ShipGlobal — AI export co-pilot for Indian MSMEs

1. One-liner

An AI-powered export documentation and compliance tool that helps Indian small manufacturers go from purchase order to shipped container without a customs broker.

2. Trend signal — why now?

This is not speculative. The numbers are screaming:

• MSME exports have tripled in 4 years — from ₹3.95 lakh crore in FY21 to ₹12.39 lakh crore in FY25 (source: SignalX.ai). More small manufacturers are exporting than ever before, and most are doing it for the first time.
• DGFT is actively pushing MSME e-commerce exports. The government published an official “E-Commerce Export Handbook for MSMEs” and the Foreign Trade Policy 2023 explicitly positions MSMEs as priority beneficiaries. The infrastructure is being built (ICEGATE, e-BRC) but the tools to use it are stuck in 2015.
• AI is now good enough to parse trade documents. Shipping bills, commercial invoices, packing lists, HS code classification — these are structured document tasks that LLMs handle well. Two years ago this would have needed custom ML. Now it’s an API call.
• Existing tools are form-fillers, not co-pilots. ExpoMaster, Expodite, and Eformz help you fill out export documents manually. None of them auto-classify HS codes, check RoDTEP eligibility, validate against DGFT regulations, or flag compliance errors before submission. They’re digital versions of paper forms.

3. The opportunity

India has 63 million MSMEs. Of those, roughly 600,000 have an active IEC (Importer-Exporter Code). Most rely on customs house agents (CHAs) who charge ₹2,000-5,000 per shipment for documentation, often make errors that cause delays, and provide zero visibility into the process.

The gap: a first-time or small-volume exporter needs to produce a commercial invoice, packing list, shipping bill, certificate of origin, and GST LUT/bond — and get the HS codes right, claim RoDTEP correctly, and file through ICEGATE. Today, this either requires a CHA or hours of manual work plus anxiety about compliance mistakes.

An AI co-pilot that ingests a purchase order and auto-generates all required documents, flags errors, and guides the exporter through ICEGATE filing would save 3-5 hours per shipment and eliminate CHA dependency for straightforward exports.

4. Target market

• Primary customer: Indian MSME manufacturers and traders who export 5-50 shipments per month — textiles, handicrafts, auto parts, chemicals, agricultural products. Typically a business doing ₹1-20 crore in annual exports.
• Why they buy: They’re paying ₹2,000-5,000 per shipment to a CHA for documentation that’s often late or error-prone. At 20 shipments/month, that’s ₹40,000-100,000/month on documentation alone. A ₹4,999/month tool that handles it faster and more accurately is an obvious trade.
• Rough TAM reasoning: 600,000 active IEC holders. Roughly 100,000 are MSMEs exporting regularly. At ₹4,999/month, that’s a ₹6,000 crore ($720M) addressable market. Even capturing 0.05% is ₹3 crore ($360K) — and this is conservative because new exporters are entering the market rapidly.
• Why now for them: Government is making exports more accessible (GeM, ONDC for exports, streamlined IEC process), but the compliance complexity hasn’t decreased — it’s shifted from physical to digital. Someone who could rely on a CHA’s institutional knowledge now needs to navigate ICEGATE themselves.

5. Product sketch (MVP)

• Purchase order ingestion — upload a PO (PDF, image, or email forward), AI extracts buyer details, product descriptions, quantities, pricing, and Incoterms
• Auto HS code classification — AI suggests the correct 8-digit HS code based on product description, with confidence scores and alternatives. This is where 60% of compliance errors happen.
• Document auto-generation — commercial invoice, packing list, and shipping bill pre-filled from PO data, formatted per DGFT/customs requirements
• RoDTEP and incentive checker — automatically checks eligibility for export incentives (RoDTEP, MEIS successor schemes) based on HS code and destination, calculates expected benefit
• Compliance validator — pre-submission check against current DGFT notifications, GST export rules, and destination country import requirements. Flags issues before they cause delays.
• ICEGATE filing guide — step-by-step walkthrough for filing the shipping bill on ICEGATE, with field-by-field help (full API integration in v2 when ICEGATE APIs mature)

