 it had 16,232 subscriptions and 0.9% share; by mid-2025 it had 16,366 and 0.8%. So the base grew about 73% over five years, but market share fell by about 47% because Kenya’s fixed market expanded far faster than Liquid’s line count. That pattern is economically consistent with a company prioritising enterprise and wholesale yield over consumer acquisition volume. If Liquid were trying to win the retail fixed war, this is the wrong output. If it were trying to harvest enterprise margin from a selective customer base while the residential market commoditised around larger access brands, this is exactly the output one would expect. That is an inference from the numbers, not a disclosed management statement, but it is the cleanest explanation the public data supports.

The company Kenya sees is a route-and-uplink business

Once Liquid Kenya is recognized as an enterprise-carrier subsidiary rather than a consumer ISP, the route map becomes the business model. Liquid’s public record in East Africa is dominated by fibre-route announcements, not subscriber campaigns. In 2017 the group upgraded the East Africa Fibre Ring to 100G, explicitly linking Nairobi and Mombasa in Kenya with Kampala, Tororo and Kigali, and positioning the ring as a fully redundant regional loop with automatic rerouting around cuts and outages. In 2024 it announced an upgrade of the 1,300 km Mombasa–Busia route with additional multi-terabit capacity and resilience. In 2023 it launched a terrestrial Mombasa-to-Johannesburg route through southern and central Africa, and in the same year announced a Kenya–Ethiopia cross-border fibre route that it said gives Ethiopian businesses access to data centres and cloud in Nairobi without traffic needing to leave the continent. In 2021 it launched a Mombasa-to-Muanda route across the DRC that it said reduced east–west latency by 20 milliseconds. None of those announcements prove Kenya-specific profitability. They do prove what asset Liquid is selling: route optionality.

That route optionality is where enterprise fibre margin originates. On a metro or national fibre network, the expensive part is not the extra megabit; it is the right-of-way, trenching or duct lease, building entry, power, restoration, and fault-management capability that make the route credible to a bank, government agency, hospital, or hyperscaler. Once the route exists, incremental capacity additions through DWDM, port upgrades, and wavelength sales can be cheap relative to the sunk civil cost. The economic payoff therefore goes to the operator that can keep routes utilised, resilient, and commercially attached to high-value customers. Liquid’s Kenyan public story is almost entirely aligned with that model. The route upgrades do not sound like consumer access investments; they sound like enterprise backhaul monetisation.

The KETRACO partnership is one of the clearest public clues about cost structure. In October 2017, Liquid Kenya announced a 10-year partnership with the Kenya Electricity Transmission Company to operate KETRACO’s optical ground wire fibre. KETRACO said the arrangement would begin by upgrading fibre connections along high-voltage lines and then extend into remote Kenyan areas and neighboring countries including Ethiopia, South Sudan, Uganda, Tanzania, Rwanda, eastern Congo and Burundi. KETRACO’s 2017/18 annual report said the contract was on a revenue-sharing arrangement and would expand Liquid’s network while adding resilience. Liquid’s 2018/19 Kenya sustainability report then stated that long-term leases of this kind save about 30% of the cost of building fibre networks. Economically, that is decisive. It means Liquid can convert utility right-of-way into lower-cost backhaul rather than overbuilding every kilometre from scratch. That is a margin weapon.

KETRACO’s own 2023 newsletter provides another useful signal: it states that Liquid Telecom had leased fibre optic on the Rabai–Isinya–Embakasi transmission line. That is commercially meaningful because Rabai sits in the Mombasa-coast corridor, and Embakasi is Nairobi-side aggregation territory. In other words, the utility-fibre strategy is not abstract. It appears in identified transmission corridors that matter for moving imported subsea capacity inland to Nairobi and then further to upcountry or cross-border routes. The public record does not show the economics of those leases. It does show how Liquid can convert non-telecom infrastructure into telecom margin.

