AvisaSAT — cancellation sentry for Mexican accounting firms

1. One-liner

Watches every client’s Buzón Tributario and catches incoming REP cancellations before 3-day silence kills a paid invoice’s deduction.

2. Trend signal — why now?

On January 1, 2026, Mexico’s tax authority (SAT) changed the rules. Under rule 2.7.1.35 of the Resolución Miscelánea Fiscal 2026 (published Dec 28, 2025), a CFDI con Complemento de Pago — the REP, the receipt that proves a payment cleared an invoice — can no longer be cancelled unilaterally. The issuer now files a cancellation request, SAT notifies the recipient through their Buzón Tributario, and the recipient has 3 business days to accept or reject. If they say nothing, the cancellation is deemed accepted and the REP flips to “Cancelado por plazo vencido.”

That last sentence is the whole business. A company that already paid an invoice can have the supporting REP silently cancelled out from under it — purely by not checking a government inbox for three days — and lose the deductibility of that expense. The fiscal press is blunt about the fix: “Es indispensable monitorear de forma periódica el Buzón Tributario” and “Debe implementarse una rutina diaria que consulte el Buzón Tributario para detectar solicitudes de cancelación entrantes” (ContadorMx, e-Digital Sicofi, Jan 2026). The recommended solution today is a manual daily routine. Nobody hand-checks a federal inbox every single day across 80 client RFCs — which is exactly the gap.

Sanctions for the broader cancellation regime are deferred to Jan 1, 2027 (transitorio Cuarto, RMF 2026), so the deductibility-loss exposure is live now and the formal penalty pressure lands in ~6 months. Perfect timing to sell ahead of the deadline.

Provenance:

• Signal 1 (demand): New SAT rule 2.7.1.35 makes REP cancellation require recipient acceptance; silence within 3 business days = tácit cancellation of a paid invoice, losing deductibility — fiscal press calls for a “daily Buzón routine” — https://contadormx.com/cancelacion-cfdi-complemento-de-pago-sat-2026/ — 2026-01 / https://www.e-digital.com.mx/portal/la-cancelacion-de-complementos-de-pago-dejo-de-ser-directa/ — 2026-01
• Signal 2 (feasibility): Third-party APIs already automate programmatic access to the SAT Buzón Tributario and CFDI/XML retrieval (CSFacturación Buzón API, Heru fiscal APIs), so daily polling across many RFCs is off-the-shelf — https://developers.csfacturacion.com/api-buzon-tributario-sat — 2026 / https://www.heru.app/empresas/tecnologia-fiscal/ — 2026
• Signal 3 (economic): Large paid accounting-SaaS market for Mexican PyMEs already spending 138–599 MXN/mo (Alegra, Facturama, Bind ERP, Contadigital, QuickBooks MX) — money is moving in fiscal tooling — https://programascontabilidad.com/comparativas-de-software/software-de-contabilidad-mexico/ — 2026 Category: Regulatory arbitrage

3. The opportunity

The incumbents — Facturama, Alegra, Bind ERP, Contadigital, QuickBooks MX — sell invoicing and accounting suites. They are built around the issuer flow: stamp a CFDI, generate the REP, keep books. The Jan 2026 rule created a brand-new recipient-side obligation that none of them treat as a first-class product: continuously watch every RFC’s Buzón Tributario for incoming REP-cancellation requests and force a decision inside 72 hours.

That’s a monitoring-and-alerting problem, not an invoicing problem. The big suites bury it inside dashboards nobody opens daily. A focused tool that does one job — “a cancellation just hit client X’s Buzón, you have until Thursday 6pm, accept or reject” — beats a buried tab in a general ledger. The incumbent weakness is altitude: they’re horizontal accounting platforms; this is a sharp, time-boxed, money-on-the-line watch that a despacho contable feels every week.

10× better = a despacho stops manually logging into the SAT portal across dozens of client RFCs and instead gets a single triaged queue: incoming cancellations, deadline countdowns, reconciliation status, and a one-click decision with the rationale logged for the audit file.

