Logistics teams at mid-market freight forwarders and enterprise shippers manually track shipments by logging into 12+ carrier portals daily, copying ETAs into shared Excel trackers, and refreshing browser tabs to catch exceptions. A logistics coordinator at a 3PL managing 400 active ocean shipments spends 2.8 hours per day (n=23, surveyed Aug 2025) switching between CMA CGM, Maersk, MSC, and Hapag-Lloyd portals to answer "where is my container?" questions from customers. The shipment goes dark between departure and arrival because carriers publish updates inconsistently — some every 12 hours, some every 3 days — and exceptions (port congestion, customs holds, vessel delays) surface 48-72 hours after they occur, when it's too late to reroute or notify the customer proactively.
This feature provides real-time multi-modal shipment tracking by polling carrier APIs and AIS vessel data every 15 minutes, calculating predictive ETAs using historical delay patterns, and triggering geofence alerts when shipments deviate from expected routes or timelines. The business case: 23 coordinators × 2.8 hrs/day × $28/hr (source: Glassdoor freight coordinator salary, India market) × 250 working days = $451,200/year recoverable + exception-driven customer escalations reduced from 18/week to <5/week (source: Q3 Zendesk tickets tagged "shipment delay"), saving ~$67K/year in customer success labor (13 hrs/week × $52/hr CS rate × 50 weeks). Total: $518K/year. If adoption reaches only 50% of target coordinators: $259K/year — still 3.7× the estimated 6-month build cost ($70K all-in, using Regional Cost Benchmarks: 2 India-based engineers × $4K/mo × 6 months + $22K infrastructure/tooling).
This is a unified tracking dashboard that aggregates carrier API data, AIS vessel telemetry, and historical delay patterns into a single pane with predictive ETAs, geofence alerts, and exception workflows. It is not a TMS, a booking platform, or a freight rate engine — any booking or rate negotiation happens in the customer's existing TMS or ERP, and GoTrack ingests container/booking references via API or CSV upload to begin tracking.
Project44 is hired to provide "control tower" visibility across ocean, air, and ground shipments by polling carrier APIs and EDI feeds, then surfacing exceptions in a dashboard — but customers report 18-24 week implementation timelines, complex API onboarding, and high cost ($40K+ annual contract minimums) that price out mid-market 3PLs. FourKites solves the same problem with deeper North American trucking coverage (70K+ carriers integrated) and real-time ELD data, but ocean/air coverage relies on carrier EDI partnerships that are slow to onboard and often miss regional carriers in APAC and EMEA. Shipa Freight (now part of Agility) offers shipment tracking as part of a digital freight forwarding platform, but it's bundled with booking and rate management — customers who already have a TMS can't adopt Shipa tracking standalone without migrating their entire booking workflow.
┌──────────────────────────────────────────────┬────────────────┬───────────────┬──────────────────────────────┐
│ Capability │ Project44 │ FourKites │ GoComet GoTrack │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ Multi-modal coverage (ocean/air/road) │ ✅ │ ✅ │ ✅ │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ Real-time AIS vessel tracking (not just API) │ ✅ │ ❌ │ ✅ │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ Predictive ETA using historical delay data │ ✅ │ ✅ │ ✅ │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ Geofencing + route deviation alerts │ ✅ │ ✅ │ ✅ │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ Standalone (no TMS/booking migration needed) │ ✅ │ ✅ │ ✅ │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ Self-serve onboarding (<7 days to first track) │ ❌ │ ❌ │ ✅ (unique) │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ CO₂ emissions tracking per shipment │ ✅ │ ❌ │ ✅ │
├──────────────────────────────────────────────┼────────────────┼───────────────┼──────────────────────────────┤
│ WHERE WE LOSE: Carrier network breadth │ ✅ 1,200+ │ ✅ 1,100+ │ ❌ ~340 carriers (Phase 1) │
│ (number of pre-integrated carriers) │ carriers │ carriers │ vs ✅ competitors │
└──────────────────────────────────────────────┴────────────────┴───────────────┴──────────────────────────────┘```
**Our wedge is self-serve onboarding and mid-market pricing** because Project44 and FourKites target enterprise customers with 18+ week professional services engagements and $40K+ contracts, leaving mid-market freight forwarders (50-500 employees, $10M-$100M revenue) underserved. GoTrack enables a logistics manager to upload a CSV of container/AWB numbers, auto-detect carriers via reference format, and begin tracking within 24 hours — no implementation project, no API keys to provision, no $40K minimum. We lose on carrier breadth (340 vs 1,200+), but we cover 89% of global ocean TEU volume (Maersk, MSC, CMA CGM, Hapag-Lloyd, ONE, Evergreen, Yang Ming, HMM, COSCO, PIL) and the top 12 air cargo carriers (source: Alphaliner Top 100, IATA World Air Transport Statistics 2024). For the mid-market customer tracking 200-800 shipments/month, breadth beyond top 20 carriers delivers diminishing returns — they'd rather have fast onboarding and $12K/year pricing than 1,200 carrier integrations they'll never use.
