Quick summary
Paraguay enters 2026 as LATAM's most dynamic economy (GDP +6.6% in 2025, BCP forecast of 4.2% for 2026) and the bloc's densest e-invoicing window. DNIT is rolling out SIFEN in four waves by taxpayer group — group 15 starts on March 2, 2026 — and from January 2, 2026 e-invoicing is mandatory for every state supplier. If your Odoo isn't tuned for Kuatia / e-Kuatia, KUDE and DNIT validations, you have 4 to 12 months to adapt.
- SIFEN 2026 rolls by groups: 15 — Mar 2, 16 — Jun 1, 17 — Sep 1, 18 — Dec 1.
- From January 2, 2026 e-invoicing is mandatory for all state suppliers (DNIT resolution N.°41/2025).
- Kuatia = the XML document (DTE). KUDE = its graphical representation with a QR code. They are two different things — confusing them costs you a tax credit.
- e-Kuatia is the regime via your own software or a PAC (for mid-size and large filers). e-Kuatia'i is DNIT's simplified portal for micro-emitters.
- Odoo +
l10n_pycovers the base, but out of the box it isn't enough: you need PAC integration, a KUDE template, pre-send validation, and contingency mode. - Electronic timbrado takes 5–10 business days to issue — the critical path to launch.
- Paraguay has just two Gold Odoo partners (Interfaces S.A. + SATI). The market stays a blue ocean for serious implementations.
Why 2026 is SIFEN's hinge year
SIFEN (Sistema Integrado de Facturación Electrónica Nacional) is Paraguay's national platform for electronic tax documents. It is run by DNIT (Dirección Nacional de Ingresos Tributarios), the agency that in 2024 absorbed part of the former SET's functions under the tax reform. Until 2023 Kuatias were issued by hand from the portal; since then Paraguay has been migrating to a fully electronic flow with XML exchange, digital signing, and DNIT server-side validation. For the regional comparison, look at Odoo in Peru (SUNAT 2026) and Odoo in Chile (SII 2026).
Terms people mix up
- Kuatia — the Electronic Tax Document (DTE) in XML format, digitally signed by the issuer. This is "what gets sent to DNIT and carries legal force."
- e-Kuatia — the electronic issuance regime via your own software or a PAC (Authorized Certification Provider). Aimed at mid-size and large filers with ≥ 500 documents/month or warehouse/CRM integration.
- e-Kuatia'i — the simplified regime for micro-emitters: issuance via a web form on the DNIT portal, no integration, free.
- KUDE (Kuatia Documento Electrónico — representation) — the graphical view of the Kuatia in PDF or HTML with a QR code, delivered to the customer. Legal force sits with the XML Kuatia, not the KUDE. Confusing the two is the most expensive mistake: someone prints the KUDE, treats it as "issued," and if the XML never reached DNIT the document legally doesn't exist.
DNIT timeline for 2026
DNIT publishes mandatory dates in waves; taxpayers are bucketed by revenue and regime. The 2026 operating calendar looks like this:
| Group | Go-live date | Who's in |
|---|---|---|
| State suppliers | January 2, 2026 | All state suppliers (DNIT resolution N.°41/2025) |
| Group 15 | March 2, 2026 | Mid-to-large taxpayers |
| Group 16 | June 1, 2026 | Next wave |
| Group 17 | September 1, 2026 | — |
| Group 18 | December 1, 2026 | Closes the main waves |
From a compliance standpoint these are four independent stress windows in a single year, plus an extra one for state contractors. Each window triggers a wave of "how do I get into SIFEN" searches: by Q1 part of the market is already on fire, and by Q4 the rest catches up.
The macro matters for sizing the investment: Paraguay's GDP grew 6.6% in 2025 (a 12-year high), BCP forecasts 4.2% for 2026 and the IMF 3.7%. The guaraní appreciated 20.34% over 12 months (from PYG 7,811/USD at the end of 2024 to PYG 6,370 in April 2026). Compliance investment today is "cheap" in real USD, and SMBs are scaling from 10–15 to 40+ headcount — the exact point where Excel and Bejerman LA stop holding up.
