
Logistics

Average shipping cost reduction delivered to client accounts
More client accounts analysable per analyst per week
8 weeks → 3 days
Time to first savings assessment for a new client
Seven Billion's models run on every shipment we touch, instantly identifying cost exceptions across hundreds of millions in parcel spend.

Chase Flashman
Co-Founder & CEO
ShipSigma

About
A logistics tech firm cut client shipping spend 25% on average, gave each analyst 4x the accounts and shrank savings assessments from 8 weeks to 3 days.
Industry
Logistics
Company size
10 – 50 employees
Founded
2018
Business Context
Business context and operating model
A fast-growing logistics-tech company outgrowing its own analyst bench The client is a US logistics technology company that helps high-volume parcel shippers reduce UPS and FedEx spend through invoice auditing, rate benchmarking, and carrier contract negotiation. Growth had been strong — a multi-year run on the Inc. 5000 fastest-growing companies list — and every new client meant another parcel invoice history to audit and another rate structure to model. Each savings assessment required an analyst to manually parse carrier invoices, benchmark rates against comparable shippers, and identify accessorial charges and surcharges the client was paying that they didn't need to be.
Problem Definition & Framework
Current State
Failure Modes
Decision Criteria
KPI Baseline
Problem definition and framework
Manual rate analysis capped how fast the business could grow A single client rate-and-route assessment could take an analyst the better part of two months from data intake to signed recommendation, most of it spent manually reconciling invoice line items against carrier rate cards and service-level agreements. As deal volume grew, the analyst bench became the constraint on revenue growth rather than sales pipeline — a good problem to have, but one that put a hard ceiling on how many client accounts the business could onboard in a given quarter.
Decision framed: Where is each shipper overpaying across base rates, accessorials, zones, dimensional weight, and service choices?
Framework used: Parcel invoice normalization, rate benchmarking, and route/service-tier optimization
Baseline mapped: current process, failure modes, handoffs, data owners, and decision latency.
Success criteria tied to business KPIs, adoption behavior, and operating cadence changes.
How it works
Before / After

Architecture

Process Flow

Data Pipeline

Results

Delivery Timeline

Engagement Method & Process Diagram
01 Discovery
02 Data Audit
03 Baseline
04 Prototype
05 Pilot
06 Production
07 KPI Monitoring
Engagement method and process diagram
Process diagram: Carrier invoices -> line-item classifier -> rate-card benchmark library -> OR-Tools route/service optimizer -> ranked savings opportunities -> analyst negotiation workflow
Engagement method: Invoice taxonomy design; billing-data ingestion; savings model validation; analyst review workflow; client assessment automation; savings accuracy monitoring.
Delivery pattern: discovery, data audit, baseline measurement, prototype, pilot workflow, production rollout, and KPI monitoring.
An automated rate-modelling and route-cost engine We built a parcel invoice parsing pipeline that ingests raw carrier billing data and automatically classifies every line item — base rate, fuel surcharge, accessorial, dimensional-weight adjustment — then benchmarks it against a rate library built from comparable shipper profiles. An OR-Tools-based route and service-tier optimiser layers on top, modelling the cost impact of shifting specific shipment profiles between carrier services or zones. The output is a ranked list of savings opportunities with the dollar impact of each, replacing what used to be a manually assembled spreadsheet with a system that regenerates the full assessment automatically as new invoice data arrives.
Architecture, Results & Measurement
Solution architecture, results, and measurement
From an eight-week bottleneck to a same-week turnaround Time to first savings assessment for a new client fell from roughly eight weeks to three days, and each analyst could carry roughly four times the client volume without a corresponding increase in headcount. Clients moved through the firm's own onboarding pipeline faster as a direct result. The underlying savings delivered to client accounts held at an average of 25% off baseline shipping spend — the automation did not trade accuracy for speed, it removed the manual bottleneck standing between analysis and delivery.
Measured outcome: 25%: Average shipping cost reduction delivered to client accounts
Measured outcome: 4x: More client accounts analysable per analyst per week
Measured outcome: 8 weeks → 3 days: Time to first savings assessment for a new client
Measurement loop: compare weekly operating behavior against baseline, surface exceptions, and review model/workflow drift with business owners.

We didn't build this to replace our analysts' judgment on a negotiation — we built it so they stop spending their time reconciling invoice line items and start spending it on the conversations that actually move a deal.
Head of Analytics, US logistics tech firm
KEEP READING
Explore more studies on AI, analytics, and enterprise intelligence.










