SB StartupBasket
All ideas
74 /100 GO Low complexity

HandoffLedger — KSeF delivery ledger for Polish exporters

Proves a Polish exporter actually handed each KSeF invoice to its foreign buyer, and when they opened it.

Compliance SaaS EU Compliance-drivenSMBAI-agentMultilingualSolo-builder
— views
Evaluation Scores
74/100

GO

Overall Score

15
Problem
12
Demand
13
Build
11
Distrib.
11
Revenue
8
Time
4
Defense

HandoffLedger

1. One-liner

Proves a Polish exporter actually handed each KSeF invoice to its foreign buyer, and when they opened it.

2. Trend signal — why now?

On 1 April 2026 every Polish VAT taxpayer outside the micro bracket became obliged to issue invoices through KSeF — including invoices to German, French, American and Chinese customers. The Ministry left one thing conspicuously unautomated: the foreign buyer cannot log in to KSeF.

That creates a legally-mandated manual step that did not exist in March 2026. The seller must issue the XML into KSeF, then separately deliver a human-readable copy carrying the KSeF number and QR verification code, through a channel the buyer has agreed to. Per art. 106gb(4), if the delivery method is not agreed, the recipient has not formally received the invoice. A tax analysis of the provision calls the rules “illogical,” noting they “impose obligations primarily on invoice issuers rather than equally on buyers.”

The volume math is already documented by Polish integrators: dozens of foreign invoices a month means “godziny dodatkowej pracy” — hours of extra work — and at hundreds of invoices “automation is needed.”

The clock matters. Throughout 2026 penalties are suspended; from 1 January 2027 sanctions reach up to 100% of the VAT on an invoice mishandled outside the system. 2026 is the year an exporter buys insurance, because 2027 is the year it stops being free to be sloppy.

Provenance:

3. The opportunity

Every KSeF vendor in Poland is fighting over the same thing: getting the XML into the government system. That’s the compliance event they all instrument, because it’s the one that returns a receipt — the UPO, timestamped and signed by the state.

The leg after KSeF has no receipt at all. The seller emails a PDF and hopes. There is no artifact proving what was sent, to whom, in what format, on what date, under what agreed method, or whether anyone opened it. For domestic trade that gap doesn’t matter — the buyer pulls the invoice from KSeF himself and the state’s own log is the proof. For export and WDT it matters enormously, because the state’s log stops at the border and the seller’s obligation doesn’t.

So the exporter ends up with a perfect record of the half of the transaction the government can already see, and no record of the half only he is responsible for. That’s exactly backwards from where the liability sits.

The incumbents won’t fix this, and the reason is structural rather than lazy. Comarch, Sovos, Tungsten and the ERP vendors sell KSeF connectivity to finance departments as a transmission feature — did the XML clear validation, yes or no. Delivery-to-the-buyer is where their scope ends and “your email client” begins. Meanwhile the Polish integrator tools that do automate the foreign send treat it as a send-and-forget action: generate PDF, attach QR, dispatch, done. Nobody keeps the ledger, because a ledger is only valuable if someone later disputes what happened — and in 2026, with penalties suspended, nobody has been disputed yet.

That’s the arbitrage. The pain is legally live now, the consequence arrives in January 2027, and the category’s attention is pointed at the wrong leg of the journey.