What you have to configure in Odoo
The base localization is the l10n_py module in the OCA (Odoo Community Association) repository. It ships the chart of accounts, the VAT structure (10% / 5%), document types and tax regimes. But out of the box l10n_py only delivers the foundation — a real implementation needs 3–5 custom modules on top. If you're inheriting a half-built Odoo, run an audit on the current instance before touching the configuration.
Minimum stack for SIFEN compliance
l10n_py(OCA) — chart of accounts, taxes, and partner structure with RUC.- PAC integration or a direct hookup to DNIT webservices (for e-Kuatia). The Paraguayan PAC pool is narrow — Gosocket, Edicom, a handful of local providers. Pick by monthly volume and SLA.
- XML digital signature — Odoo generates the signed XML using the issuer certificate provided by DNIT.
- Contingency mode — for when DNIT is down or the PAC is offline. The document gets generated locally with a contingencia flag and sent later. Without it, point-of-sale freezes.
- KUDE generator — PDF/HTML with a QR code for the customer. Usually a custom QWeb template, because Odoo's standard invoice report does not meet DNIT's mandatory-field list.
- Pre-send validation — before pushing to DNIT, check every field (valid RUC, correct CDC, VAT category consistent with the SKU). Skip this and DNIT tells you about the error — wasted cycle and downtime.
SIFEN document types
DNIT distinguishes 7+ DTE types; Odoo has to emit each one with the right references:
- Factura electrónica (FE) — Kuatia for a regular sale.
- Credit / debit note (NC/ND) — always linked by reference to the originating Kuatia.
- Autofactura — special cases where the buyer issues the document on the seller's behalf (typical for agro purchases from small producers).
- Nota de remisión — goods-transport document.
- Factura virtual — for specific regimes.
- Comprobante interno — internal document.
- DUI (Despacho Único de Importación) — customs document the Kuatia must reference on imports.
NC/ND is the part that breaks most often: if a debit note points to the wrong CDC (Código de Control), DNIT rejects it and the recalculation goes back to manual. For a Ciudad del Este importer with DUI references, each linking error costs 1–2 hours of review.
Taxes: IVA, IRP, IRACIS
- IVA: 10% (standard) / 5% (basic agro goods, medicines, specific categories). Set per-SKU via product categories. The typical error is applying 10% to an item that should run at 5%, or the other way around.
- IRP (personal income tax): progressive 8% / 9% / 10% by income. Reformed since 2023. In Odoo this lives in payroll with custom calculation rules.
- IRACIS (corporate tax on commercial, industrial and service income): the corporate tax. It has to fall out of Odoo accounting automatically — not through manual Excel exports.
Multi-currency for border operations
Paraguayan companies, especially in Ciudad del Este and Encarnación, run in PYG / USD / BRL at the same time. Odoo supports multi-currency natively, but the FX revaluation and realized FX differences need careful setup. Rates come from the BCP (Banco Central del Paraguay) and update daily. In production you want them pulled via API, not typed in.
When Odoo fits Paraguay — and when it doesn't
This section is for anyone comparing Odoo to Bejerman LA, SAP B1, or staying on QuickBooks / Excel. If your situation is closer to "half-implemented Odoo," review what we do when a project derails first — starting clean is cheaper than rescuing.
Fits: SMB of 15–80 staff growing through multiple channels
If you have 15–80 employees, USD 500k–10M in revenue, and you want sales, inventory, accounting and CRM in one loop, and you can spend USD 9k–30k in year one, Odoo fits. l10n_py covers the base and process customization is realistic in 10–14 weeks.
Fits: agroexporter with RTRS or EUDR traceability
If you export soy / wheat / beef / stevia to the EU or Brazil and need lot → silo → container → customer traceability plus certifications (RTRS for soy, GlobalGAP, EUDR from 2026), Odoo Manufacturing + Quality and custom agro modules close the loop better than Bejerman LA. Bejerman doesn't have a full Manufacturing block and most PY partners patch traceability with another tool or Excel — which collapses during the audit.