The Mombasa side is even more revealing. The CA register shows Liquid is an international gateway operator but not a submarine landing-rights operator. Yet Liquid announced in 2022 that it had partnered with PEACE Cable Company to introduce 800Gbps of additional subsea capacity in Mombasa. PeeringDB also shows AS30844 present at iColo Mombasa One, iColo Mombasa Two, and the SEACOM Mombasa CLS, in addition to Nairobi interconnection sites. Economically, this means Liquid Kenya’s international edge is based less on owning Kenyan beach infrastructure than on buying, landing, cross-connecting, and rerouting capacity efficiently through Mombasa. That is a lower-control, lower-capex model than owning the landing rights outright. It can be highly profitable if route resale and enterprise attach rates are strong, but it also means upstream dependency is structurally higher.

This distinction matters because it shapes margin ceilings. A landing-rights owner can capture more of the access rent at the shore. Liquid, by contrast, appears to capture value in the inland network: moving capacity from cable systems into metro fabrics, private lines, data centres, and cross-border routes. That often produces better returns on capital than retail last-mile expansion, because the customer is buying not raw bandwidth but a mission-critical path with service controls. But it also means Liquid’s gross margin depends on keeping the inland route more valuable than the commodity international bitstream feeding it. The company’s public behavior—route upgrades, cloud interconnects, colocation, government and enterprise case studies—suggests management understands that perfectly.

Metro fibre margin is won in buildings, not in press releases

The least glamorous part of enterprise fibre is where most of the margin is actually defended: building entry, service handover, SLA compliance, and fault discipline. Liquid’s own Kenyan Master Services Agreement is unusually revealing here. The agreement defines NRC as non-recurring installation and related fees and MRC as monthly recurring charges. It says order forms roll into successive 12-month renewal terms unless notice is given, allows Liquid to continue invoicing MRCs and minimum usage charges during suspension, and imposes explicit early termination fees: if a service is cancelled with 12 months or less remaining, the customer may owe 50% of fees for the remaining initial term; if more than 12 months remain, the customer may owe 50% for the first 12 months and 25% thereafter. The contract also says the service commencement date can follow Liquid’s successful testing at the demarcation point and gives the customer only a short window to reject for demonstrable failures. These are textbook enterprise-fibre switching-cost mechanics.

The same document also quietly shows why enterprise service delivery is operationally difficult. It notes that provisioning can be delayed because buildings are not ready, because power or space is unavailable, because customer migration from another provider is incomplete, or because change-control approval for current services is pending. That is commercially important. Enterprise fibre is not sold like prepaid data bundles. Every building is a mini-project with civil works, landlord approvals, downtime windows, demarcation tests, and migration sequencing. The operator that manages those frictions with the fewest escalations wins durability, not just the order.

Liquid’s enterprise support process is also visible publicly. Its support page lists Kenya escalation contacts, requires calls to go through an enterprise service desk, targets ticket assignment within 15 minutes, regular status updates every three hours, and root-cause analysis reports on request in less than 10 business days. It also reserves the right to charge customers for false-fault dispatches if the issue was not on Liquid’s own network or service infrastructure. That combination—fast ticketing, formal RCA, and contractual demarcation of blame—is classic enterprise-carrier behavior. It does not prove best-in-class operational performance. It does prove that Liquid is designed for a world where fault governance matters commercially.

The NOC footprint fits that picture. Liquid’s 2017/18 offering memorandum said the group’s primary network operating centres were located in Harare, Nairobi, and Johannesburg. Its later 2021/22 financing memorandum also surfaced “NOC” references in search results, consistent with that architecture. A Nairobi NOC is not just overhead; it is a monetisable asset when selling multi-site service assurance into East Africa. The enterprise customer does not pay a premium because the fibre exists. They pay because someone can isolate a fault fast, reroute traffic, and produce an RCA before a CIO meeting becomes a procurement event.