WHO / JTBD: When a logistics coordinator at a freight forwarder receives a "where is my shipment?" inquiry from a customer, they want to provide an accurate ETA and flag any exceptions (customs holds, vessel delays, port congestion) — so they can answer the customer in under 3 minutes and proactively notify them of delays before the customer calls.
WHERE IT BREAKS: Today, the coordinator opens the CMA CGM portal, enters the container number, waits 8-12 seconds for the page to load, checks the last movement event (often 36+ hours stale), then repeats the process across Maersk, MSC, Hapag-Lloyd, and 9 other carrier portals because the customer's shipment might be on any of them. If the shipment involves transshipment (40% of ocean shipments globally, source: UNCTAD 2024 Review of Maritime Transport), the coordinator must track the feeder vessel separately using a different portal or manually search MarineTraffic for AIS coordinates. There is no single source of truth: each carrier's portal uses different terminology ("Departed", "Vessel Sailed", "Gate Out"), shows different granularity (some show gate-in/gate-out at port, some only show "In Transit"), and updates on different cadences (Maersk: every 6 hours, regional carriers: every 24-48 hours). When an exception occurs (e.g., vessel diverted due to port congestion, customs inspection triggered), the coordinator learns about it only when the customer calls angry 2-3 days later, because carriers don't push alerts — the coordinator has to manually refresh portals to catch status changes.
WHAT IT COSTS:
┌─────────────────────────────────────────────────────┬───────────────┬──────────────────────┬─────────────────────────────────────────────┐
│ Metric │ Baseline │ Measured Method │ Aggregate Annual Impact │
├─────────────────────────────────────────────────────┼───────────────┼──────────────────────┼─────────────────────────────────────────────┤
│ Time spent per coordinator on manual portal checks │ 2.8 hrs/day │ n=23, time-tracking │ 23 coord × 2.8 hrs × $28/hr × 250 days │
│ │ │ study, Aug 2025 │ = $451,200/year recoverable labor │
├─────────────────────────────────────────────────────┼───────────────┼──────────────────────┼─────────────────────────────────────────────┤
│ Exception-driven escalations (customer calls CS) │ 18/week │ Q3 Zendesk tickets │ 18/wk × 43 min avg handle time × $52/hr CS │
│ │ │ tagged "shipment │ × 50 weeks = $66,924/year │
│ │ │ delay", n=178 │ │
├─────────────────────────────────────────────────────┼───────────────┼──────────────────────┼─────────────────────────────────────────────┤
│ Missed proactive reroute opportunities (carrier │ 4.2/month │ Aug-Oct 2025 ops │ 4.2/mo × $1,840 avg extra drayage/storage │
│ exceptions discovered >48 hrs late) │ │ incident log, n=13 │ × 12 months = $92,736/year │
└─────────────────────────────────────────────────────┴───────────────┴──────────────────────┴─────────────────────────────────────────────┘```
**Total quantified cost:** $610,860/year across 23 coordinators managing ~9,200 active shipments annually (400 concurrent, 21-day avg ocean transit). Business case equation: 23 coordinators × 2.8 hrs/day × $28/hr × 250 days + (18 escalations/week × 0.72 hrs × $52/hr × 50 weeks) + (4.2 incidents/month × $1,840 × 12) = **$611K/year recoverable value**.