Fits with caveats: Ciudad del Este importer with DUI and multi-currency
If you bring electronics / perfume / watches across the Friendship Bridge, run 3+ currencies (PYG / USD / BRL / CNY), integrate Tigo Money, and every Kuatia has to reference a DUI, Odoo handles it — but this is not a boxed project. You need a custom Tigo Money integration (there's no official connector), a module to link Kuatia → DUI, and multi-currency tuned for border operations. Realistic year-one budget: USD 30k–45k, not 9k. If you also want cross-border dashboards, layer on Odoo + Analytics.
Doesn't fit: micro-emitter with 50 invoices a month
If it's 1–3 people and 50 invoices per month, Odoo is overkill. Use e-Kuatia'i: DNIT's free portal. There is internet dependency but zero capex. Odoo earns its place when inventory, sales reps and CRM are already in play and you want to automate processes — not just emit invoices.
Doesn't fit: team operating 100% in Guaraní
Odoo supports Spanish and English, not Guaraní. If your operational team works in Guaraní (common in rural operations around Concepción or Caaguazú), you will need Spanish training or a translation layer — operational risk and a source of errors. Alternative: local software with a bilingual UI.
5 typical implementation errors
#1. "We installed l10n_py and we're done"
The OCA module l10n_py is a foundation, not a finished SIFEN solution. Without a PAC connector, a KUDE template and pre-send validation, you end up with an Odoo that is "localized in theory" but Kuatias never reach DNIT. Real localization = l10n_py + 3–5 custom modules + 2–3 weeks of testing against the DNIT staging environment.
#2. KUDE vs Kuatia confusion
The accountant sees the printed KUDE and assumes the invoice "went out." But if the XML Kuatia wasn't accepted by DNIT (because the PAC dropped, the CDC was wrong, or a field was missing), there is no document — the customer can't deduct it and the company can't book revenue. The "accepted by DNIT" state has to live in the Odoo workflow, not in a side Excel.
#3. VAT misassigned by SKU
10% vs 5% is a common mistake for importers with mixed catalogs. With 2,800 SKUs, manual tax setup is a direct path to errors and downstream adjustments. Configure product categories with automatic rate assignment, then run a targeted audit on 50 SKUs.
#4. No contingency mode
DNIT goes down periodically (maintenance, overnight windows, incidents). If a register is open at that moment, the sale has to flow with a contingencia flag and be sent later. Without contingency in Odoo the sale blocks, the queue grows, and revenue doesn't book. This is base compliance, not a nice-to-have.
#5. Certificate and timbrado with no buffer
Electronic timbrado takes 5–10 business days; the certificate is a separate process. Starting configuration "the week before the group cutover" doesn't work. Realistic plan: kick off 6–8 weeks before the mandatory date.
Case: an agroexporter in Alto Paraná
Anonymous case. Company with 45 permanent employees + 120 seasonal, 6 silos, 28,000 tonnes of capacity, exporting soy and wheat to the EU and Brazil. Before Odoo: Bejerman LA + an agro module assembled by 4 different freelancers + Excel for traceability + an external facturador for Kuatia.
Trigger. The 2024 RTRS certification audit failed — there was no documented "farm → customer" traceability. EU contracts worth USD 1.8M were frozen pending resolution. In parallel, 12% of Kuatias were bouncing from DNIT after the April 2024 validation changes (the Bejerman partner shipped the module update 6 weeks later; during that window invoices ran with a delay).
What we did. Migration to Odoo 17 Enterprise (Manufacturing + Quality + Inventory + l10n_py + custom agro modules) over 5 months. Traceability tractor → silo → container → port → customer. A mobile app for producer partners with an offline mode (Itapúa zones with no signal) and sync when back on the network. Kuatia + e-Kuatia with pre-send validation. IVA 5% / 10% by SKU. Automated RTRS report straight from Odoo. Multi-currency PYG / USD / BRL.