Liquid’s own customer stories show what that premium actually buys. For Aga Khan Hospital Mombasa, Liquid sold a 99.99% uptime guaranteed connection and a 24Mbps high-speed link connecting the main hospital to outreach clinics, enabling telemedicine and video-based consultations. For KenTrade, Liquid provided an MPLS management link, a hosted environment including servers, storage, links and security equipment, plus redundant point-to-point links between primary and secondary sites for replication, failover and high availability. For Unga Group, Liquid’s value was not simply “internet.” Unga had suffered frequent cable cuts and connectivity outages at a plant in Nakuru that could take databases and connectivity offline for up to three days. Liquid’s answer was colocation in an ADC Tier III facility plus Office 365 migration, with the case study claiming ROI in less than a year from better continuity, security and productivity. These are not consumer use cases. They are uptime use cases.

That matters because uptime contracts are where enterprise fibre resists commoditisation. A residential customer may switch providers for a lower monthly bill or a promotional router. A hospital, trade platform, bank, port user, or manufacturer with replicated workloads, site-to-site private lines, and compliance exposure switches far less easily. Once the provider also supplies colocation, cloud connectivity, collaboration tools and security, the cost of changing is not just the early-termination fee in the contract. It is project risk. That is why enterprise fibre margins can remain attractive even when access bandwidth becomes cheap. The hard thing to replace is not bandwidth. It is a trusted operational stack.

Cloud adjacency is the real upsell

If route control is the base business, cloud adjacency is the margin expansion layer. Liquid’s Kenyan public record shows a long-running attempt to push beyond transit into colocation, cloud access and managed services. In 2017, East Africa Data Centre in Nairobi received Tier III certification from Uptime Institute, which Liquid described at the time as making it the first such certified facility in East and Central Africa. In 2018, Africa Data Centres opened an additional floor at the Nairobi facility with 500 square metres of additional rack space. The group later continued to describe East Africa Data Centre Limited as a related company under common control, and customer stories repeatedly position ADC’s Nairobi facility as the place where Liquid can move clients from on-premise fragility into carrier-neutral, better-powered environments.

This matters because metro fibre alone tends toward commodity economics unless it is attached to scarce destinations. Data centres are scarce destinations. So are cloud on-ramps. Liquid’s Kenya record shows both. The 2018 Strathmore Business School analytics-centre partnership said Liquid would provide dedicated rack space and colocation in East Africa Data Centre and a direct fibre link between the SBS campus and EADC. The 2018/19 Kenya sustainability report says Liquid provides colocation and cloud services through its own carrier-neutral EADC in Nairobi. These are small details on the surface, but they reveal the logic: once the carrier controls both the route and the destination, it can charge for end-to-end service quality rather than for transport alone.

The public interconnection footprint reinforces the same point. PeeringDB shows AS30844 present in Kenya at Africa Data Centres Nairobi NBO1, PAIX Nairobi, iXAfrica NBOX1, iColo Nairobi One, iColo Mombasa One, iColo Mombasa Two, and the SEACOM Mombasa CLS. The NBO1 facility page also lists cloud and platform names such as Amazon, Microsoft, Netflix and Huawei Cloud among networks present there, alongside Liquid. That does not prove Liquid has a commercial interconnect contract with every platform present in every facility. It does prove Liquid sits in the buildings where such cross-connect economics become possible. For enterprise carriers, that adjacency is the difference between being a bandwidth wholesaler and being a cloud-access broker.

Liquid’s cloud products make the monetisation path explicit. The Kenya-local Azure Stack Hub deployment in Nairobi was pitched around local data-regulatory requirements and support for latency-sensitive applications. Liquid later secured AWS Direct Connect Delivery Partner status, allowing private connectivity to AWS that bypasses the public internet. It has marketed Microsoft Azure Peering Service as a low-latency, high-reliability route into Microsoft SaaS, and in 2024 Liquid C2 became Africa’s first Google Cloud Interconnect provider. In June 2026, Liquid C2 said it had achieved gold-level status in Google’s Verified Peering Provider programme. These are partly group-level capabilities rather than Kenya-only disclosures, but they are still economically relevant to Liquid Kenya because the Kenyan unit is one of the group’s enterprise access and interconnection nodes.