**JTBD statement:** "When I need to respond to a customer inquiry about shipment location, I want to see real-time position, accurate ETA, and any active exceptions in one dashboard — so I can answer in under 3 minutes and proactively notify customers of delays before they escalate."
GoTrack exposes two primary API endpoints for customers to submit shipments for tracking and retrieve tracking events:
POST /api/v1/shipments — Submit a shipment for tracking
{
"reference_number": "MAEU123456789",
"reference_type": "container" | "booking" | "bill_of_lading" | "airway_bill",
"carrier_code": "MAEU" | "MSCU" | "CMDU" | null,
"origin_port_code": "CNSHA",
"destination_port_code": "USLAX",
"transport_mode": "ocean" | "air" | "road",
"customer_reference": "PO-2847-ACME",
"notify_email": "logistics@acmecorp.com",
"geofence_alerts_enabled": true
}
Response: 201 Created with shipment_id (UUID) and initial tracking status. If carrier_code is null, GoTrack auto-detects carrier from reference format (e.g., "MAEU" prefix → Maersk, "MSCU" → MSC). Polling begins within 60 seconds.
GET /api/v1/shipments/{shipment_id}/events — Retrieve tracking events
{
"shipment_id": "f47ac10b-58cc-4372-a567-0e02b2c3d479",
"reference_number": "MAEU123456789",
"carrier": "Maersk",
"transport_mode": "ocean",
"status": "in_transit" | "delivered" | "exception" | "delayed",
"current_location": {
"latitude": 22.5431,
"longitude": 114.0579,
"location_name": "Hong Kong",
"timestamp": "2025-06-14T08:23:00Z",
"source": "carrier_api" | "ais"
},
"carrier_published_eta": "2025-06-18T14:00:00Z",
"adjusted_eta": "2025-06-19T09:30:00Z",
"eta_confidence": "high" | "medium" | "low",
"events": [
{
"timestamp": "2025-06-12T04:15:00Z",
"event_type": "departed",
"location": "Shanghai, China",
"description": "Container departed from port",
"source": "carrier_api"
},
{
"timestamp": "2025-06-14T08:23:00Z",
"event_type": "position_update",
"location": "22.5431, 114.0579 (Hong Kong waters)",
"description": "Vessel at sea - AIS position",
"source": "ais"
}
],
"exceptions": [
{
"exception_type": "route_deviation",
"severity": "medium",
"detected_at": "2025-06-14T02:10:00Z",
"description": "Vessel 51nm off expected route for 6.2 hours - likely weather detour",
"acknowledged": false
}
],
"emissions": {
"co2_kg": 284.7,
"methodology": "GLEC v3.0",
"distance_km": 2588
}
}
┌─────────────────────────────────────────────────────────────────────────────────────────────┐
│ GoTrack Dashboard [+ Add Shipment] [Upload CSV] [Settings] [Help] │
├─────────────────────────────────────────────────────────────────────────────────────────────┤
│ │
│ Active Shipments: 387 │ Exceptions: 12 │ Delivered (7d): 94 │ Avg Delay: +22 hrs │
│ │
│ ┌─ Filters ─────────────────────────────────────────────────────────────────────────────┐ │
│ │ Mode: [All] [Ocean] [Air] [Road] Status: [All] [In Transit] [Exception] [Delayed] │ │
│ │ Carrier: [All carriers ▼] Origin: [All ▼] Destination: [All ▼] │ │
│ └───────────────────────────────────────────────────────────────────────────────────────┘ │
│ │
│ ┌─ Shipment List ─────────────────────────────────────────────────────────────────────┐ │
│ │ Ref Number │ Carrier │ Origin → Dest │ Status │ ETA │ CO₂ │ │
│ ├──────────────────┼──────────┼──────────────────┼─────────────┼─────────────┼───────┤ │
│ │ 🔴 MAEU48573921 │ Maersk │ CNSHA → USLAX │ Exception │ Jun 19 09:30│ 285kg │ → │
│ │ │ │ At Sea (HK) │ Route dev. │ +18h delay │ │ │
│ ├──────────────────┼──────────┼──────────────────┼─────────────┼─────────────┼───────┤ │
│ │ 🟡 MSCU7392847 │ MSC │ SGSIN → DEHAM │ Delayed │ Jun 21 14:00│ 412kg │ → │
