The AI Chatbot for Freight Forwarders: Chat With Your Whole Company

Company Knowledge Lives Everywhere Except Where You Need It

Ask a forwarding office a simple question — what rate did we last quote this customer for Shanghai–Rotterdam FCL? — and watch what happens. Someone searches their inbox. Someone else checks a spreadsheet. A third person says the answer is with a colleague who is on leave. The information exists; it just lives in a specific inbox, a specific folder, or a specific person's head.

That is the daily tax of running a business on email and tribal knowledge. Every lookup is an interruption. Every answer depends on who is at their desk. And when the person who knows leaves the company, the knowledge leaves with them.

Zavin's chatbot exists to end that lookup tax. Because the platform already reads the company's email, creates its shipments, files its documents, and parses its rates, all of that operational knowledge sits in one structured place — and the chatbot is how anyone in the company asks it a question.

Ask Your Company Anything, in Plain Language

The chatbot works the way people already ask each other questions across a desk. No query language, no report builder, no hunting through menus — just a question, and an answer grounded in your company's actual records.

Questions operators ask it every day:

• •“What’s the status of the Hamburg shipment for this customer?” — answered from live shipment and milestone data.
• •“Which quotes are still unanswered from last week?” — answered from the quote pipeline, with links to each one.
• •“What rate did we get from our Ningbo agent for 40HC to Long Beach, and until when is it valid?”
• •“Which shipments arriving this week are still missing an HBL or a customs document?”
• •“Who is our contact at the destination agent in Rotterdam, and what did they last email us?”
• •“What Incoterm did we agree with this shipper — FOB or CIF — and where is that confirmed?”

What the Chatbot Can See

A chatbot is only as useful as the data behind it. Zavin's chatbot sits on top of the full operating system, so its answers draw on the same structured records the rest of the platform runs on: emails and their attachments, shipments and milestones, quotes and their outcomes, customer and partner records, and the company-wide rate database.

That includes the freight-specific detail that generic assistants miss. It knows the difference between an HBL and an MBL, reads the AWB number off a pre-alert, and understands that a booking on a Jebel Ali–Felixstowe lane under DAP terms carries different responsibilities than the same lane under FOB. When it answers, it cites the underlying record — the email, the document, the shipment file — so the answer is checkable, not just plausible.

Permission-Aware Answers, With a Full Audit Trail

The obvious risk of a company-wide brain is that it becomes a company-wide leak. Zavin closes that risk at the foundation: the chatbot enforces the same field-level access control as every other part of the platform — across every object, branch, and country.

Permission-aware means the chatbot never shows data the asker couldn't access themselves. A sales rep who can't see buy rates won't get buy rates out of the chatbot, no matter how the question is phrased. An operator in one branch can't pull another branch's customer list. Margin fields, salary data, and restricted accounts stay invisible to anyone without the right role — the chatbot answers from what that person is permitted to see, and nothing more.

Every question and answer is logged in an audit trail, so administrators can see who asked what and which records the answer drew on. Access rules are managed centrally: change a role or scope once, and the chatbot's behavior changes with it.

Track and Trace That Answers Itself, 24/7

“Where’s my container?” is the most-asked question in freight forwarding, and it usually costs an operator a lookup, a carrier site visit, and a reply email. With Zavin, that question is answered instantly, 24/7 — from live carrier and shipment data, without a human in the loop.

Internally, the team gets the same speed: vessel and ETA questions, milestone status, and exception checks are a one-line question instead of a system lookup. And because the chatbot works around the clock, a question asked from a different time zone at midnight gets the same answer it would at midday — which matters in a business where the origin office is asleep when the destination office is busiest. Customers feel it too: the tracking requests that used to queue overnight are answered the moment they arrive, and the operators who used to answer them get their mornings back for exceptions and quotes.

Institutional Knowledge That Survives Staff Turnover

In most forwarding companies, the real operating manual is unwritten. Which agent to use for project cargo into Dammam, how a key customer wants their pre-alerts formatted, why a certain lane always books through a specific co-loader — that knowledge lives in veterans' heads, and it walks out the door when they resign or retire.

Zavin changes what happens to that knowledge. Because the platform captures the company's work as structured records — every email, quote, shipment, and rate — the history stays queryable long after the person who handled it has moved on. A new hire can ask how a customer's last shipments were routed, what was quoted, and what went wrong, and get answers on day one instead of after months of shadowing.

The company brain doesn't replace experienced people. It makes their experience a company asset instead of a personal one.

Fed by the Email Core, Connected to Every Module

The chatbot is only as smart as the data flowing into it, and that flow starts with AI Email Automation — the core of the platform. As the email module reads and acts on inbound mail, it produces the structured records the chatbot answers from: RFQs become quotes, bookings become shipments, rate sheets become rate database entries, and attachments become filed documents.

The other modules deepen the picture. Shipment Management supplies live job files and milestones. Rate Intelligence supplies validities and lane history. The CRM supplies customer context and commitments. The chatbot is where all of it becomes conversational — one place to ask, whatever module the answer lives in.

Why a Legacy TMS or a Generic Chatbot Can't Do This

A legacy TMS holds some of this data, but it only knows what someone typed in — and answering a question means knowing which screen, which filter, and which module to open. The email history, where most of the real story lives, isn't in the TMS at all.

A generic AI chatbot has the opposite problem: it converses fluently but knows nothing about your company. Bolting one onto a document folder gets you summaries without operational truth — no live shipment status, no rate validities, no permission model, no audit trail.

Zavin's chatbot works because the operating system underneath it does the hard part first: capturing freight work as structured, permission-controlled data. The conversation layer is only trustworthy because the data layer is.

Live in Under 14 Days

The chatbot deploys with the rest of the platform: connect your inboxes, and the company brain starts building itself from real traffic — no data migration, no IT project. Most teams are live in under 14 days, and the chatbot gets more useful every week as more emails, shipments, quotes, and rates flow through.

Because Zavin works alongside CargoWise, Magaya, and other systems, you don't have to replace your existing TMS to start asking your company questions. Teams typically begin with tracking and quote lookups, then lean on it for history, documents, and handovers as coverage grows.