4. Target market

  • Primary customer: Finance manager or head bookkeeper at a Polish exporter — 10–150 employees, 5M–150M PLN revenue, issuing 30–500 foreign invoices a month into Germany, Czechia, France, Netherlands, UK, USA. Manufacturing subcontractors, furniture and auto-parts makers, IT and creative services firms billing EU clients, e-commerce sellers doing WDT. Also the ~1,000-strong tier of Polish accounting offices (biura rachunkowe) that carry foreign-invoice clients and inherit the exposure.
  • Why they buy: Not to send the PDF — their invoicing software may already do that. They buy because when a German customer’s AP department says “we never received invoice 4471, that’s why it’s 60 days unpaid,” or when a 2027 audit asks how invoices were made available outside the system, the exporter currently has an email folder and a shrug. Cash and penalty exposure both sit on the same missing artifact.
  • Rough TAM reasoning: ~2.3M Polish SMEs; ~4.8% export goods, giving a six-figure population of exporting firms. The serviceable slice — enough foreign invoice volume to feel the pain, not so large they have a Comarch implementation — is realistically 15,000–40,000 firms plus the accounting offices serving them. At 249 PLN/month, 3,000 customers is roughly $2.4M ARR. This is a bounded market, which is the point: too small for a VC-backed compliance suite, correctly sized for a two-person team.
  • Why now for them: The obligation began 1 April 2026. The penalty regime begins 1 January 2027. Anyone who has been muddling through since April has 2026 to build a record — and no record for the months already elapsed unless someone reconstructs it.

5. Product sketch (MVP)

  • Delivery ledger — an append-only record per invoice: KSeF number and UPO timestamp, the rendered document hash, recipient address, channel, send time, delivery/open/download events, and the agreed-method basis being relied on.
  • Agreed-method capture — a one-time counterparty onboarding flow (email or short web form, in the buyer’s language) where the foreign buyer nominates and confirms their delivery channel and address. Stores the confirmation as the art. 106gb evidence. This is the artifact nobody currently has.
  • Bilingual document rendering — FA(3) XML into a clean PDF in Polish + the buyer’s language (DE/EN/FR/CS/NL), with the compliant QR code, so a German AP clerk can actually process it without a translation round-trip.
  • Tracked handoff — sends via the buyer’s agreed channel with open/download tracking and a verification landing page where the buyer confirms receipt in one click.
  • Silence alerts — flags invoices that were issued into KSeF but never opened by the buyer after N days. That’s the payment-delay early warning, and it’s the feature that gets used weekly rather than at audit time.
  • Reconciliation sweep — pulls the exporter’s KSeF issued-invoice list and shows every foreign-buyer invoice with no corresponding delivery record. Backfills the gap since April 2026.
  • Audit export — one PDF/ZIP per period: every foreign invoice, its KSeF ID, its delivery evidence chain. Hand it to the tax office or the client’s accountant.
  • Dispute pack — for a single contested invoice, a one-page evidence sheet for the buyer’s AP department.

6. AI angle — what’s load-bearing

Honest answer: AI is the assistant here, not the engine. The ledger, the QR rendering and the tracking are deterministic plumbing, and I’d rather say that than dress it up.

Where AI genuinely earns its place:

  • Counterparty resolution across messy data. Foreign buyer records in Polish SME systems are inconsistent — the same German customer appears as three entities with different VAT-ID formatting, addresses and AP contacts. Matching invoices to the right agreed-delivery-method record, and spotting when an invoice is about to go to a stale contact, is fuzzy entity resolution over dirty multilingual data. Rules get you 70%; that last 30% is where the missed deliveries hide.
  • Inbound AP correspondence triage. Foreign AP departments reply in German, French, Czech, English — “please resend in our portal,” “PO number missing,” “we need a paper copy.” Reading those replies, classifying them, and updating the delivery record automatically is the difference between a ledger that stays current and one that rots after month two.
  • Schema-error explanation. Export/WDT invoices fail on predictable but obscure fields — P_12_ marked “0 EX” vs “0 WDT”, KodKraju, NBP conversion date. Translating a rejection into “your intra-EU sale is tagged as an export, fix this field” is a small thing that saves a bookkeeper a call to their accountant.

If you removed the AI you’d still have a sellable compliance product — a worse one, with more manual data hygiene. I’m scoring this accordingly rather than pretending otherwise. The moat here is regulatory precision and workflow position, not model cleverness.