6. AI angle — what’s load-bearing

Three things that are impossible without AI:

1. HS code classification from natural language. India’s export tariff schedule has 11,000+ HS codes. A manufacturer describes their product as “cotton printed bedsheets, 200 thread count, for home use.” The AI maps this to 6302.21.10, checks against recent DGFT notifications, and flags if the classification has changed. Getting this wrong means incorrect duty, rejected RoDTEP claims, or customs holds. Today, exporters either memorize codes, consult a CHA, or guess.
2. Intelligent document extraction and cross-validation. A purchase order from a buyer in Germany looks nothing like one from a buyer in UAE. The AI normalizes both into a standard internal format, cross-references against the exporter’s GST registration, validates Incoterms consistency, and generates all downstream documents. This replaces 2-3 hours of manual data entry per shipment.
3. Regulatory change monitoring. DGFT issues notifications constantly — HS code reclassifications, RoDTEP rate changes, new export restrictions. The AI monitors these and proactively alerts exporters: “Your top product (HS 6302.21.10) had its RoDTEP rate revised from 2.6% to 1.8% effective April 1. Your next shipment’s incentive drops by ₹12,400.”

Remove the AI and you have ExpoMaster — a form-filling tool. The intelligence is the product.

7. Localization angle (if any)

This is India-only by design. The entire value prop is India-specific:

• Regulatory framework: DGFT, ICEGATE, GST export refunds, RoDTEP, certificates of origin — these are uniquely Indian systems. No global trade tool handles them well.
• Language: Interface in English + Hindi. Many MSME exporters in Surat, Tirupur, Ludhiana, and Moradabad are more comfortable in Hindi or regional languages.
• Pricing: ₹2,499/month for up to 10 shipments, ₹4,999/month for up to 30, ₹9,999/month unlimited. This undercuts CHA fees by 50-70% while providing better accuracy.
• Payment: UPI and net banking. Annual plan discounts for the price-sensitive MSME buyer.
• Distribution: Export Promotion Councils (EPCs), District Industries Centres (DICs), and MSME trade fairs are physical distribution channels that global SaaS ignores.

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

• Pricing: ₹2,499/month (10 shipments), ₹4,999/month (30 shipments), ₹9,999/month (unlimited)
• ACV: Blended average of ₹60,000/year (~$720) per MSME
• Rough math to $1M ARR: 1,400 MSMEs × $720/year = $1.008M ARR
• Rough math to $5M ARR: 4,500 MSMEs at blended $1,100/year (mix shifts toward higher plans as exporters grow volume) = $4.95M
• Expansion path: Customs broker partnership program (CHAs white-label the tool for their clients), import documentation module, trade finance/LC document preparation, expansion to other South Asian markets (Bangladesh, Vietnam) with similar export complexity

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

1. Free HS code classifier tool. Build a standalone web tool: paste your product description, get the correct HS code with RoDTEP rate and duty drawback eligibility. This is the #1 pain point and the most searchable. SEO-optimized for “HS code finder India”, “RoDTEP rate check”, “export HS code classification.” Target: 1,000 monthly users within 3 months, convert 5% to paid.
2. Export Promotion Council partnerships. India has 30+ EPCs (Apparel Export Promotion Council, Engineering Export Promotion Council, etc.). Offer a free workshop on “AI tools for export compliance” at 5 EPC events. These are attended by exactly the target customer — MSME exporters looking for help.
3. Tirupur and Surat clusters. Tirupur (textiles, 5,000+ export units) and Surat (diamonds and textiles, 3,000+ exporters) are India’s densest export clusters. A 2-person sales blitz — visit 50 exporters in each city, demo the HS code tool, close 10-15 each. These clusters have tight networks; 5 happy customers generate 20 referrals.
4. CHA partnership play. Customs brokers handle documentation for 50-200 clients each. Offer 10 progressive CHAs a white-label version at ₹999/month per CHA — they use it to serve their clients faster and keep the margin. Each CHA brings 50+ end-clients into the ecosystem.
5. LinkedIn and Twitter/X targeting. India’s export community is active on LinkedIn. Founders, export managers, and trade consultants post about DGFT notifications, RoDTEP changes, and compliance issues daily. Targeted content + the free HS code tool drives organic signups.