Cloud adjacency raises enterprise fibre margin in three ways. First, it lifts ARPU because the customer now buys private connectivity, security, or hybrid-cloud architecture instead of just internet access. Second, it reduces churn because cloud migration increases business-process dependence on the carrier’s network path. Third, it improves price defence because the product shifts from “Mbps delivered” to “latency, resilience, sovereignty, and security delivered.” You can see this in Liquid’s own white paper on cloud connectivity, which frames private cloud access as more reliable, lower-latency and more secure than public internet paths. You can also see it in customer outcomes like Unga, where the gain came from colocating servers and moving collaboration workloads, not from buying a bigger commodity link.

There is also a group-revenue clue here. In FY2025, audited group revenue totaled $693.5 million. Of that, the group classified roughly 70.5% as Network, 16.6% as C2 cloud/cyber/managed services, 5.2% as Dataport undersea and international wholesale, and 7.7% as voice traffic. That is not Kenya segmentation, so it should not be read as a Kenya P&L. But it does show where Liquid, as a group, thinks the real revenue engines sit: enterprise and network services first, then cloud/cyber adjacency, with wholesale submarine and international business important but not dominant in reported external revenue. Liquid Kenya’s local asset mix—enterprise access, colocation, cloud services, cross-border routing—fits that architecture almost exactly.

The competitive position is narrow but real

Liquid Kenya’s weakness is obvious from regulator tables: it is tiny in subscriber terms. Liquid Kenya’s strength is that the regulator tables measure the wrong thing for its best business. In fixed subscriptions, it sits beneath Safaricom, Jamii, Wananchi, Poa, Ahadi, Vilcom, Mawingu, and roughly around the same scale as Starlink and Dimension Data. In pure access-line terms, that is not impressive. But the company’s public product mix is not primarily fibre-to-the-home. It is IP transit, private leased line, Metro Access Circuit, DIA, cloud connectivity, colocation, cybersecurity, and cross-border Ethernet. Liquid’s wholesale pages explicitly target ISPs, carriers and content providers, promising scalable pricing per Mbps on metro access, highly available private leased circuits, and direct reach into exchanges and cloud services. That is a different competitive arena.

The interconnection telemetry suggests the wholesale position is serious, not ornamental. PeeringDB shows AS30844 with an open peering policy and presence across multiple African and European exchanges, including KIXP in Nairobi. On KIXP, PeeringDB shows Liquid at 20G; bgp.tools currently shows 40 Gbps at KIXP and says AS30844 peers with about 1,733 networks and has 19 upstream carriers. Hurricane Electric lists 25 internet exchanges and a long list of major upstreams, including Level 3, Tata, Zayo, PCCW, Cogent, GTT, Airtel, Hurricane Electric, Telstra, Sparkle and Orange. Those third-party measurements are market signals, not audited traffic figures, and they do not prove Kenyan revenue or local port utilisation. But they do strongly support the claim that Liquid’s network is engineered as a serious peering and transit platform rather than a provincial access network.

That in turn helps explain why a small Kenyan access base can still support useful margins. Liquid can win where buyers value one or more of the following more than a lower monthly port price: redundant cross-border routes; direct or near-direct cloud access; enterprise-grade escalation; local colocation with private connectivity; government procurement credibility; and bundled cyber/cloud/voice services. Its public customer references cut across exactly those themes: county governments for public Wi-Fi, Office of the President-linked connectivity projects, KenTrade for redundant hosted infrastructure, hospitals for telemedicine, Strathmore for analytics hosting, Twiga for IoT, and Unga for business continuity. Again, these references do not disclose contract size or current status in every case. But they show that Liquid’s sales motion is enterprise and institutional, not promotional retail.