4. Target market

• Primary customer: Small-to-mid despachos contables (accounting firms) in Mexico, 1–10 contadores, each managing 20–200 PyME clients’ fiscal compliance. Secondary: the in-house admin/finance person at a single mid-size PyME who handles their own SAT obligations.
• Why they buy: In their words (synthesized from fiscal-press guidance, Jan–Apr 2026): “la omisión al responder puede provocar cancelaciones automáticas de REP que respaldan operaciones ya pagadas” — i.e. miss a Buzón notice and a client loses a legitimate deduction. The contador eats the blame. They will pay to never be the reason a client got a non-deductible expense.
• Rough TAM reasoning: Mexico has millions of PyMEs (INEGI DENUE) and a large, fragmented accounting-services sector (Data México). Even a conservative slice — say 30,000–50,000 small despachos that each carry the new daily-monitoring burden — at a few hundred pesos/month is a comfortable >$5M ARR ceiling without needing the whole market.
• Why now for them: The rule is 5 months old, enforcement-by-deduction-loss is already live, and formal sanctions arrive Jan 1, 2027. Right now every despacho is improvising a manual routine. Sell into that anxiety before a 2027 deadline scramble.

5. Product sketch (MVP)

• Connect each client RFC once (via existing SAT-credential / PAC API integration); AvisaSAT polls the Buzón Tributario daily across all of them.
• Cancellation queue: every incoming REP-cancellation request shown as a card — issuer, invoice, amount, the underlying payment, and a live countdown to the 3-business-day deadline.
• One-click decide: Accept or Reject with a required short rationale; the decision and timestamp are saved to an audit file per client.
• Payment reconciliation check: matches the REP’s UUID to the recorded payment so the contador can see “this REP supports a payment you actually received” before rejecting a cancellation.
• Deadline escalation: WhatsApp + email alerts at request, T-1 day, and T-2 hours so nothing lapses into tácit cancellation.
• Issuer-side companion: when your client wants to cancel a REP they issued, draft the coordination message to the counterparty and track the 3-day status.
• Monthly exposure report: per client, REPs at risk, decisions made, and deductibility protected — the artifact the despacho shows its client to justify the fee.

6. AI angle — what’s load-bearing

Two places AI does real work, not decoration:

1. Reconciliation matching. Tying an incoming REP to the right payment is messy — short payments, multiple open invoices, foreign currency, missing exchange-rate fields, wrong UUID linkage. An LLM/extraction layer reads the CFDI XML + the client’s payment records and produces a confident “this REP backs payment #4471, MXN 38,200, paid Mar 12” with a flag when it can’t. Pure rules choke on the edge cases; this is where the 2-minute-vs-2-hour collapse happens.
2. Decision recommendation. Given the reconciliation result and history, the tool recommends Accept vs Reject and writes the rationale in plain Spanish for the audit file. Remove the AI and you’re back to a contador manually cross-referencing XMLs against bank statements under a 72-hour clock — which is the status quo we’re replacing.

If you stripped the AI out, this degrades to a glorified inbox poller. The matching + recommendation is what makes it worth paying for.

7. Localization angle

This is the localization play — it cannot exist outside Mexico:

• Regulatory: Built entirely around SAT rule 2.7.1.35, the Buzón Tributario, CFDI 4.0, and the REP/Complemento de Pago. Deep Mexico-specific fiscal knowledge is the product.
• Language: Spanish-first UI, Spanish rationales, Spanish alerts. Non-negotiable.
• Payment rails / pricing: Priced in MXN at a tier that sits beside Alegra/Contadigital (138–599 MXN/mo). A 249–399 MXN/mo add-on is an easy yes for a despacho; a US-dollar tool would be mispriced.
• Distribution: WhatsApp-native alerts (the default business channel in Mexico) and selling through colegios de contadores and fiscal-content creators.