10. Build complexity — justification

Medium. Core stack: Next.js dashboard, Supabase backend, LLM API (Claude/GPT-4) for document extraction and HS code classification, and a rules engine for Indian trade regulations. The HS code classification needs a curated dataset of India’s customs tariff (publicly available from CBIC) as grounding data for the LLM. Document template generation is straightforward. The ICEGATE integration is initially a guided walkthrough (not API), keeping scope manageable. Two builders, 10-12 weeks. The main risk is getting HS code classification accuracy above 90% — this needs extensive testing against real export products.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Standard SaaS; assists with documentation, doesn’t file on behalf of exporter
Ethical — no harm / dark patterns	✅	Helps MSMEs comply with export regulations; reduces errors and delays
Market exists (evidence above)	✅	600K IEC holders, MSME exports tripled in 4 years, US incumbents proving pricing model
1–5 person team can build this	✅	2 builders for v1; standard web + LLM stack + curated tariff data
Launchable with <$50K / ₹40L	✅	LLM APIs + standard infra + travel for cluster sales visits

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	16/20	Exporters paying ₹2,000–5,000 per shipment to CHAs for documentation that’s often wrong. Pain is financial, recurring, and directly tied to revenue. First-time exporters especially feel it. Not maxed because experienced exporters have workarounds.
Demand evidence	15	12/15	MSME exports tripled in 4 years — hard data. US incumbents charging $75–399/mo prove the model. DGFT actively promoting MSME exports. Slight deduction: no direct signal that Indian MSMEs will pay for SaaS vs. continue using CHAs.
Build feasibility	15	10/15	Standard web + LLM stack. But HS code classification accuracy is a genuine engineering challenge — India’s tariff schedule is complex, LLMs can hallucinate codes. Needs curated grounding data and extensive testing. Not a simple build.
Distribution clarity	15	12/15	Export clusters (Tirupur, Surat) are named, dense, and networked. EPC partnerships are concrete. CHA white-label is smart. But MSMEs are slow enterprise sales — feet-on-street, not viral.
Revenue mechanics	15	12/15	ACV of $720 is solid. Directly undercuts CHA fees by 50–70%, so value prop is clear. 1,400 customers for $1M ARR is achievable. Main risk: seasonal volume fluctuation could drive churn.
Time to first revenue	10	5/10	Free HS code tool can launch in 2 weeks. But paid conversion requires full documentation suite (10–12 weeks build) + cluster sales visits. Realistically 4–5 months to first paying customer.
Defensibility	10	6/10	Regulatory specialization + HS classification accuracy improves with usage data. CHA partnerships create distribution lock-in. But a funded trade-tech startup could replicate. Government building its own tool is a nonzero risk.
Total	100	73/100

13. Qualitative modifiers

Founder-fit tags

technical-heavy · sales-heavy · domain-expertise-required

Needs strong LLM engineering for HS code classification, willingness to do sales blitzes in export clusters (Tirupur, Surat, Ludhiana), and deep understanding of Indian export compliance (DGFT, ICEGATE, RoDTEP, GST export rules).