The biggest competitive threat is not that Liquid lacks a business. It is that the easiest parts of that business are getting crowded while the hardest parts require capital and balance-sheet patience. On the access side, Kenya’s fixed-data market grew from 609,611 subscriptions in mid-2020 to over 2.14 million subscription lines across narrowband and broadband speed tiers by mid-2025, a roughly 3.5x increase. Liquid’s line count rose far more slowly, which means it is not taking share in the volume segments. Starlink, which barely existed in the Kenyan fixed rankings a short time ago, already matched Liquid’s 0.8% share in CA’s mid-2025 fixed-data table. Safaricom and Jamii remain far larger. If enterprise access in Kenya were to become heavily price-led and less SLA-led, Liquid’s narrow local base would be a problem.

The counterargument is that enterprise fibre is still a scarcity market, not a pure scale market. There are many operators that can sell bandwidth into a Nairobi office park. There are fewer that can sell a properly restored path from an office in Nairobi to a replicated environment in a carrier-grade data centre, then onward via private cloud access, with cross-border redundancy to Uganda or Ethiopia and a formal RCA if something fails. Liquid’s public footprint suggests it is in that smaller club. The company looks economically weaker than consumer champions if you compare subscriber counts, and stronger than those same champions if you compare route richness, facility presence, product breadth, and regional peering depth. That is why the right category for Liquid Kenya is not “small ISP.” It is “enterprise infrastructure node inside a continental carrier.”

What the evidence proves and what it does not

The public record proves several things with high confidence. It proves that Liquid Telecommunications Kenya Limited is a Kenyan telecommunications operating entity trading as Liquid Intelligent Technologies; that it sits inside the Liquid/Cassava group; that it holds Kenyan licences suited to international gateway, tier-2 facilities and application/content services; that it is not listed in the CA’s submarine landing-rights register; that Liquid’s Kenyan business is tied to KETRACO utility-fibre access, Mombasa subsea capacity partnerships, local data-centre adjacency, and group-level cloud interconnect products; and that it sells uptime-sensitive solutions to enterprises, public entities and carriers. It does not prove Kenya-specific EBITDA, local capex intensity, contract concentration, customer retention rate, or the precise split of Kenyan revenue between access, wholesale and cloud-linked services. Those figures are not publicly disclosed in the sources reviewed here.

There are also significant financial caveats. The group’s 2025 audited statements say there is a material uncertainty around the refinancing project because, without completion, the group would not be able to repay its bond at maturity in September 2026. The 2024 and 2025 records also show that Liquid Telecommunications Kenya Limited is one of the guarantors of the group’s senior secured notes and revolving credit facility. Separately, the 2024 statements say the Kenyan subsidiary had $23.8 million of tax losses for which no deferred tax asset was recognised, and the holdco also carried a $66.7 million long-term intercompany receivable from the Kenyan subsidiary, unsecured at SOFR + 3.75%, repayable in February 2026. None of that means the local operation is distressed. It does mean Kenya should be analysed as part of a leveraged continental structure rather than as a clean standalone utility. That is a real commercial risk, especially when enterprise customers care about long-term service continuity.

There is also a network-identity nuance worth keeping straight. AFRINIC’s membership list includes Liquid Telecommunications Kenya Limited in Kenya, and the historical public trail associates the entity with the region’s internet-numbering system. But current public peering and BGP visibility are centered on AS30844, branded at various times as Liquid Telecommunications Ltd or Liquid Intelligent Technologies, rather than on a clearly visible Kenya-specific public ASN as the main interconnection face. Third-party telemetry on Hurricane Electric even labels one prefix description as “LIT Kenya,” which suggests Kenyan traffic engineering inside the shared AS30844 fabric. Commercially, that implies the Kenya subsidiary likely benefits from being integrated into a larger group backbone rather than standing alone in routing policy. What it does not prove is that every Kenyan customer path is carried only on AS30844 or that the old AFRINIC resource history is economically irrelevant. It is a network-structure signal, not a complete map.