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

• Pricing: Tiered by number of client RFCs monitored. Despacho Starter 399 MXN/mo (up to 25 RFCs), Pro 899 MXN/mo (up to 100 RFCs), plus a single-PyME tier at 149 MXN/mo.
• ACV: Blended ~7,200 MXN/year (~$390 USD) per despacho account, higher as RFC counts grow.
• Rough math to $1M ARR: ~$1M USD ≈ ~18.5M MXN. At a 7,200 MXN blended ACV that’s ~2,560 paying despacho accounts — a low-single-digit percent of the small-despacho universe. Achievable.
• Rough math to $5M ARR: ~12,000–13,000 accounts or the same base at a higher average tier as despachos add RFCs and you upsell the issuer-side module + a broader Buzón-notice watch (not just REP cancellations, but all SAT notifications). Expansion is the realistic route to $5M, not raw logo count.
• Expansion path: Land on REP-cancellation watch → expand to full Buzón Tributario monitoring (all SAT notices, requerimientos, multas) → per-RFC usage growth as the despacho’s client book grows. Natural seat/usage expansion.

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

• Colegios de contadores & fiscal Facebook/WhatsApp groups. Mexican contadores cluster in dense professional communities. Post a free “REP cancellation deadline calculator” + a one-page explainer of the 2027 sanction timeline; gate the daily-watch tool behind a trial. These groups have thousands of exactly-right members.
• Fiscal content creators / despacho YouTubers. ContadorMx-style channels already publish the “monitorea tu Buzón diario” advice. Sponsor 5–10 of them to demo the tool as the way to actually do what they’re telling people to do. Their audience is 100% target.
• Direct outbound to despacho directories. Scrape the SAT directory of registered contadores públicos and despacho listings (DENUE / Cronoshare), send a Spanish Loom showing a real REP cancellation getting caught with hours to spare. Time-bounded fear (“you have 3 days”) makes this convert.
• PAC channel. Mid-size PACs that lack a cancellation-watch feature could white-label or refer it; one PAC partnership delivers a batch of despachos at once.

The first 100 are a 3–4 week sprint through one or two big contador communities. The hook writes itself: a paid invoice quietly losing its deduction.

10. Build complexity — justification

Medium. The hard parts are off-the-shelf: Buzón Tributario access and CFDI/XML retrieval already exist as third-party APIs (CSFacturación, Heru), and the alerting/queue UI is standard web + WhatsApp Business API. The custom work is (a) reliable daily polling and deadline state-machine across many RFCs without missing the 72-hour window, and (b) the reconciliation/recommendation layer. A pair could ship a credible v1 in ~10–14 weeks. Not Low because handling SAT credentials securely and getting the multi-RFC polling/reconciliation right takes real engineering discipline.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Helping taxpayers respond to SAT notices is squarely legal; uses sanctioned API access.
Ethical — no harm / dark patterns	✅	Protects taxpayers from silent deduction loss. No dark patterns.
Market exists (evidence above)	✅	Live 2026 rule + large paid accounting-SaaS spend + explicit “monitor daily” guidance.
1–5 person team can build this	✅	Pair, ~10–14 weeks on existing APIs.
Launchable with <$50K / ₹40L	✅	Software + API costs only; sub-$15K to launch.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	16/20	Money-on-the-line and recurring (every cancellation request, 72-hr clock). Just shy of hair-on-fire because most days nothing arrives — the pain is the exposure, not constant.
Demand evidence	15	11/15	Strong: fresh mandatory rule, large paid tooling market, explicit press guidance to monitor daily. Held below 13 because customer voice is expert/press synthesis, not yet verbatim despacho complaints.
Build feasibility	15	11/15	Core APIs off-the-shelf; reconciliation + multi-RFC deadline reliability is the real work. ~10–14 weeks.
Distribution clarity	15	11/15	Named channels (colegios, fiscal creators, SAT/DENUE directories, PACs) with a sharp fear-based hook. Conversion still unproven.
Revenue mechanics	15	12/15	Pricing benchmarked against incumbents; ~2,560 accounts to $1M is credible. $5M needs expansion, not just logos.
Time to first revenue	10	8/10	Trial-to-paid in weeks; urgency-driven. Below 9 because despachos buy on monthly cycles.
Defensibility	10	5/10	Execution + fiscal-rule depth + despacho workflow lock-in (audit files accumulate). A PAC could bolt this on — moat is focus and speed, not structural.
Total	100	74/100

13. Qualitative modifiers

Founder-fit tags

domain-expertise-required · technical-heavy

You need a Mexico fiscal-domain partner (a contador or ex-PAC person) who lives and breathes CFDI/Buzón rules, paired with a builder comfortable with SAT/PAC API integration and a reliable scheduling/state-machine backend.

