Softpiper Solutions

Logistics · Customer conversations

Shipment-exception desk and customer updates

Route and resolve “where is my shipment” queries with tools on your TMS and carrier APIs, send proactive WhatsApp updates and give ops a clear exception queue.

  • ClaudeAnthropic
  • GeminiGoogle
  • Typical timeline: 6–8 weeks

The problem

Delays and exceptions trigger floods of “where is my shipment” queries, and ops teams triage them across several TMS and WMS screens.

  • Customers chase updates the system already has.
  • Agents switch between TMS, WMS and carrier portals for every query.
  • Exceptions are found late, after the customer complains.

How we solve it

A desk where a fast model routes every query, a stronger model resolves it with tools on your TMS, WMS and carrier APIs, and customers get proactive WhatsApp updates before they ask.

It plugs into TMS / WMS, Carrier APIs, WhatsApp Business, BigQuery or Postgres and Slack, so nothing you rely on today is replaced. The architecture below shows each step and which model does the work.

Architecture

From your stack, through the AI layer, to numbers you can track

Each step names the model or system doing the work. Writes to your systems wait for a person to approve them.
Shipment-exception desk and customer updates. Your systems (TMS / WMS, Carrier APIs, WhatsApp Business, BigQuery or Postgres and Slack) connect to Softpiper's AI layer, which runs these steps: Route (Claude Haiku); Resolve (Claude Sonnet · tools); Voice line (Gemini Live API); Location context (Grounding with Google Maps); Proactive updates (WhatsApp Cloud API); Ops queue and alerts (Slack · BigQuery). Outcomes are measured as where-is-my-order contact rate, exception resolution time, on-time notification rate and agent handle time.

Your stack

  • TMS / WMS
  • Carrier APIs
  • WhatsApp Business
  • BigQuery or Postgres
  • Slack

AI layer

Softpiper AI layer

  • Claude
  • Gemini
  1. Route, using Claude Haiku

    Inbound messages and calls are classified by intent and urgency.

  2. Resolve, using Claude Sonnet · tools

    Status, ETAs and exceptions are read from the TMS, WMS and carrier APIs.

  3. Voice line, using Gemini Live API

    Drivers and customers call in and get answers by voice.

  4. Location context, using Grounding with Google Maps

    Addresses and delays get map context.

  5. Proactive updates, using WhatsApp Cloud API

    Customers get a WhatsApp update as soon as an exception is detected.

  6. Ops queue and alerts, using Slack · BigQuery

    Claims, damage and unresolved cases reach ops in Slack with full context; metrics land in BigQuery or Postgres.

Outcomes you measure

  • Where-is-my-order contact rate
  • Exception resolution time
  • On-time notification rate
  • Agent handle time

Why this model

Claude and Gemini, each where it leads

We choose per task on capability, cost and where your data must stay. Final versions are picked at kickoff.
ClaudeAnthropic

A small Claude model routes high volumes cheaply, and a larger one resolves exceptions by calling your TMS, WMS and carrier tools in sequence.

GeminiGoogle

The Gemini Live API runs a voice line for drivers and customers, and Grounding with Google Maps adds location context. Teams on BigQuery can use Gemini's data insights for ops dashboards.

Delivery

What you get, how long it takes, and how we keep it safe

Guardrails

  • ETAs come only from system data, never from the model.
  • Claims and damage cases always go to a person.
  • Proactive messages follow opt-in and template rules.
  • Every action is logged against the shipment.

Plus the guardrails on every build: facts from your systems, approvals for writes and money, a human one step away, audit logs and evals. See all

Deliverables

  • Routing and resolution agents
  • TMS/WMS and carrier tools
  • Proactive WhatsApp updates
  • Optional voice line
  • Ops exception queue and dashboard

Typical timeline

  1. Routing, resolution tools and updates6–8 weeks

A planning range for a first production release. We confirm scope, price and dates in a tailored proposal.

KPIs to track

  • Where-is-my-order contact rate
  • Exception resolution time
  • On-time notification rate
  • Agent handle time
  • Customer satisfaction

Goals we agree with you and measure from a baseline. They are what we track, not results we promise.

Related services

The services behind this playbook

More playbooks

Related playbooks

All playbooks

Pip · Softpiper's AI

Ask Pip about this playbook

Ask Pip how this would work for you

Pip is Softpiper's AI. Tell it what you run today and it will walk through how this playbook fits, what we would build first and a tailored price range. Or book a call with the team.

Book a call