7. Localization angle

This is the localization play — it only exists because of one country’s statute. Poland-first and Poland-specific: FA(3) schema, KSeF 2.0 API, art. 106gb(4), the official QR spec, NBP conversion rates, Polish-language UI for the seller, and buyer-language rendering outward. Pricing in PLN at a level a 40-person exporter approves without a committee.

The bigger prize is that the pattern generalizes on a known timetable. Every country running a clearance-model mandate creates the same orphaned leg the moment a domestic seller invoices a foreign buyer who has no access to the clearance platform. Belgium went Peppol-mandatory in January 2026; Poland’s micro-enterprises join January 2027; more mandates land after. Same product, different schema adapter — but only after Poland is won. Trying to be pan-European on day one is how this loses to Sovos.

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

  • Pricing: Tiered on foreign-invoice volume. Starter 149 PLN/mo (up to 50 foreign invoices), Standard 249 PLN/mo (up to 200), Pro 499 PLN/mo (up to 1,000). Accounting-office plan at 899 PLN/mo covering 10 client entities. Sits deliberately at the top of the observed 30–200 PLN/mo band for KSeF tooling, because this is sold as exposure cover rather than as an invoicing utility.
  • ACV: 3,000 PLN ($750) blended for direct SMEs; 10,800 PLN ($2,700) for accounting offices.
  • Rough math to $1M ARR: ~1,250 SME customers at 249 PLN/mo ≈ 3.7M PLN ≈ $930K. Mixing in 80 accounting offices gets there with ~1,000 direct customers.
  • Rough math to $5M ARR: Needs ~4,000 direct customers plus 300 accounting offices, or Poland at ~2,500 customers plus a second mandate market (Belgium, or Poland’s own micro tier from Jan 2027) contributing the rest. Reachable, but assume the second country is required.
  • Expansion path: Volume tier upgrades as export activity grows; per-entity pricing at accounting offices; a paid archival tier (KSeF invoices must be retrievable for years and the delivery evidence should live exactly as long); later, the AP-side mirror — helping Polish buyers document PDF invoices received outside KSeF, which the KR Group analysis shows is its own live problem with double-recording risk.

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

  • The accounting offices are the channel, not the customer. Poland’s biura rachunkowe already field the “how do I invoice my German client now” question daily, and they carry the client’s exposure. Recruit 25 offices with a revenue share and a free multi-entity dashboard; each brings 5–15 exporting clients. This is the single highest-yield path and it’s a phone-and-coffee motion, not a marketing one.
  • Scrape the exporter list and target by evidence of foreign trade. Polish company registries plus EU VIES VAT-ID validation identify firms doing intra-EU trade; industry directories (furniture, auto parts, IT services) narrow it further. For each, send a Polish-language one-pager with a specific hook: “Since April you’ve issued X invoices to EU buyers. Where’s your proof you delivered them?” 2,000 targeted, 3–5% reply.
  • Free reconciliation audit as the wedge. Connect KSeF read-only, get a free report: every foreign-buyer invoice since April 2026 with no delivery evidence. It’s genuinely useful standalone, it’s a two-minute setup, and the number it returns — “you have 340 invoices with no delivery record” — is the sales argument. Converting the report into ongoing coverage is the paid product.
  • Ride the January 2027 penalty deadline with the tax-advisory crowd. Polish tax blogs, doradcy podatkowi, KSeF-focused LinkedIn and the VATupdate/Infor readership are already publishing on art. 106gb ambiguity. Co-author practical guidance with 2–3 tax advisors; they need content, and the product is the natural CTA. Q4 2026 is peak attention.
  • Target the freshly-mandated micro tier before January 2027. They get mandated 1 Jan 2027 and will be shopping in Q4 2026. Same message, one quarter later, to an audience that hasn’t yet chosen a vendor.

