Why can't manufacturers rely on guaranteed, accountable pickup logistics for outbound deliveries?

Finished goods sit at the factory gate because the truck didn't show and nobody is accountable. Here's "Gate Pass", a dispatch desk that one person could build with Claude in a weekend.

The problem

Who feels it: Small and mid-sized manufacturers who ship finished goods every day through a mix of local transporters, brokers and booking apps. The dispatch clerk chases drivers on the phone, a missed pickup means a missed customer delivery date, and there is rarely a written record of which transporter failed and how often.

Logistics is a big line item for the country. A DPIIT report prepared by NCAER puts India's logistics cost at about 7.97% of GDP in 2023-24, with road transport at Rs 11.03 per tonne-km against Rs 1.96 for rail (KNN India).
Structure helps. In an IIT Delhi C-DEP study, 95% of surveyed MSMEs reported more on-time transport of goods and 73% reported lower transport costs after using tech-enabled goods transport services (Tribune).
Businesses are the main users. The same study says business users are about 71% of users and about 97% of intra-city goods orders on these services (Tribune).
The aggregate hides the last mile. An ISB professor writing on the study notes that the national figure masks persistent inefficiencies in intra-city movement, and that MSMEs account for nearly 45% of manufacturing output (Daily Pioneer).

Who's already working on it

Booking platforms supply the truck: BlackBuck for intercity trucking, Porter and similar apps for intra-city moves, and comparison guides such as Tata Nexarc's for MSMEs. Large shippers use transport management systems. We didn't find a simple tool that sits on the shipper's side across all transporters and keeps score on who actually turned up.

The solution idea

"Gate Pass" is a dispatch desk for small factories that books each pickup with a confirmed slot, chases the transporter automatically, escalates to a backup before the delay costs a delivery, and keeps a reliability scorecard for every carrier.

Slot bookingCreate a pickup from a sales order with weight, vehicle type, dock and time window, and send it to a preferred transporter for a written WhatsApp or email confirmation.

Auto-chase and escalationAsks the driver for vehicle number and ETA at set times; if the truck has not reported by the slot, offers the load to the next transporter on the list or a booking app.

Reliability scorecardRecords on-time arrival, no-shows, rate changes and damage per carrier, so allocation and rate talks rely on data.

Accountability clause packDrafts a simple service-level agreement with pickup windows, no-show charges and escalation contacts that a small factory can actually send.

Daily dispatch summaryOne message to the owner each evening: what left, what is stuck, and which customer needs a heads-up.

How Claude builds it

Which Claude capabilities do the work

Claude CodeBuilds the web app, the WhatsApp and email integration, the database and the tests.

Claude API in the productReads free-text replies from drivers and transporters, extracts vehicle number, ETA and issues, and drafts customer delay notices.

SkillsA "dispatch-playbook" skill with escalation rules, message templates in English and Hindi, and the SLA clause library.

ArtifactsA clickable dispatch board and scorecard mock-up to show factory owners before the build.

Connectors (MCP)Gmail and Google Sheets for pilots that already run dispatch in a sheet; Vercel to deploy.

Scheduled tasksSlot-time checks that trigger chasing and escalation, plus the evening summary and a weekly carrier report.

Weekend build plan

Saturday morning — interviews and data: Talk to three dispatch clerks. Collect a month of pickups from one factory's register or sheet, including no-shows.
Saturday midday — prototype: Ask Claude for an artifact with the dispatch board and scorecard. Show it to the clerks.
Saturday afternoon — core app: Use the starter prompt below in Claude Code to build pickups, transporters, slots and the status timeline.
Saturday evening — messaging: Connect WhatsApp Business or email, and have the Claude API parse replies into structured status updates. Test on real message samples.
Sunday morning — escalation: Add the scheduled slot checks, the backup-transporter ladder and customer delay drafts.
Sunday afternoon — scorecard: Compute on-time and no-show rates per carrier with plain code and unit tests. Add the SLA clause pack.
Sunday evening — ship and pilot: Deploy to Vercel and onboard two factories. Ask the clerk to log every pickup for two weeks.

Starter prompt for Claude Code

Build a Next.js + TypeScript + Postgres web app called "Gate Pass" for small Indian manufacturers.

Entities: Factory, Transporter (priority order, contacts, vehicle types), Pickup (order ref, weight, vehicle type, dock,
slot_start, slot_end, status), StatusEvent.

Flow: clerk creates a pickup -> app sends a confirmation request to the first transporter (WhatsApp Business API or email)
-> Claude API parses each free-text reply into JSON (confirmed, vehicle_number, driver_phone, eta, issue)
-> scheduled job checks 60 and 15 minutes before slot_start; if not confirmed or no vehicle number,
offer the load to the next transporter and alert the clerk
-> on arrival or no-show, record the event.

Rules:
- Scorecard maths (on-time %, no-show count, average delay) is deterministic code with unit tests, never model output.
- Never auto-cancel a confirmed truck; escalation only offers the load and asks the clerk to approve.
- Store every message with a timestamp so disputes have a record.
- Templates in English and Hindi live in /skills/dispatch-playbook.md.

Start by proposing the schema and file structure, then build pickup creation and the status timeline first.
Add fixtures of 20 real-style transporter replies and a test that checks the parser output.

Monetisation

Suggested pricing, not yet validated.

Free: up to 30 pickups a month with the scorecard.
Rs 1,499 a month per dispatch point: unlimited pickups, auto-chase, escalation and summaries.
Rs 4,999 a month for multi-plant factories with carrier reports.
Channel: referral fees from booking platforms when a backup load is placed with them.

Riskiest assumption, and how to test it

Assumption: The idea assumes that a written slot, automatic chasing and a visible scorecard change transporter behaviour. Small transporters may ignore a small shipper anyway when a better load appears.

Test first: Before building, run it manually for one factory for two weeks: a shared sheet, slot confirmations sent by message and a call at the 60-minute mark. Compare no-shows and average delay with the previous month's register. A reasonable bar: a clear drop in no-shows and a clerk who wants to keep the process.

Messaging drivers at scale needs WhatsApp Business API approval and consent. No-show charges only work if they are in a signed agreement, so check the clause pack with a lawyer.