The clearest commercial judgement, then, is this: Liquid Kenya is strategically better than it looks on subscriber tables and structurally narrower than it sounds in corporate branding. It is not the Kenyan mass market. It is a Kenyan enterprise and wholesale bridgehead. Its margin depends on whether it can keep selling scarce things—resilient route combinations, cloud-adjacent connectivity, and corporate uptime—faster than those things are commoditised by larger access brands, LEO alternatives, and hyperscaler direct presence. Public evidence supports a positive but disciplined view: Liquid Kenya looks like a high-quality enterprise fibre and cloud-adjacent infrastructure node with moderate local scale, meaningful route advantages, and material parent-structure dependence. That is a good business category. It is not the same as a dominant one.

Evidence ledger

AFRINIC Membership List URL: https://www.afrinic.net/nomcom?catid=39&id=2230%3Aafrinic-membership-list-all&view=article Source type: RIR membership directory. What it supports: Liquid Telecommunications Kenya Limited appears as an AFRINIC-linked organisation in Kenya. What it does not prove: It does not prove current headquarters, financial health, or commercial scale. Why it matters economically: It confirms the company is a real network-resource participant, which is a basic prerequisite for carrier economics.

AFRINIC Voter Register 2025 URL: https://election.afrinic.net/home/election-2025/about-election-2025/voters-register?start=280 Source type: RIR election register. What it supports: Liquid Telecommunications Kenya Limited had Mathew Chigwende publicly listed as designated voter and CTO in 2025. What it does not prove: It does not prove statutory directorship or local management hierarchy beyond this registration context. Why it matters economically: It gives a public accountability marker for responsible technical leadership.

Liquid Kenya Master Services Agreement 2025 URL: https://liquid.tech/wp-content/uploads/2025/02/Liquid-Intelligent-Technologies-Kenya-Master-Services-Agreement-2025.pdf Source type: Company contract document. What it supports: Legal entity name, trading name, registration number C.41705, Kenyan law, NRC/MRC structure, renewal terms, early termination charges, SLA framing and service handover mechanics. What it does not prove: It does not prove how often these clauses are enforced or realised in revenue. Why it matters economically: It shows directly how switching costs and recurring revenue are contractually engineered.

Communications Authority of Kenya ULF License Register 2026 URL: https://www.ca.go.ke/sites/default/files/2026-05/REGISTER%20OF%20UNIFIED%20LICENSING%20FRAMEWORK%20LICENSEES.pdf Source type: Regulator register. What it supports: Liquid Kenya appears as International Gateway Operator, Network Facilities Provider Tier 2, Application Service Provider and Content Service Provider; it does not appear in the same document’s submarine cable landing rights list or NFP-T1 list. What it does not prove: It does not prove absence of every other regulatory approval outside this published register. Why it matters economically: The licence stack identifies the company’s position in the value chain and its likely margin pool.

CA Sector Statistics Q4 2024/25 URL: https://www.ca.go.ke/sites/default/files/2025-09/Sector%20Statistics%20Report%20Q4%202024-2025_1.pdf Source type: Regulator market report. What it supports: Liquid Kenya had 16,366 fixed-data subscriptions and 0.8% share; Kenya had 76.69 million SIMs; fixed sub leaders were Safaricom and Jamii; international bandwidth data shows national subsea capacity composition. What it does not prove: It does not separate enterprise from residential lines within Liquid’s 16,366 subscriptions. Why it matters economically: It proves Liquid is not a SIM-scale or volume-fixed player and sets the competitive baseline.

CA Sector Statistics Q4 2019/20 and Q2 2024/25 URLs: https://www.ca.go.ke/sites/default/files/2024-03/Sector%20Statistics%20Report%20Q4%202019-2020_0.pdf and https://www.ca.go.ke/sites/default/files/2025-03/Sector%20Statistics%20Report%20Q2%202024-2025.pdf Source type: Regulator market reports. What they support: Liquid Kenya moved from 9,444 fixed subscriptions and 1.5% share in 2020 to 16,232 and 0.9% by late 2024; Kenyan fixed-speed tiers show a meaningful high-speed segment above 100 Mbps. What they do not prove: They do not disclose the profitability of those subscriptions. Why they matter economically: They show share dilution amid market expansion, supporting the enterprise-yield thesis over a retail-scale thesis.