10. Build complexity — justification

Low. The government publishes the FA(3) schema, the KSeF 2.0 REST API and the QR specification; open-source FA(3)→PDF renderers with QR verification already exist and are permissively licensed, so the hardest-looking piece is largely solved. The rest is a standard web app: OAuth-style KSeF authorization, a document store, an email/tracking pipeline, and an append-only event log. A competent pair ships a credible v1 in 8–10 weeks. The real work isn’t engineering — it’s getting the legal interpretation of “agreed delivery method” right, which needs a Polish tax advisor on retainer rather than another developer.

11. Gating checklist

GatePass?Note
Legal in target market✅Helps taxpayers meet an existing obligation; uses published government APIs. No licence required.
Ethical — no harm / dark patterns✅Open-tracking on B2B invoices is standard commercial practice; buyer explicitly confirms channel during onboarding.
Market exists (evidence above)✅Mandate live since 1 April 2026; commercial tooling already selling; documented manual burden.
1–5 person team can build this✅Two people, 8–10 weeks, plus a tax advisor.
Launchable with <$50K / ₹40L✅Well under — advisor fees are the largest line item.

12. Feasibility score

AxisWeightScoreNotes
Problem intensity2015/20Legally mandated, felt weekly, with real money attached — but through 2026 penalties are suspended, so it’s urgent-on-paper before it’s urgent-in-the-gut. The unpaid-invoice angle is what makes it bite today.
Demand evidence1512/15Strong documentary evidence: the mandate, the published burden, existing paid tooling, integrators already automating adjacent steps. Docked because the evidence is from advisors and vendors rather than exporters complaining in their own words — I could not source verbatim end-user quotes.
Build feasibility1513/15Published schema and API, open-source renderers, standard web stack. 8–10 weeks for a pair.
Distribution clarity1511/15The accounting-office channel is concrete, named and warm, and the free-audit wedge is strong. Docked because it’s a Polish-language, relationship-led motion — a non-Polish founder cannot run it, and conversion through offices is unproven.
Revenue mechanics1511/15Pricing benchmarked against observed 30–200 PLN/mo tooling, and 1,250 customers for $1M ARR is achievable in-country. Docked because $5M almost certainly needs a second market, and PLN pricing caps ACV.
Time to first revenue108/10The free reconciliation audit converts fast and the deadline creates its own urgency; realistically 6–8 weeks from launch to first paid accounts.
Defensibility104/10Honest score. No proprietary data, no network effect, and any Polish KSeF vendor could bolt a delivery log onto an existing product in a quarter. What you get is regulatory precision, the accumulated agreed-method records that are painful to re-collect, and a head start. That’s an execution moat.
Total10074/100

13. Qualitative modifiers

Founder-fit tags

technical-heavy · domain-expertise-required

This needs a Polish-speaking founder or co-founder. Not a preference — the customer conversations, the accounting-office channel, the statutory interpretation and the marketing are all in Polish, and the legal nuance is the product. A foreign founder should not attempt this without a local partner who owns the go-to-market.

Key assumptions to validate (3–5)

  1. Assumption: Exporters perceive missing delivery evidence as a real risk before penalties start, rather than a 2027 problem to defer. How to test: 30 structured calls with finance managers at Polish exporters. Ask what they’d show if a buyer disputed receipt today. Count how many have anything beyond a sent-items folder — and, more importantly, how many care.
  2. Assumption: The unpaid-invoice angle, not the audit angle, is what actually closes deals in 2026. How to test: Split the cold outreach: half penalty-framed, half “your German customer says they never got it” framed. Compare reply and demo-booking rates.
  3. Assumption: Accounting offices will actively resell rather than politely nod. How to test: Pitch 15 biura rachunkowe; require a signed referral agreement plus one named client introduction to count as a yes. Nodding is not signal.
  4. Assumption: The “agreed delivery method” record is legally meaningful enough that a tax advisor will endorse it in writing. How to test: Commission written opinions from two independent Polish tax advisors on whether the captured confirmation satisfies art. 106gb(4). If they hedge, the compliance framing collapses and only the AR/payment framing survives.
  5. Assumption: Existing invoicing vendors won’t ship a good-enough delivery log within 6 months. How to test: Audit the release notes and roadmaps of the top 8 Polish invoicing/KSeF products quarterly.

