JuaSync — reverse-invoice clerk for Kenyan SMB buyers

1. One-liner

Turns informal-supplier receipts into KRA-compliant reverse invoices and chases supplier sign-off so Kenyan SMBs keep every deduction.

2. Trend signal — why now?

Three things landed in the last six months and they all point at the same buyer-side pain.

The rule flipped on January 1, 2026. KRA now validates every income and expense line in a tax return directly against eTIMS data. Under Income Tax Act Section 16(1)(c), any expense not backed by a valid eTIMS invoice is disallowed — “added back” to your profit and taxed. You spend the cash, then you pay tax on it like it was income. A small furniture maker in Gikomba buying timber from informal suppliers faces losing legitimate deductions outright.

Reverse-invoicing guidelines were published mid-2026. KRA’s escape hatch: the buyer raises the invoice on behalf of a supplier who can’t (informal traders, jua kali, farmers, sub-KSh-5M-turnover vendors). Practically, the buyer keys the transaction into eCitizen (7 steps), the supplier approves via eCitizen or USSD *222# (5 steps), and the result becomes the compliant tax document. Per transaction. By hand. Manufacturers have reportedly hired up to 100 staff just to reconcile mismatched invoices.

KRA softened — but didn’t solve. In May–June 2026 KRA made the supplier-PIN field optional in iTax and opened a one-time window to declare non-eTIMS expenses via a manual CSV upload. That’s relief, not a fix — the CSV is hand-built, the window is transitional, and the permanent compliant path for ongoing B2B with informal suppliers is still reverse invoicing. Existing software is all seller-side POS (Risiti KSh 1,000/mo, CuteProfit KSh 2,000–10,000/mo). Nobody is purpose-built for the buyer chasing reverse invoices.

Provenance:

• Signal 1 (demand): KRA validates every 2026 return line against eTIMS; expenses without a valid eTIMS invoice are disallowed and taxed as profit — direct, recurring buyer pain. — https://kenyammfcalculator.co.ke/kra-etims-merchant-tax-compliance-certificate-expense-validation-2026/ — 2026-06
• Signal 2 (feasibility): KRA published reverse-invoicing / buyer-initiated invoicing guidelines and the eCitizen + USSD *222# approval flow, defining a concrete process to build against. — https://www.vatupdate.com/2026/06/07/kenya-introduces-reverse-invoicing-under-etims-expanding-control-over-supplier-compliance/ — 2026-06-07
• Signal 3 (economic): An eTIMS software market already exists and charges money (Risiti KSh 1,000/mo; CuteProfit KSh 2,000–10,000/mo; ERPs from KSh 260,000) — money is moving into eTIMS tooling, all of it seller-side. — https://getrisiti.com/blog/etims-for-small-business-kenya — 2026 Category: Regulatory arbitrage

3. The opportunity

The whole eTIMS software market is built for the seller — issue your invoice, print a QR, ring it through a POS. But the rule that actually destroys money in 2026 hits the buyer: your deduction dies if your supplier didn’t issue a compliant invoice, and a huge slice of Kenyan SMB supply chains is informal — jua kali artisans, farmers, roadside vendors, matatu operators, small sub-contractors who have no PIN, no smartphone, no will to learn eCitizen.

The “incumbent” here is the manual reverse-invoicing process itself — 12 keystroke-heavy steps split across buyer and supplier, per transaction, run at month-end by an overworked bookkeeper, rejected when a PIN or VAT flag is wrong. An AI-first tool collapses that into: snap the receipt → draft the reverse invoice → auto-chase the supplier’s approval over SMS/USSD → roll the lot into a clean month-end package (or the non-eTIMS CSV where reverse invoicing isn’t viable). Same compliance outcome, 95% less keying.

4. Target market

• Primary customer: Kenyan SMBs that buy regularly from informal suppliers — small manufacturers, building contractors, agri-processors, restaurants, hardware/retail — roughly 5–50 staff, KSh 5M–200M annual turnover, in Nairobi, Mombasa, Nakuru, Kisumu, Eldoret. Secondary buyer: the independent accountant/bookkeeper running 10–50 such SME clients, who eats this pain every filing cycle.
• Why they buy: “I paid the supplier, the supplier won’t go on eTIMS, and now KRA wants to tax me on the money I already spent.” It’s a direct, quantifiable cash loss — disallowed expense × 30% corporate tax — felt every month and brutally at filing.
• Rough TAM reasoning: ~250K VAT-registered businesses in Kenya plus 511K+ micro/small taxpayers onboarded since 2024; 8.1M active taxpayers overall. Even a few percent of the VAT-registered base — say 5,000–15,000 buyers with meaningful informal-supplier exposure — is a healthy bootstrap market. Accountants multiply reach: one firm = dozens of buyers.
• Why now for them: The disallowance rule went live January 1, 2026 and the first full filing season under it is happening right now (2025 returns filed in 2026, June deadlines). The pain is not theoretical — it’s this quarter’s tax bill.