Liquid Telecommunications Holdings AFS 2025 URL: https://liquid.tech/wp-content/uploads/2025/10/Liquid-Telecommunications-Holdings-Ltd-AFS-2025.pdf Source type: Audited group financial statements. What it supports: Kenya is an active telecommunications subsidiary; the group reports 100% economic interest; group revenue mix is dominated by Network and C2; the group faces a refinancing-related going-concern uncertainty around September 2026 bonds; East Africa Data Centre is a related company under common control. What it does not prove: It does not disclose Kenya standalone revenue or EBITDA. Why it matters economically: It ties the Kenyan business to group ownership, strategy, and balance-sheet risk.

Liquid Telecommunications Holdings AFS 2024 URL: https://liquid.tech/wp-content/uploads/2024/07/LTH-AFS-2024-v16-Signed-with-AR.pdf Source type: Audited group financial statements. What it supports: Kenya had $23.8 million of tax losses with no deferred tax asset recognised; the parent had a $66.7 million long-term receivable from the Kenya subsidiary; the Kenya subsidiary guarantees group debt. What it does not prove: It does not prove local insolvency or poor operating service quality. Why it matters economically: It shows that local operations sit inside a leveraged internal-capital structure.

KETRACO–Liquid partnership trail URLs: https://liquid.tech/about-us/news/liquid_intelligent_technologies_and_ketraco_partner_to_build_fibre_network_across_kenya_and_east_africa/, https://www.ketraco.co.ke/information-center/media-center/news/liquid-telecom-and-ketraco-partner-build-fibre-network-across, https://www.ketraco.co.ke/sites/default/files/publications/Annual%20Report%202017-2018_0.pdf, https://www.ketraco.co.ke/sites/default/files/publications/KETRACO%20The%20Grid%20-%20Issue%2010.pdf Source type: Company and utility records. What it supports: 10-year OPGW partnership, revenue-sharing model, route expansion into remote/cross-border corridors, and a specific Rabai–Isinya–Embakasi lease signal. What it does not prove: It does not show exact lease economics or current utilisation. Why it matters economically: It explains how Liquid lowers route-build cost and widens coverage without owning every trench.

Route announcements from Liquid URLs: https://liquid.tech/about-us/news/liquid_intelligent_technologies_upgrades_east_africa_fibre_ring_to_100g_delivering_faster_speeds_across_rwanda_uganda_and_kenya/, https://liquid.tech/about-us/news/announces-the-upgrade-of-its-1300km-fibre-route/, https://liquid.tech/about-us/news/liquid-dataport-launches-the-first-terrestrial-data-superhighway-connecting-mombasa-to-johannesburg/, https://liquid.tech/about-us/news/liquid-intelligent-technologies-announces-two-new-cross-border-fibre-routes/, https://liquid.tech/about-us/news/liquid_intelligent_technologies_announces_a_new_asiausa_global_internet_transit_route_via_africa/ Source type: Company network announcements. What they support: 100G East Africa ring, Mombasa–Busia upgrade, Mombasa–Johannesburg route, Kenya–Ethiopia route, and Mombasa–Muanda east–west route. What they do not prove: They do not prove local route profitability or customer fill rates. Why they matter economically: They show the company’s core product is route optionality and resilience.

Cloud and data-centre trail URLs: https://liquid.tech/about-us/news/liquid_intelligent_technologies_accelerates_kenyas_digital_transformation_with_azure_stack_cloud_solutions/, https://liquid.tech/about-us/news/liquid_cloud_brings_amazon_web_service_direct_connect_to_business_customers_across_africa/, https://liquid.tech/about-us/news/liquid-c2-becomes-the-first-provider-in-africa-to-support-hybrid-network-connections-to-google-cloud/, https://liquid.tech/about-us/news/liquid-c2-secures-gold-level-in-googles-verified-peering-provider-programme-providing-improved-cloud-connectivity-in-africa/, https://liquid.tech/about-us/news/east_africa_data_centre_receives_tier_iii_certification/ Source type: Company announcements. What they support: Azure Stack in Nairobi, AWS Direct Connect partnership, Google Cloud Interconnect, Google VPP accreditation, and EADC Tier III pedigree. What they do not prove: They do not disclose Kenya-specific cloud revenue. Why they matter economically: They show how Liquid tries to move from commodity transit into stickier cloud-adjacent revenue.

