The problem it solves
B2B logistics is where enterprise money quietly leaks. A single pharma shipment misses a hospital’s 11 AM dock slot, and the whole truck sits until the next day — dock-demurrage, driver overtime, a chargeback, and an SLA credit all fire at once. A distributor’s 24-ton line-haul rolls at 68% utilisation because trunk planning happens in a spreadsheet that no one trusts past Thursday. An e-way bill timestamp slips by an hour and the shipment gets held at an inter-state checkpoint. Customer promise was “next working day” — reality is “next week, maybe.”
The hardest part is that every one of these failures is recoverable if anyone knew in time. B2B freight teams do not lack effort; they lack signal. They inherit a patchwork of SAP, a legacy TMS, three carriers with different track-and-trace, a customer portal that is read-only, and a WhatsApp group for exceptions. When a shipment slips, the loss is eight handoffs and forty minutes away from anyone who could have intervened. By the time it surfaces, it has already cost the invoice — and usually, the customer relationship is already being renegotiated on the next QBR slide.
What it is
B2B Logistics Optimization is Shipsy’s vertical stack for enterprise B2B shipping — trunk, middle-mile, last-mile, and customer experience — wired together as a single operational loop. It pairs Astra for trunk and last-mile planning, Multi-Carrier Intelligence for 3PL allocation, Clara for customer-experience resolution, Nexa for freight-invoice enforcement, and Vera for carrier-dispute automation — all running over a unified B2B TMS.
What makes it different: it treats a B2B shipment as a multi-leg, SLA-bound, appointment-driven, compliance-regulated job — not as a single tracking number. Enterprise SLAs, dock appointments, e-way-bill compliance, and customer-specific settlement rules are first-class inputs, not afterthoughts. The platform is already live in premium B2B express networks, pharma distribution, industrial freight, and building-material supply chains.
It is also new-age in the sense that matters commercially: every workflow that used to sit in an ops person’s inbox — appointment confirmations, chargeback rebuttals, invoice reconciliation, customer ETA updates, dock-slot negotiations — runs as an autonomous agent with a confidence-tiered handoff to humans. The shipper’s organisational structure shifts from a large ops team managing exceptions manually to a smaller ops team managing agents and policy. This is not a dashboard story. It is a headcount-ratio story, and it is the single biggest reason enterprises choose to consolidate onto this stack.
Core capabilities
| Capability | What it does |
|---|---|
| Enterprise SLA enforcement | Each customer’s SLA tree — delivery window, on-time commitment, dock-appointment tolerance, reefer temperature band, document package — is modelled and enforced shipment by shipment. Breaches are predicted, not reported. |
| Appointment delivery orchestration | Dock-slot booking integrated with top customer portals (e.g. Carrefour, Walmart Retail Link, Reliance Retail Portal) or handled via agent-driven call/email to smaller receivers. Slot compliance tracked per shipment. |
| Multi-modal trunk + last-mile planning | Trunk legs (FTL, LTL, air, rail) planned jointly with last-mile legs. A single optimisation run balances line-haul utilisation against last-mile window adherence — not two disconnected decisions. |
| E-way bill + compliance automation | E-way bill generation, extension, and cancellation automated at the shipment level. Hazmat, FSSAI, GST, and ADR flows handled as workflow templates per customer and per lane. |
| Appointment-aware last-mile routing | Micro-Cluster Route Optimization treats a dock appointment as a hard time-window constraint with a validated geometry. The route is built around the appointment, not adjusted afterwards. |
| Hybrid own-fleet + 3PL orchestration | Trunk and last-mile allocated dynamically between own fleet, dedicated 3PLs, and spot carriers — via Multi-Carrier Intelligence. Volume-commitment tracking ensures contracted volumes are fulfilled without over-service. |
| Customer-experience resolution | Clara handles customer queries, NDR (non-delivery report) rescues, and proactive ETA comms for B2B customers — who want formal, consistent, documented communication. 17+ incident types auto-detected. |
| Settlement + dispute automation | Nexa reconciles carrier invoices against contracts and PODs; Vera handles disputes autonomously up to high-dollar confidence thresholds. At scale, 50%+ reduction in manual settlement effort is typical. |
| ePOD + document packaging | Electronic proof of delivery with signature, photo, and document capture. Customer-specific document packages (LR, invoice, temperature log, quality certificate) bundled at the right handoff point. |
| Returnable packaging intelligence | RPM tracking for crates, kegs, totes, and pallets — reconciled against customer accounts to prevent asset leak, a major profit drain in B2B distribution. |
| Customer portal + EDI | Read/write customer portals with role-based access (procurement, receiving, finance). EDI (850/856/810/214) handled natively. |
| Exception command surface | Atlas sits over the stack as the autonomous control tower — predicts breaches, triggers next-best actions, coordinates agents. The ops team manages by exception, not by dashboard. |
How it works
The stack is organised as four horizontal loops — plan, execute, serve, settle — wired together by a shared event bus and governed by Atlas. Plan generates trunk and last-mile routes against SLA and compliance constraints. Execute runs shipments across own fleet and 3PLs, with live visibility. Serve manages customer-experience resolution via Clara. Settle closes the financial loop via Nexa and Vera. All four loops share a common data fabric: one shipment record, one SLA tree, one set of events, one audit trail. No stitching. No CSV reconciliation at month-end.