5. Product sketch (MVP)

• Snap or forward a supplier receipt (photo, M-Pesa SMS, WhatsApp) → AI extracts vendor, items, amounts, VAT status.
• One-tap reverse invoice draft pre-filled to eCitizen’s required fields; flags missing/risky data (no PIN, VAT mismatch) before submission to cut rejections.
• Supplier approval chaser: sends the supplier a plain-language SMS/WhatsApp with the exact *222# approval steps, nudges until approved, tracks status (pending / approved / rejected).
• Month-end package: every expense sorted into reverse-invoiced vs. non-eTIMS-CSV vs. already-compliant, with the ready-to-upload non-eTIMS CSV generated for you.
• Deduction-at-risk dashboard: running KSh tally of expenses that will be “added back” if not fixed before filing — the loss made visible.
• Accountant multi-client view: one login, all SME clients, bulk-chase suppliers, per-client filing pack.
• M-Pesa Paybill/Till statement import to auto-match payments to receipts and surface undocumented spend.

6. AI angle — what’s load-bearing

AI does the two jobs a clerk hates. One: extraction. Kenyan informal receipts are chaos — handwritten chits, faded thermal slips, M-Pesa confirmation SMS, WhatsApp photos, Swahili/English mix. Vision + LLM parsing turns that into structured invoice fields reliably; without it you’re back to manual keying and the product is pointless. Two: triage and drafting. The model decides per expense whether it’s reverse-invoiceable, CSV-only, or already compliant, drafts the correct document, and writes the supplier the right approval instructions. Remove the AI and you’ve rebuilt a data-entry form — which is exactly the status quo that’s failing. The AI is the product, not a sticker.

7. Localization angle

This is the localization play — it cannot be a generic global tool. It’s bolted to Kenyan rails end to end: KRA eTIMS / iTax, the eCitizen reverse-invoicing flow, USSD *222# supplier approval, M-Pesa Paybill/Till reconciliation, Swahili-English receipts, and KSh pricing tuned to SME wallets (a KSh 1,500/mo tier works where a $49 tool doesn’t). The regulatory specificity is the moat — a global invoicing SaaS has no reason and no path to model Kenya’s reverse-invoicing rules. Natural template to replicate later into Nigeria (FIRS/NRS, small-supplier rollout 2027), Tanzania, Uganda as their mandates bite — but Kenya first, deep.

8. Business model — path to $1M–$5M ARR

• Pricing: SMB self-serve KSh 1,500/mo ($11.50) entry, KSh 3,500/mo ($27) for higher volume + M-Pesa reconciliation. Accountant/firm tier KSh 8,000–20,000/mo (~$60–155) for multi-client.
• ACV: Blended ~$180/yr SMB, ~$1,000/yr accountant firm.
• Rough math to $1M ARR: ~3,500 SMBs at $180 = $630K, plus ~370 accountant firms at $1,000 = $370K → ~$1M. Or ~5,500 SMBs alone.
• Rough math to $5M ARR: ~18,000 SMBs + 1,500 firms, or expansion into a second African mandate market (Nigeria/Tanzania) using the same engine. Realistic only if the accountant channel compounds and a second geography opens.
• Expansion path: usage tiers by receipt volume; add M-Pesa reconciliation, then full mini-bookkeeping; firm seats grow with their client books; cross-border replication.

9. Go-to-market wedge — first 100 customers

• Accountants are the wedge. Kenya has thousands of small accounting/bookkeeping firms drowning in eTIMS reconciliation. Scrape ICPAK member directories + Google Maps for “accountant / tax consultant Nairobi/Mombasa/Nakuru,” send a personalized WhatsApp/Loom showing their exact month-end reverse-invoice grind collapsed to minutes. One signed firm = 10–50 downstream SMBs.
• Ride the filing-season panic. Target the live June 2026 / quarterly deadlines: run targeted posts in Kenyan SME and accountant Facebook/WhatsApp groups (large, active communities around KRA changes) with a free “deduction-at-risk calculator” — upload your expenses, see the KSh you’re about to lose. Gated to sign-up.
• Sector beachhead. Pick one informal-supplier-heavy vertical — furniture makers (Gikomba), building contractors, or fresh-produce restaurants — and saturate it via WhatsApp referral, since these owners cluster and talk.
• Content on the rule. The web is full of confused “what is reverse invoicing” searches right now — rank for the how-to, convert to the do-it-for-you tool.

10. Build complexity — justification

Medium. Off-the-shelf carries most of it: vision+LLM receipt extraction, standard web/mobile stack, SMS/WhatsApp gateways (Africa’s Talking, Twilio), M-Pesa Daraja API. The custom work is the eTIMS/iTax/eCitizen integration surface and reverse-invoice field mapping — these are documented but fiddly, KRA APIs are known to be flaky, and the supplier-approval chase needs a reliable USSD/SMS workflow. No research breakthrough, no proprietary dataset. A 2–3 person team with one Kenyan tax-domain advisor ships a credible v1 in ~3–4 months.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Facilitates KRA-defined reverse invoicing; no license needed to help taxpayers comply.
Ethical — no harm / dark patterns	✅	Helps SMBs keep legitimate deductions and stay compliant. No exploitation.
Market exists (evidence above)	✅	Live rule since Jan 2026, paying seller-side market, active complaints.
1–5 person team can build this	✅	2–3 builders + domain advisor, ~3–4 months.
Launchable with <$50K / ₹40L	✅	API + cloud + outreach; well under cap.