│ │ │ │ Port of Singapore│ Port cong. │ +9h delay │ │ │
│ ├──────────────────┼──────────┼──────────────────┼─────────────┼─────────────┼───────┤ │
│ │ 🟢 157-38472093 │ Lufthansa│ FRALH → GBLBA │ In Transit │ Jun 15 06:20│ 1.8t │ → │
│ │ │ Cargo │ In flight │ On time │ No delay │ │ │
│ ├──────────────────┼──────────┼──────────────────┼─────────────┼─────────────┼───────┤ │
│ │ 🟢 CMDU9384756 │ CMA CGM │ USNYC → BEANR │ In Transit │ Jun 22 11:00│ 318kg │ → │
│ │ │ │ At Sea (N. Atl.) │ On time │ No delay │ │ │
│ └──────────────────┴──────────┴──────────────────┴─────────────┴─────────────┴───────┘ │
│ │
│ [1] [2] [3] ... [16] Showing 1-25 of 387 │
└─────────────────────────────────────────────────────────────────────────────────────────────┘
┌─────────────────────────────────────────────────────────────────────────────────────────────┐
│ ← Back to Dashboard Shipment MAEU48573921 [Acknowledge Exception] │
├─────────────────────────────────────────────────────────────────────────────────────────────┤
│ │
│ ┌─ Status Overview ──────────────────────────────────────────────────────────────────────┐ │
│ │ │ │
│ │ Container: MAEU48573921 │ Booking: 482938475 │ Customer PO: PO-2847-ACME │ │
│ │ Carrier: Maersk │ Vessel: MSC GÜLSÜN (IMO 9839738) │ │
│ │ Route: Shanghai, China → Los Angeles, USA │ │
│ │ │ │
│ │ 🔴 EXCEPTION: Route Deviation Detected │ │
│ │ Vessel 51nm off expected route for 6.2 hours (detected Jun 14 02:10 UTC) │ │
│ │ Likely cause: Weather detour or traffic separation routing near Hong Kong │ │
│ │ │ │
│ │ Carrier ETA: Jun 18 14:00 │ Adjusted ETA: Jun 19 09:30 (+18 hours delay) │ │
│ │ Confidence: Medium (historical delay data: 12 similar routes, avg +16h) │ │
│ │ │ │
│ │ CO₂ Emissions: 284.7 kg (GLEC v3.0, 2,588 km) │ │
│ └─────────────────────────────────────────────────────────────────────────────────────────┘ │
│ │
│ ┌─ Live Map ─────────────────────────────────────────────────────────────────────────────┐ │
│ │ │ │
│ │ Shanghai ●──────────────────●──────────────────────────────● Los Angeles │ │
│ │ 🚢 Current position │ │
│ │ (22.5431, 114.0579) │ │
│ │ Last updated: 8 min ago (AIS) │ │
│ │ │ │
│ │ Expected route corridor (50nm buffer) shown in gray │ │
│ │ Vessel currently 51nm outside corridor (route deviation) │ │
│ └─────────────────────────────────────────────────────────────────────────────────────────┘ │
│ │
│ ┌─ Event Timeline ───────────────────────────────────────────────────────────────────────┐ │
│ │ │ │
│ │ Jun 14 08:23 UTC │ Position Update (AIS) │ 22.5431, 114.0579 (Hong Kong waters) │ │
│ │ Jun 14 02:10 UTC │ 🔴 Exception Detected │ Route deviation - 51nm off course │ │
│ │ Jun 13 18:45 UTC │ Position Update (AIS) │ At sea, 180nm east of Hong Kong │ │
│ │ Jun 12 04:15 UTC │ Departed │ Shanghai, China (Yangshan Port) │ │
│ │ Jun 11 22:30 UTC │ Loaded on Vessel │ Container loaded onto MSC GÜLSÜN │ │
│ │ Jun 11 14:00 UTC │ Gate In │ Arrived at Shanghai port terminal │ │
│ │ │ │
│ └─────────────────────────────────────────────────────────────────────────────────────────┘ │
│ │
│ [Export Events CSV] [Set Alert] [Notify Customer] [Add Note] │
└─────────────────────────────────────────────────────────────────────────────────────────────┘
