Over the past quarter we verified 291 Kenyan M-PESA paybills against official sources. This report summarises what we found about the underlying payments rail: how concentrated it is, which sectors actually publish their numbers, where it’s rapidly consolidating, and where it’s breaking.
Headline numbers
- 291 M-PESA paybill entries verified across 15 sectors.
- 145 fully verified (50%), 132 verified-with-caveat (45%), 12 still under review (4%), 2 flagged as wrong or defunct (1%).
- 32 distinct entities now route their payments through the single eCitizen paybill 222222up from 8 in our previous count. Kenya is in the middle of the fastest single-paybill consolidation we’ve seen.
- 7 universities checked, 5 had wrong numbers in the most widely-shared online directories. Real Kenyan university fees paybills cluster in the 30xxxx range.
- 1 KE insurer (Resolution Insurance) is still listed on most Kenyan paybill directories despite being in court-ordered liquidation since December 2023.
1. Sector concentration
The verified directory breaks down by sector roughly as follows:
- Government46 entries (16%). Skewed heavily by the eCitizen consolidation (see Section 2).
- School / education34 entries (12%). Universities, KMTC colleges, SOPs, professional bodies.
- Bank31 entries (11%). All majors covered; the gap is microfinance.
- Religious28 entries (10%). Churches, mosques, Hindu/Sikh temples, with M-PESA paybills increasingly published in service announcements.
- Lending26 entries (9%). Mobile lenders, asset-finance, SACCO-linked.
- SACCO25 entries (9%). The five largest SACCOs by deposit value.
- Insurance22 entries (8%). General + Life, with a known data-quality issue (see Section 5).
- Hospital19 entries (7%). Heavy bias to private: public hospitals route through eCitizen.
- Betting13 entries (4%). All GRA-licensed operators.
- Utility, retail, telecom, transport, media, othertogether 47 entries (16%).
2. The eCitizen consolidation: Kenya’s most consequential payments story of 2026
When we started the verification effort in 2025, eight unrelated government services routed through paybill 222222 with different account-number suffixes. As of May 2026, we count 32 distinct entities using the same paybill: a four-fold increase. The list now includes:
- KRA (taxes, fines, customs)
- NTSA (driving licence, vehicle inspection)
- Most county-level revenue (Mombasa, Kisumu, Nyeri, Makueni, Embu and others)
- Public referral hospitals (Moi Teaching and Referral, in this quarter)
- SHA (National Hospital Insurance Fund replacement)
- Government professional registrations (KMPDU, EBK, BORAQS, etc.)
- Public university fees in transition (some counties on, most still on legacy numbers)
Why this matters. A single paybill paying multiple government surfaces is operationally useful: one rail, one user mental model. But it concentrates risk: any rate limiting or outage on 222222 hits a wide surface area. We saw this in March 2026 when a 90-minute interruption affected KRA VAT filings, NTSA renewals, and county parking simultaneously. We’d expect Safaricom + Treasury to be working on resilience for this number specifically.
What changes for Kenyans. The account number is now doing more work than the paybill anyone moving public payments should expect to verify the account suffix carefully (a typo on the 222222 account field doesn’t pay the right surface).
3. The reverse-lookup data: most-shared paybill numbers
Outside the eCitizen aggregator, three numbers recur disproportionately across unrelated entities: a directional signal of legacy bank-to-business pooling:
- 247247 (Equity Bank corporate paybill): 5 distinct businesses route via Equity’s shared rail.
- 303030 (Absa-linked aggregator): 5 distinct businesses.
- 220220 (Pesapal payment-service aggregator): 2 distinct merchant clusters.
We treat these as expected, not anomalies: they reflect Pesapal’s growth as a payments aggregator and the Equity/Absa decisions to pool merchant payments on a shared paybill with the account number doing the routing.
4. Sectors where data quality is excellent
Some Kenyan sectors publish their M-PESA paybills clearly and stably:
- Banking. Every major bank publishes its paybill prominently on the relevant product page. We had 0 conflicts across the 31 verified bank entries.
- Telecom. Safaricom, Airtel, Telkom each publish paybill numbers in service guides. We hit 0 disputed entries.
- Major utilities. KPLC (888880, 888888 prepaid) and Nairobi Water are stable. Secondary water companies (county-level) show drift: see Section 5.
- GRA-licensed betting operators. All 13 publish paybills on their landing page within one click. We had 1 conflict (SportyBet, since corrected).
5. Five systemic data-quality clusters we discovered
These are the sectors where the most widely-shared Kenyan paybill directories online were materially wrong: typically because they were sourced from a single bad data drop years ago and never re-verified.
5a. Universities
Of 5 universities we sampled with cross-source verification, 5 out of 5 showed material conflict between our existing entries and the actual numbers in use. Real Kenyan university fee paybills cluster in the 30xxxx range: KCA 300078, KeMU 300112, Zetech 303072. The numbers shared across multiple aggregator-style Kenyan directories (565555, 898220, 898100, 844200) corresponded to nothing we could verify against the universities themselves. We treat the whole university block as suspect data that needs re-sourcing from a clean origin.