Key assumptions to validate (3–5)

1. Assumption: HS code classification accuracy exceeds 90% using LLM + curated tariff grounding data. How to test: Build the classifier, test against 200 known product-code pairs across textiles, engineering goods, and chemicals.
2. Assumption: MSMEs will pay ₹2,499–4,999/mo for SaaS instead of continuing with CHAs. How to test: Demo to 30 exporters in Tirupur/Surat; count how many commit to paid pilot.
3. Assumption: Free HS code tool drives 1,000+ monthly users within 3 months. How to test: Launch, SEO-optimize, share in LinkedIn export communities; track usage.
4. Assumption: CHA white-label partnerships bring 50+ end-clients per CHA. How to test: Sign 3 CHAs on a pilot; measure their client adoption over 60 days.

Risk flags

1. HS code accuracy: A 5% misclassification rate means wrong duties, rejected incentive claims, or customs holds. Trust is everything in compliance — accuracy must be 95%+ before launch.
2. Slow MSME sales cycles: Convincing a textile exporter in Tirupur to change workflows requires in-person demos and hand-holding. CAC could be ₹5,000–10,000, which is tight at ₹2,499/mo.
3. Government builds it: ICEGATE is improving. If DGFT builds an AI documentation assistant, the tool becomes redundant. Government tools are historically bad at UX, but the risk is real.

14. Structured verdict

Score:                  73/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Full-stack dev with LLM experience + export trade domain knowledge; comfortable doing sales blitzes in Indian manufacturing clusters
Time to revenue:        4–5 months (free HS tool in 2 weeks, paid suite requires full build + cluster sales)
Capital to launch:      ₹20–30L ($25–35K) — LLM APIs + tariff data curation + infra + travel for Tirupur/Surat sales blitzes
Top 3 assumptions to validate first:
  1. HS code classification accuracy exceeds 90% — test 200 product-code pairs
  2. MSMEs will pay ₹2,499+/mo — demo to 30 exporters, count paid commitments
  3. Free HS code tool drives organic signups — launch and track 90-day usage
Kill criteria:
  - HS code classification accuracy below 85% after grounding data optimization
  - Fewer than 3/30 exporters commit to paid pilot
  - Free HS code tool gets fewer than 200 monthly users in first 90 days

15. Risks & unknowns — top 3 things that could kill this

1. HS code classification accuracy. If the AI misclassifies even 5% of products, the exporter gets wrong duty rates, rejected incentive claims, or customs holds. Trust is everything in compliance. The tool needs 95%+ accuracy before launch, which requires extensive testing with real export products across multiple categories. This is the single biggest technical risk.
2. MSME sales cycles are slow. Unlike a SaaS tool that a developer can sign up for in 5 minutes, convincing a textile exporter in Tirupur to change their documentation workflow requires demos, hand-holding, and often an in-person visit. CAC could be ₹5,000-10,000 per customer, which is tight against a ₹2,499/month plan. The CHA partnership route is the mitigation — but it adds a layer of dependency.
3. Government builds it. ICEGATE is slowly improving. If DGFT builds an AI-powered documentation assistant (they’ve published the handbook, they clearly care about MSME exports), the tool becomes redundant. Mitigation: government tools are historically terrible at UX and move slowly. But the risk is non-zero.

16. Next step — 1-week validation sprint

• Day 1-2: Build a free HS code classification tool. User enters a product description in plain English/Hindi, gets the 8-digit HS code, applicable duty rates, and RoDTEP eligibility. Use India’s customs tariff data as grounding for the LLM. Test accuracy against 200 known product-code pairs across textiles, engineering goods, and chemicals.
• Day 3-4: Share the tool with 3 MSME exporters (personal network or LinkedIn outreach in Tirupur/Surat). Ask them to classify 20 of their actual products. Measure: accuracy rate, time saved vs. manual lookup, and willingness to pay for a full documentation suite.
• Day 5: Evaluate — did HS code accuracy exceed 90%? Did the exporters say “this saves me real time”? Did anyone ask “can it also generate my shipping bill?” If yes to all three, proceed to build the full MVP. If accuracy is below 85%, the idea needs either better grounding data or a narrower product scope (e.g., textiles-only to start).