Auto-detection of carrier from reference format: Customers often don't know which carrier is handling a container (especially with co-loading or transshipment). GoTrack parses the reference number prefix (first 4 characters) against a lookup table of 340 carrier SCAC/BIC codes to auto-assign carrier. If ambiguous or not found, we poll the top 5 carriers by volume (Maersk, MSC, CMA CGM, COSCO, Hapag-Lloyd) in parallel until one returns a valid tracking result. This adds 8-12 seconds to initial tracking setup but eliminates a major onboarding friction point observed in competitor products ("user doesn't know if container is on Maersk or ONE because their booking was with a freight forwarder, not the carrier").
Adjusted ETA display priority: Carrier-published ETA is always shown alongside adjusted ETA (our predictive calculation). Adjusted ETA is visually emphasized (larger font, color-coded confidence badge) because internal testing shows carrier ETA is late 33% of the time by >12 hours on transpacific routes. If no historical data exists for the route, we display carrier ETA with a "No delay data available" label to set expectations — we do NOT hide carrier ETA or show only our prediction, because transparency builds trust during the initial 6-month period when we're collecting delay data.
Exception severity classification: Exceptions are classified as High (customs hold, vessel breakdown, port strike), Medium (route deviation, port congestion, weather delay >12h), or Low (minor schedule slip <6h, gate delay). High-severity exceptions trigger email + SMS alerts immediately; Medium triggers email only; Low is logged but doesn't push alerts unless the shipment is flagged as "high priority" by the user. This prevents alert fatigue — in pilot testing (n=8 beta customers, 340 shipments, Oct 2025), users acknowledged 87% of High alerts within 30 minutes but only 34% of Low alerts, and 3 customers requested the ability to mute Low alerts entirely.
Emissions calculation transparency: CO₂ figure is always accompanied by methodology label ("GLEC v3.0") and distance in km. This is non-negotiable: customers subject to CSRD or CDP Supply Chain reporting need auditable emission data, and hiding methodology reduces trust. We do NOT round emissions to whole numbers (display "284.7 kg" not "285 kg") because enterprise customers export this data to sustainability reporting tools that expect precision.
In scope (Phase 1):
Out of scope (Phase 1):
Decision 1: Polling Frequency for Carrier APIs Choice Made: Poll carrier APIs every 15 minutes for active shipments (status = "In Transit" or "At Origin"), every 6 hours for delivered/completed shipments. Rationale: Carriers rate-limit APIs (Maersk: 500 req/hour, CMA CGM: 300 req/hour). At 15-min intervals, we stay under limits for up to 1,875 concurrent active shipments per customer before needing to batch. Rejected: 5-min polling (would hit rate limits at ~600 shipments, blocking mid-market customers) and 30-min polling (too slow to catch exceptions — a 2-hour port delay wouldn't surface for 30+ minutes, reducing proactive alert value). Delivered shipments drop to 6-hour polling to conserve API quota for active tracking.
────────────────────────────────────────
Decision 2: AIS Data Source Choice Made: Primary source = MarineTraffic API (paid tier, $1,200/month for 50K position requests), fallback = VesselFinder API if MarineTraffic is unavailable for >10 minutes. Rationale: AIS fills the gap when carriers don't publish updates (e.g., vessel at sea between ports). MarineTraffic covers 99.2% of commercial vessels >300 GT (source: MarineTraffic coverage report, 2024) and provides 5-minute position updates. Rejected: free AIS sources (30-min update lag, 72% vessel coverage) and dual-sourcing every request (2× cost for marginal reliability gain). Fallback to VesselFinder adds redundancy without doubling baseline cost.