5b. Counties (the transitional story)
Kenyan county revenue is fragmented and transitional. No county we sampled reliably publishes a single clean static M-PESA paybill. Larger counties (Mombasa, Kisumu) are migrating to eCitizen (222222); smaller counties point users at the ICRMS Pesaflow aggregator at varying stages of completeness. Any directory listing a "county paybill" as a stable number is almost certainly wrong by 2026 standards. The correct primitive is the county’s eServices portal URL, not a paybill number.
5c. Insurance
Seven insurance companies in our sample carried numbers that appeared nowhere in independent sources. Where we could find a number, it didn’t match what was on the bad directory cluster. Insurance publishes paybills the least openly of any vertical we tracked: most carry payment instructions behind a login or a download form, not in scrapeable HTML. Resolution Insurance is particularly concerning: it’s in court-ordered liquidation since December 2023 but still appears on most Kenyan paybill directories as if it were a working insurer. We’ve flagged it red so users see the warning before paying.
5d. SACCOs
Four major Kenyan SACCOs (Hazina, Sheria, Imarisha, Kenya Defence Forces) had wrong numbers in our original dataset that we’ve since corrected. Sheria’s case was particularly instructive: an upstream directory had 064700 where the correct number is 964700: a single-digit transposition (0 vs 9) that propagated across at least three secondary sources before we caught it.
5e. Hospitals (especially private)
Aga Khan (Mombasa and Kisumu), Avenue Healthcare, Mediheal, Kijabe Hospital, Mombasa Hospital common pattern: paybills used at point-of-service are not published on the hospital website. We could verify Kijabe (paybill 512900 per their FAQ) and Moi Teaching and Referral (now on eCitizen 222222), but most private hospital paybills can only be confirmed against an actual bill. This is the verification gap that needs reader help, not more automated search.
6. Defunct / flagged entities
Two entries are flagged red in the directory:
- Resolution Insurance Companyin court-ordered liquidation since December 2023. Claims are processed by the appointed liquidator, not the company. The fact that this entity still appears on most Kenyan paybill aggregator sites suggests how rarely these directories get re-verified.
- Kiambu County (legacy)the original placeholder number 123456 has been flagged because it’s clearly a developer-test value that propagated into a "real" directory. Kiambu now routes through ICRMS / eCitizen depending on revenue type.
We’d expect 5-10 more "defunct entity" flags as we revisit the long tail of small Kenyan institutions that appeared in 2018-2020 paybill data drops.
7. What we’d like to see fixed
Three recommendations the underlying data points toward:
- Safaricom should publish the canonical M-PESA paybill registry. Right now this is verifiable only through direct query to a Safaricom relationship manager or by paying a small amount and reading the SMS confirmation. A read-only API exposing “number → entity name + sector” (omitting account-format details) would eliminate the secondary-directory drift entirely.
- Universities should standardise their fee paybill publication. The pattern of "ask the bursar" produces drift; a one-click "Pay fees" landing page would solve it.
- Insurance carriers should publish premium paybills openly. The "hidden behind login" pattern serves no one and creates room for scam-paybill social engineering.
Methodology and access
Each verified entry in our directory carries at least one source URL and a retrieval timestamp. The methodology has three tiers:
- Verified (green check, 145 entries): confirmed against an official source, the company’s own website, a regulated body’s gazette, a verified social account, or a publicly-confirmed bill scan.
- Verified-with-caveat (amber, 132 entries): consistent across multiple high-quality secondary sources, but not asserted on a scrapeable official page. Treated as provisional pending official confirmation.
- Needs-review (grey, 12 entries): residual long-tail where we have a number and our secondary directories agreed, but cross-source verification was inconclusive.
- Flagged (red, 2 entries): actively wrong or pointing to a defunct entity.
For the underlying CSV (sector, slug, primary number, secondary networks, verification tier, sources, last-verified date), email interconpublishers@gmail.com. Journalists and researchers are welcome to cite the report and any of the figures above without attribution gymnastics — just please link back to the source URL.
Common questions
How was the verification done?
A mix of direct browser scraping of official company sites, cross-referencing across 6+ independent directories, conflict-detection (any number appearing in <2 sources got flagged), duplicate-number detection (a single number shared by unrelated entities is a strong wrongness signal), and human verification against bills/statements where available. Every verified entry carries at least one URL-and-retrieval-date source. The 132 "verified-with-caveat" entries are provisional: supported by multiple secondary sources but not by a scrapeable official assertion.
Why are some numbers marked "provisional"?
Kenyan institutions vary wildly in how openly they publish their paybills in scrapeable HTML. KPLC, KRA, and Safaricom publish prominently. Insurance companies, SACCOs, and most hospitals don't. Where we have a number that's consistent across multiple high-quality secondary sources (industry directories, official social posts) but not asserted on the company's own scrapeable webpage, we mark it provisional with an amber check. It signals "we believe this, but the official site doesn't make it easy to verify."
Is this report free to cite?
Yes. Cite as: "paybillke, Q2 2026 State of Kenyan Paybill Payments, May 2026, https://paybillke.com/guides/paybillke-q2-2026-state-of-kenya-payments". For interview requests, methodology questions, or access to the underlying CSV, email interconpublishers@gmail.com.
Related
- The full verified paybill directory (291 entries)
- Capital Deployment KenyaKSh 20M to where, modelled across 9 strategies
- Debate Modetwo-strategy comparison artifact
- Our editorial process
- Our verification policy