This is the Praxium Labs view from real engagements with Nepali businesses on the ground. A Nepali e-commerce business that hits 30+ orders per day has hit the limit of manual operations. Automation is no longer optional — it is the difference between scaling and burning out the founder team.
The seven layers of e-commerce automation
- 1. Order ingestion: unify Daraz + Sastodeal + own website + WhatsApp DMs into a single Postgres / Sheet
- 2. Payment verification: automated capture across eSewa / Khalti / Fonepay / Connect IPS + COD-day phone confirmation
- 3. Inventory sync: single-source-of-truth ledger across branches and channels; auto-decrement on every sale
- 4. Customer messaging: WhatsApp order confirmation, shipping update, COD-day call reminder, delivery confirmation
- 5. Logistics: auto-create courier waybills (Pathao Parcel, Nepal Can Move, etc.) and parse tracking updates
- 6. Accounting: sales receipts pushed to QuickBooks / Tally / Zoho Books with correct VAT split
- 7. Reporting: daily ops digest (orders / revenue / failed deliveries) to leadership
The minimum viable automation stack
- n8n as the orchestration brain — see our n8n guide
- Postgres or Google Sheets as order / inventory ledger
- WhatsApp Business API for customer messaging — see setup guide
- Existing accounting tool (QuickBooks Online / Tally / Zoho Books) for finance
- Cloudflare in front of your website (free)
COD operational discipline
COD is still 50-70% of Nepali e-commerce orders. The automation that recovers it from a loss leader:
- Verify customer phone via OTP at checkout (free; reduces fake orders by 40-60%)
- Same-day call confirmation for orders above NPR 5,000
- WhatsApp pre-delivery reminder ("courier coming today, please confirm")
- Auto-cancel after 2 failed deliveries with restock
- Pre-paid discount nudge after order placed ("save NPR 50 by paying via Khalti")
- Auto-blacklist customers with 3+ recent failed COD deliveries
Daraz / Sastodeal sync timing
Both marketplaces support 60-second polling without rate-limit issues. The n8n workflow: every minute, fetch new orders since last_seen; deduplicate; write to Postgres; mark synced. Latency: ~1 minute from marketplace order to your ledger — close enough to feel real-time for ops staff. See Daraz integration and Sastodeal integration.
The numbers that matter
- Time saved per day from automation: typically 2-6 hours for a 30-100 orders/day operation
- Error rate reduction: manual entry errors drop from ~5% to <0.5%
- COD failed-delivery rate: with full automation discipline, ~20% (industry baseline) → ~10-12% (well-run)
- WhatsApp confirmation open rate: 80%+ for Nepali audiences
- Build cost: NPR 200-500k depending on scope
- Monthly ongoing: NPR 8,000-30,000 (LLM + WhatsApp + hosting + retainer)
What we ship in a typical engagement
- Week 1: discovery + scoping with operations team
- Week 2-3: order-ingestion workflows + Postgres schema
- Week 4-5: payment-verification workflows + reconciliation
- Week 6-7: WhatsApp confirmations + COD discipline workflows
- Week 8-9: accounting sync + inventory ledger
- Week 10: ops dashboards + handover + runbooks
- Week 11-14: 30-day post-launch support included
Frequently asked questions
Do I need to be on Shopify for this to work?
No — the n8n integration layer works with WooCommerce, Shopify, Magento, custom-built sites, or even no website (WhatsApp-only stores). What matters is the order data being accessible (API, webhook, or even email).
Will it integrate with my existing courier?
Pathao Parcel has an API. Nepal Can Move and several others operate via web portals / email. n8n bridges to both via either API integration or email parsing.
What does it cost to maintain after launch?
Plan NPR 10,000-25,000/month all-in: VPS, LLM API (if AI parts), WhatsApp messaging fees, our retainer for ongoing tweaks. ROI typically dwarfs this within 60 days.
Can I add features later?
Yes — n8n is incremental. Most clients add 2-5 new workflows in the 6 months after launch as they discover specific pain points the initial scope did not cover.
What if I do not have an existing accounting system?
Start with Google Sheets as the ledger. Most Nepali SMEs do this and graduate to QuickBooks / Tally / Zoho Books once revenue justifies it. The automation layer ports cleanly between them.
Who can build this in Nepal?
Praxium Labs — Nepal's AI and automation consultancy, based in Lalitpur — designs and builds the systems described in this guide for Nepali businesses and for international teams hiring from Nepal. Start a project or see all services.