────────────────────────────────────────
Decision 3: Predictive ETA Algorithm Choice Made: Use historical delay data (24 months rolling window, grouped by carrier + route + season) to adjust carrier-published ETA. Formula: Adjusted ETA = Carrier ETA + (Median Historical Delay for [Carrier, Origin Port, Dest Port, Month]). If no historical data exists (new route), display carrier ETA with "Estimated — no delay data" label. Rationale: Carrier-published ETAs are optimistic (average 18% late on transpacific routes, source: SeaIntelligence Global Liner Performance report Q2 2025). Historical adjustment improves accuracy from 67% on-time (±24 hrs) to 84% on-time based on internal testing (n=340 shipments, Aug-Oct 2025 backtest). Rejected: ML-based ETA (requires 18+ months of training data we don't have yet; defer to Phase 1.2) and real-time weather/congestion API integration (adds $8K/month for Windward or MarineInsight API; ROI unclear until customer validates adjusted ETA value).
────────────────────────────────────────
Decision 4: Geofence Alert Trigger Logic Choice Made: Trigger "route deviation" alert if vessel's AIS position is >50 nautical miles from expected route corridor for >6 hours. Expected route = great circle path between last known port and destination port, ±50nm buffer to account for normal navigation. Suppress alerts if vessel is within 20nm of any commercial port (likely transshipment stop, not deviation). Rationale: 50nm threshold balances false positives (weather detours, traffic separation schemes) with real exceptions (unplanned port calls, sanctions evasion). 6-hour duration filter prevents alerts during temporary course changes. Rejected: 25nm threshold (generated 34% false positive rate in backtest due to TSS routing near Singapore/Suez) and 100nm threshold (missed 3 actual diversions in backtest where vessel diverted to alternate port 60nm off-route).
────────────────────────────────────────
Decision 5: Emissions Calculation Methodology Choice Made: Use GLEC Framework v3.0 (Global Logistics Emissions Council) for CO₂ calculation: Ocean = (TEU × distance in km × emission factor 0.011 kg CO₂/TEU-km), Air = (chargeable weight in kg × distance × 0.602 kg CO₂/kg-km). Emission factors are GLEC defaults; we do NOT customize by vessel type/age or airline fleet mix in Phase 1. Rationale: GLEC is the industry standard (adopted by Smart Freight Centre, CDP Supply Chain Program). Using default factors ships Phase 1 in 6 weeks; vessel-specific factors (requires integrating IMO vessel database + actual fuel consumption data) adds 8 weeks and delivers <12% accuracy improvement (source: Smart Freight Centre GLEC Framework white paper, 2023). Rejected: proprietary emission models (not auditable, reduces customer trust) and excluding emissions entirely (customers in EU are subject to CSRD reporting requirements starting Jan 2025; emissions data is table stakes for enterprise deals).
────────────────────────────────────────
Decision 6: Phase 1 Carrier Coverage Choice Made: Integrate 18 ocean carriers (covering 89% global TEU volume), 12 air cargo carriers (covering 76% global air freight tonnage), and 0 trucking carriers in Phase 1. Trucking deferred to Phase 1.2. Rationale: Ocean and air deliver the highest customer pain (2.8 hrs/day manual tracking, 18 escalations/week). Trucking is more fragmented (70K+ carriers in North America alone per FourKites) and requires ELD/GPS integrations, not just API polling — estimated 16 additional weeks. Mid-market customers track ocean/air internationally and hand off domestic trucking to regional 3PLs who already provide truck tracking. Rejected: including top 5 trucking carriers in Phase 1 (adds 12 weeks, delays ocean/air value) and reducing ocean carrier count to 10 (drops coverage to 78% TEU, misses regional Asia-Europe carriers like ONE and PIL that matter to APAC customers).