Risk flags

  1. Feature-absorption risk (the big one): This is a feature-shaped product. Comarch, iFirma, wFirma, Fakturownia or any Polish invoicing vendor could add a delivery ledger and give it away. The counter is to be sharper on the evidence chain and own the accounting-office relationship before they notice — but a fast-follower here is likely, not hypothetical.
  2. Regulatory drift: KSeF has been delayed and rewritten repeatedly. The Ministry could clarify art. 106gb in a way that makes evidence trivial, could build buyer-side foreign access, or could extend the penalty holiday past 2027 and defuse the urgency entirely.
  3. Platform dependency: Entirely dependent on the KSeF API — its uptime, its authorization model, and its terms. Rate limits and file-size ceilings are already documented pain points for high-volume issuers.
  4. Urgency gap in 2026: Penalties are suspended for the whole first year. There’s a real chance the honest answer is “we’ll deal with it in December,” which pushes the market to Q4 2026 and makes the first two quarters slow.
  5. Single-country concentration: All revenue depends on one country’s tax statute. Great focus, bad diversification.

14. Structured verdict

Score:                  74/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Polish-speaking technical founder with a tax-advisor
                        partner; or a two-person team where one owns the
                        biura rachunkowe channel
Time to revenue:        6–8 weeks from launch (free audit converts fast)
Capital to launch:      $8–15K (PLN 30–60K) — mostly tax advisory fees
Top 3 assumptions to validate first:
  1. Exporters treat missing delivery evidence as a today problem —
     30 finance-manager calls, count who has more than a sent-items folder
  2. Accounting offices will resell — 15 pitches, require signed referral
     + named client intro to count as yes
  3. Agreed-method capture satisfies art. 106gb(4) — two independent
     written tax-advisor opinions
Kill criteria:
  - Abandon if <8 of 30 exporters can't produce delivery evidence AND
    don't consider that a problem worth paying to fix
  - Abandon if fewer than 3 of 15 accounting offices sign a referral
    agreement with a named client introduction
  - Abandon if two major Polish invoicing vendors ship a delivery-evidence
    log before your v1 reaches 50 paying customers
  - Abandon if the Ministry extends the penalty holiday beyond 2027 or
    grants foreign buyers direct KSeF access

15. Next step — 1-week validation sprint

  • Day 1–2: Build the target list. Pull Polish exporters via registry plus VIES validation in three sectors (furniture, auto parts, IT services). Get to 200 named firms with a finance contact. In parallel, book calls with 8 biura rachunkowe known to carry export clients.
  • Day 3–4: Run 30 exporter calls. One question decides everything: “A German customer claims they never received invoice 4471 and it’s 60 days unpaid. Show me what you’d send them.” Record whether they have anything beyond a sent-items folder, and whether they react with concern or a shrug. Run the split-framing outreach (penalty vs unpaid-invoice) across the remaining list.
  • Day 5: Commission one written tax-advisor opinion on whether a captured buyer confirmation satisfies art. 106gb(4), and pitch the 8 accounting offices for signed referral agreements.

Falsifiable go/no-go: Go only if ≥12 of 30 exporters cannot produce delivery evidence and rate the risk ≥7/10, and ≥2 of 8 accounting offices sign a referral agreement with a named client introduction, and the tax advisor’s opinion is affirmative rather than hedged. Miss any of the three and this drops to VALIDATE — rerun in Q4 2026 when the January penalty deadline does the urgency work for you.

Interested in a detailed proposal?

Get a deep-dive with market research, competitive analysis, and implementation roadmap.

Contact us

info@startupbasket.ai