PeeringDB and BGP telemetry URLs: https://www.peeringdb.com/net/725, https://www.peeringdb.com/ix/236, https://bgp.tools/as/30844, https://bgp.he.net/as30844 Source type: Peering and routing databases. What they support: AS30844 presence at Kenyan facilities and exchanges, KIXP connectivity, broad peering, major upstreams, and multi-facility Kenyan interconnection. What they do not prove: They do not prove traffic volumes, profit margins, or exact port commitments beyond what is publicly listed. Why they matter economically: They are the best reusable public signals that Liquid is selling a serious carrier-grade backbone.

Customer and public-sector case trail URLs: https://liquid.tech/insights/customer-stories/kenya_trade_network_agency_kentrade/, https://liquid.tech/insights/customer-stories/unga_group_customer_story/, https://liquid.tech/insights/customer-stories/aga_khan_hospital_mombasa_providing_specialist_treatment_along_kenyas_coastline/, https://liquid.tech/about-us/news/liquid_intelligent_technologies_and_office_of_the_president_kenya_win_global_telecoms_award_for_wi-fi_projects/ Source type: Company case studies and news. What they support: Redundant hosted links for KenTrade, business-continuity colocation for Unga, 99.99% uptime medical links for Aga Khan, and Office-of-the-President-linked public Wi-Fi work. What they do not prove: They do not prove those contracts are all still active at the same commercial scale today. Why they matter economically: They show the kind of customer problem Liquid solves and the margin logic behind that solution.

Watchpoints

Parent refinancing and local continuity. Track whether the group completes its refinancing before the September 2026 note maturity and whether Kenya remains a guarantor in any amended debt package. This is the single most important non-operational risk to the local business.

Kenya-specific cloud monetisation. Watch for Kenyan references—not just pan-African ones—to Google Cloud Interconnect, AWS Direct Connect, or Azure Peering wins. The platform exists; the open question is how much of the high-margin cloud-connect revenue is booked through Kenya.

Mombasa dependency. Watch for any further PEACE, 2Africa, or other Mombasa capacity announcements, plus any direct evidence that Liquid acquires stronger landing-rights control rather than remaining an inland aggregator. That would change margin structure materially.

KETRACO execution depth. Watch for more named transmission corridors, any renewal or extension of the OPGW partnership, and any evidence of new utility-fibre revenue-sharing terms. The KETRACO route is one of Liquid Kenya’s most plausible cost advantages.

Facility migration and adjacency. Monitor PeeringDB and facility websites for changes in Liquid presence at NBO1, PAIX, iXAfrica, iColo and SEACOM CLS. New ports or facility exits are meaningful because the enterprise margin story depends on adjacency, not just fibres in the ground.

Kenyan fixed-share drift. Keep tracking CA fixed-data tables. If Liquid’s subscriptions stay flat while enterprise stories multiply, that would confirm a deliberate non-retail strategy; if the line count starts falling, it may suggest access compression or customer concentration risk.

Government-account freshness. Many public-sector references are historic. The next real intelligence step is to look for fresh Kenyan procurement records, framework agreements, or tender awards that show Liquid still winning government-grade uptime and hosting work. Current public case studies prove capability more than current-book status.

Kenya standalone profitability signals. Any future filing that updates the Kenyan subsidiary’s tax-loss position, intercompany funding balance, or asset carrying value will matter. A shrinking tax-loss pool and lower intercompany dependence would be constructive; the reverse would indicate that strategic importance has not yet translated into local earnings quality.