The typical shipment flow moves through five substantive decision points — allocation, dock scheduling, pickup execution, line-haul/last-mile execution, and settlement. Each is autonomous up to a defined confidence threshold; each escalates cleanly when the threshold is not met. The design philosophy is closed-loop-by-default: the same event that creates a delivery exception is the event that triggers the recovery action, the customer communication, and the settlement update — simultaneously, not sequentially.
This is what “autonomous B2B operations” looks like in practice. A delayed trunk leg is not a phone call from a driver to a dispatcher to a customer-service agent. It is an event: Astra re-plans the last-mile window, Clara sends a calibrated ETA update to the receiving customer, Atlas surfaces the breach to the control tower with the already-taken corrective action logged, and Nexa flags the dock-demurrage clause in the carrier contract. All four happen inside 90 seconds of the event landing on the bus.
Proven outcomes
| Customer type & scale | Outcome |
|---|---|
| Premium Indian B2B express network, 49 cities, 3,500+ pincodes | 16-18% cost savings on trunk + last-mile; first-attempt delivery from ~75% to 90%+; autonomous CX resolution for B2B receivers |
| Global alco-bev leader, 70+ countries, 2M+ annual distribution trips | $25M+ carrier/vendor disputes autonomously resolved; 28% reduction in LSP excessive-stay payments; route-settlement automation from 70% to 90% |
| Global biotech scaling rare-disease therapeutics across 30+ countries | Multi-country clinical-supply visibility with 21 CFR Part 11 audit trail; settlement-effort cut 50%+ |
| UK’s largest newspaper distributor, £1.1B revenue, 24,000 retail locations | Dawn-window distribution SLA held across 24,000 delivery points; exception resolution shifted from phone-based to agent-based |
Integrations
- ERP: SAP ECC / S/4HANA, Oracle EBS / Fusion, Microsoft Dynamics 365, Infor, Sage — bidirectional for PO, shipment, invoice, GR
- WMS: Manhattan, Blue Yonder, Oracle WMS, Körber, native Shipsy WMS
- Customer portals / EDI: Carrefour portal, Walmart Retail Link, Reliance Retail, BigBasket B2B, Metro B2B; EDI 850/856/810/214 native
- Compliance: Indian GST + e-way bill portal, EU ADR, US DOT, GCC customs, FSSAI, FDA 21 CFR Part 11 (pharma)
- Carriers + 3PLs: 240+ pre-built via Multi-Carrier Intelligence; includes all major Indian (Delhivery, BlueDart, TCI, Safexpress), GCC, EU, ANZ, and global express carriers
- Telematics: Wialon, Wheelseye, Samsara, MiX Telematics, Geotab — CAN + GPS ingested for fleet analytics
- Data platforms: Snowflake, Databricks, BigQuery for B2B margin analytics and customer-level profitability
Deployment
Phase 1 — Discovery (weeks 1-3). Customer-tree mapping (top 20 B2B accounts), SLA audit, compliance-workflow inventory, carrier and dock-partner inventory. Baseline on cost-to-serve per customer, FADR, on-time percentage, and settlement-cycle length.
Phase 2 — Configuration (weeks 3-6). ERP + WMS integration, customer-portal connectors, e-way-bill automation, contract and rate-card loading, SLA-tree modelling per top account. Multi-carrier adapter framework configured for current 3PL panel.
Phase 3 — Pilot (weeks 6-9). Launch on one lane family (e.g. North India metro lanes) with one customer cohort (top 5 accounts). Shadow mode on allocation and settlement for the first two weeks; cut-over on week 8. Exit criteria: 10%+ cost reduction, 5%+ FADR lift, and customer-team sign-off on CX resolution SLAs.
Phase 4 — Scale (weeks 9-16). Lane-by-lane and customer-by-customer rollout. Governance: a weekly ops review in the first 90 days, moving to a monthly margin review. Most enterprises reach full-scale by week 14-16, with financial impact visible by week 10.
Security & compliance
- SOC 2 Type II, ISO 27001, GDPR
- 21 CFR Part 11, GDP, GMP for pharma-regulated B2B lanes
- Full audit trail on every agent decision — Clara, Nexa, Vera, Astra — with the confidence score that drove the action
- Three-tier confidence scoring; high-risk actions (disputes > $50k, contract overrides, customer-SLA waivers) are human-in-the-loop by default
- Regional data residency — EU data in EU infrastructure, India data in Indian infrastructure, GCC data in GCC infrastructure
- Role-based access on customer portals; EDI certificates managed and rotated
- PII minimisation across customer-facing workflows; consent-based contact in Clara
Case study callouts
Premium Indian B2B express network · 49 cities · 3,500+ pincodes
Greenfield deployment that paired trunk optimisation, micro-cluster last-mile, and multi-carrier orchestration on a single platform. 16-18% freight-cost savings and first-attempt delivery moving from ~75% to 90%+. Customer experience shifted from a call-centre model to an agent-backed model using Clara.
Global alco-bev leader · 70+ countries · 2M+ annual trips
Deployed Vera for carrier dispute resolution — $25M+ resolved autonomously — and RPM intelligence for keg and crate reconciliation. LSP excessive-stay payments cut 28%. Route settlement automation moved from 70% to 90%.
Global biotech · 30+ countries · rare-disease therapeutics
Multi-country clinical-supply operations with 21 CFR Part 11 audit trail across trunk legs, customs, and final-mile appointment delivery to hospitals and treatment centres. Nexa cut settlement effort 50%+; Clara handled proactive ETA comms with treatment sites.