All five pass.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	16/20	Direct cash loss (disallowed expense × tax), felt monthly, brutal at filing. Not quite “daily hair-on-fire” but close, and recurring.
Demand evidence	15	12/15	Live regulation, paying adjacent market, documented reconciliation pain (“100 staff”), active SME/accountant chatter. Lacks verbatim buyer quotes begging for this exact tool.
Build feasibility	15	11/15	Mostly off-the-shelf; KRA/eCitizen integration + flaky APIs + USSD chase add real engineering. ~3–4 months.
Distribution clarity	15	11/15	Accountant channel is concrete and scrapeable; filing-season urgency is real. Conversion math still unproven.
Revenue mechanics	15	11/15	Pricing benchmarked to existing eTIMS tools and accountant retainers; $1M path needs thousands of small accounts — achievable but not trivial at KSh wallets.
Time to first revenue	10	7/10	Self-serve + accountant pilots can pay within 4–8 weeks of launch, especially mid-filing-season.
Defensibility	10	5/10	Regulatory-knowledge + workflow lock-in + accountant relationships are soft moats; a determined local clone could copy it, and KRA could automate parts away.
Total	100	73/100

13. Qualitative modifiers

Founder-fit tags

operations-heavy · domain-expertise-required (Kenyan tax/eTIMS fluency is non-negotiable; on-the-ground accountant relationships drive distribution).

Key assumptions to validate (3–5)

1. Assumption: SMB buyers feel the disallowed-expense loss sharply enough to pay KSh 1,500–3,500/mo to avoid it. How to test: 30 in-person/WhatsApp interviews with Gikomba/Nairobi SMB owners; show their actual at-risk KSh and ask for a pre-commit.
2. Assumption: Accountants will adopt and resell to their client book rather than just bill the hours themselves. How to test: Pilot with 5 small firms; measure whether they push it to ≥5 clients each within 60 days.
3. Assumption: Supplier approval via SMS/USSD nudges actually completes at a usable rate (informal suppliers cooperate). How to test: Run 100 real reverse-invoice chases manually; measure approval completion % and median time.
4. Assumption: KRA’s APIs/eCitizen flow are stable enough to automate against. How to test: Build the integration spike first; log error/downtime rates over two weeks.

Risk flags

1. Regulatory risk: KRA is actively changing the rules (PIN field made optional, CSV relief opened mid-2026). They could automate reverse invoicing or M-Pesa-linked auto-invoicing and shrink the manual gap. Mitigate by owning the receipt-to-package workflow and accountant relationship, not just the keystroke-saving.
2. Platform dependency: Hard reliance on KRA/eCitizen APIs (flaky) and Safaricom M-Pesa/USSD. Outages become your support tickets.
3. Willingness-to-pay at KSh wallets: SMB price sensitivity is real; the accountant channel is the margin insurance.

14. Structured verdict

Score:                  73/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Operations-heavy founder with deep Kenyan tax/eTIMS domain knowledge + accountant network, paired with one full-stack/AI builder
Time to revenue:        4–8 weeks post-launch (mid-filing-season pilots)
Capital to launch:      $8–15K (KSh ~1–2M) — APIs, cloud, outreach, one domain advisor
Top 3 assumptions to validate first:
  1. SMB buyers pay KSh 1,500–3,500/mo to stop disallowed-expense loss — 30 owner interviews with their real at-risk KSh + pre-commit
  2. Accountants resell to ≥5 clients each — 5-firm pilot, measure downstream adoption in 60 days
  3. Supplier USSD/SMS approval completes usefully — 100 manual reverse-invoice chases, measure completion % and time
Kill criteria:
  - Abandon if <20% of 30 interviewed SMBs will pre-commit to a paid plan
  - Abandon if supplier approval completion stays <50% after nudges (the chase is the product)
  - Abandon if KRA ships native automated reverse invoicing / M-Pesa auto-invoicing before v1, collapsing the manual gap

15. Next step — 1-week validation sprint

• Day 1–2: Build the deduction-at-risk calculator landing page (upload expenses → KSh you’ll lose). Pull ICPAK + Maps lists of 200 Nairobi accountants/SMBs.
• Day 3–4: Personalized WhatsApp/Loom outreach to 40 accountants + 40 SMB owners showing their specific reverse-invoice grind and at-risk KSh. Book interviews.
• Day 5: Run 30 interviews; manually do 10 real reverse-invoice chases end-to-end (eCitizen draft + USSD supplier approval) to measure true completion rate and time.
• Decide go / no-go on: ≥20% of interviewed SMBs verbally pre-commit to a paid plan and ≥50% of the 10 manual supplier chases complete approval within 72 hours. Falsifiable — if either fails, the wedge isn’t real.