Results:
- 2025 RTRS audit — passed on the first attempt
- Trace-back (a bag in the EU → the farm): 6 hours → 9 minutes
- Rejected Kuatias: 12% → 0% from the second month
- Month-end close: 14 days → 4 days
- 2024 export volume: +22%, zero EU claims
- Investment: USD 32k in year one, USD 12k from year two
Traceability is no longer a side Excel: the RTRS audit now comes out of the same instance that emits the electronic invoice. That is the real win.
SIFEN 2026 checklist
If you fall into group 15 / 16 / 17 / 18, or you supply the state, you have 4 to 12 months to adapt. The SIFEN 2026 Checklist (PDF, 36 items) covers: choosing a PAC, getting timbrado issued, configuring l10n_py, the KUDE template, pre-send validation, contingency mode, and the DNIT staging environment.
Grab the checklist — drop your email and it arrives immediately. Alternative: a 30-minute diagnostic on your current system. We will point to 3 places where compliance is breaking today.
Frequently asked questions
When does SIFEN become mandatory for my company?
It depends on the group. Group 15 — March 2, 2026, group 16 — June 1, 2026, group 17 — September 1, 2026, group 18 — December 1, 2026. If you supply the state, the obligation kicks in on January 2, 2026 regardless of group (DNIT resolution N.°41/2025). Validate the exact date for your RUC on the DNIT portal.
What is the difference between e-Kuatia and e-Kuatia'i?
e-Kuatia is for mid-size and large filers, via your own software (Odoo, SAP, a local ERP) or a PAC. e-Kuatia'i is for micro-emitters, via a web form on the DNIT portal, no integration, free. If you issue more than 500 invoices a month or need inventory / CRM automation, e-Kuatia is the path.
Does l10n_py cover Paraguay localization out of the box?
No. l10n_py (OCA) ships the chart of accounts, the tax structure, and basic document types. PAC integration, the KUDE template, pre-send validation, contingency mode, IRP / IRACIS payroll configuration, and BCP exchange rates all sit on top as customization. Production stack: l10n_py + 3–5 modules.
How much does an Odoo implementation cost in Paraguay?
SMB of 15–40 employees: USD 9k–30k year one, USD 10k–15k from year two (support + hosting + Enterprise licenses). Ciudad del Este importer with DUI and multi-currency: USD 30k–45k. Agroexporter with RTRS traceability: USD 25k–40k. Prices stay below PE / CL / CO because of market size and lower PAC competition.
What is KUDE — and can I just print a Kuatia and hand it over?
KUDE is the PDF/HTML of the Kuatia with the QR code, the "human-readable" version handed to the customer. Legal force sits with the XML Kuatia sent to and accepted by DNIT — not with the paper. You cannot "print a Kuatia": what you print is the KUDE as a visualization. If the XML wasn't accepted, the KUDE is legally invalid no matter how many times you print it.
Can DNIT fine me for a badly issued Kuatia?
Yes. Breaking SIFEN rules, skipping the electronic form, or loading wrong data opens tax penalties, VAT / IRACIS adjustment risk, and loss of deductibility for the counterparty. Specific amounts vary by resolution and sector; your contador or the current DNIT regulation gives you the concrete numbers.
How many Gold Odoo partners are there in Paraguay?
Two confirmed: Interfaces S.A. (interfaces.com.py, Gold continuously for over 5 years) and SATI (Asunción). This is the smallest Gold ecosystem among the 10 LATAM-sprint countries — the market stays open for new implementations and quality partners.
How does Paraguay compare with Argentina, Chile or Peru on SIFEN complexity?
Paraguay is closer to Uruguay (CFE 25.1) in calendar style: rollout by groups. XML validation, CDC and PAC logic resemble the Chilean (SII) and Peruvian (SUNAT) models, but the local partner ecosystem is much smaller. Versus Argentina (AFIP/ARCA), there are fewer document types but more rigidity on KUDE and contingency.