Key assumptions to validate (3–5)

1. Assumption: Despachos will pay ~399–899 MXN/mo specifically for cancellation-watch, separate from their existing accounting suite. How to test: 30 despacho interviews + a pre-sale landing page; target ≥10% of demoed despachos commit to a paid pilot.
2. Assumption: Reliable programmatic daily Buzón polling across many client RFCs is achievable via available APIs/credential models without breaking SAT terms. How to test: Build a 20-RFC polling prototype on a real PAC/Buzón API; confirm zero missed notifications over 30 days.
3. Assumption: Incoming REP cancellations are frequent enough that the watch feels worth paying for (not a once-a-year event). How to test: Sample 50 despachos’ last 90 days of Buzón notices; measure REP-cancellation request frequency per RFC.
4. Assumption: The recipient-side gap is real — incumbents (Alegra/Bind/Contadigital) don’t already offer a first-class cancellation-deadline watcher. How to test: Hands-on audit of the top 5 suites’ cancellation-handling UX.

Risk flags

1. Platform dependency: Built on SAT Buzón + PAC APIs. If SAT changes the Buzón access model or a PAC restricts third-party polling, the core breaks. Mitigate with 2+ API providers.
2. Incumbent fast-follow: A PAC or Alegra could ship a “cancellation alerts” feature in a quarter. Defensibility is execution + despacho relationships, so move fast and lock in workflow/audit-file habit.
3. Market timing (demand depth): If REP cancellations are rare for a given despacho, perceived value drops. Validate frequency (assumption 3) before scaling spend — and broaden to all-Buzón-notice monitoring early to widen the value.
4. Regulatory drift: SAT softens the rule or extends tácit-cancellation grace. The Jan 2027 sanction date is the live catalyst — watch for changes.

14. Structured verdict

Score:                  74/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Mexico fiscal-domain contador + technical co-founder (SAT/PAC API integration)
Time to revenue:        6–10 weeks from launch
Capital to launch:      $10–15K USD (≈ ₹8–12L)
Top 3 assumptions to validate first:
  1. Despachos pay 399–899 MXN/mo for a standalone watcher — 30 interviews + pre-sale page, ≥10% pilot commit
  2. Reliable multi-RFC daily Buzón polling via available APIs — 20-RFC prototype, 0 missed notices in 30 days
  3. REP cancellations are frequent enough to justify the fee — sample 50 despachos' 90-day Buzón history
Kill criteria:
  - Abandon if <10% of 30 demoed despachos commit to a paid pilot
  - Abandon if the median despacho sees <1 REP-cancellation request per month across its book (value too thin)
  - Abandon if SAT/PAC API terms prohibit automated third-party Buzón polling with no compliant workaround

15. Next step — 1-week validation sprint

• Day 1–2: Build a 20-RFC Buzón-polling prototype against a real PAC/Buzón API. Confirm it can detect an incoming REP-cancellation notice and compute the 3-business-day deadline correctly. Sample those RFCs’ last 90 days to measure cancellation frequency.
• Day 3–4: Interview 30 despachos contables (via 2 contador WhatsApp/Facebook groups). Ask: how do you monitor the Buzón today, how many client RFCs, have you lost or nearly lost a deduction, and would you pay 399 MXN/mo to never miss one. Put up a Spanish pre-sale landing page.
• Day 5: Go / no-go. Go only if (a) the prototype caught a real cancellation with correct deadline math, (b) median sampled despacho sees ≥1 REP-cancellation request/month, and (c) ≥3 of 30 despachos commit to a paid pilot. Anything less = the exposure is real but the willingness-to-pay isn’t — revisit pricing or pivot to full Buzón-notice monitoring.