ClassifyIQ is a research tool for customs brokers and freight forwarders working under the Harmonized Tariff Schedule. It speeds up the first-pass research and keeps the final call where it belongs — with the broker.
Why we built ClassifyIQ
Independent customs brokers spend hours every week turning thin, half-described product information into structured classification research. Vague invoice lines, missing component breakdowns, suppliers who can't explain what they're shipping — every entry starts from the same incomplete picture, and the broker still has to land on a defensible code.
We built ClassifyIQ to shrink that first-pass research effort. The tool reads what you have, suggests likely classifications with supporting reasoning, surfaces relevant duty rates and trade measures, and hands you a draft to review. The judgement stays with you.
What we believe
AI should assist, not replace, broker judgement. A model can generate options. It cannot sign an entry.
Classification must be explainable. Every suggestion should be defensible — to a client, to a colleague, and to CBP if it comes to that.
Outputs should show reasoning, alternatives, and notes for review. Not just a code. You should see why, what else was considered, and where to double-check.
Compliance tools should be affordable for small brokers. The independent practitioner shouldn't be priced out of useful tooling because they don't sit inside a Fortune-500 trade department.
What ClassifyIQ does not do
- We do not issue binding rulings. Only CBP does.
- We do not replace licensed customs brokers. We are research support, not a brokerage.
- We do not guarantee classification outcomes. Tariff schedules, agency rules, and trade measures change; every output needs to be verified against primary sources.
- We do not encourage blind reliance on AI outputs. Every report is a draft for your review.
How it works
The workflow is deliberately simple:
Product details → AI-assisted research draft → source review → broker decision → client-ready report.
You enter what you know about the product, upload supporting documents if you have them, and ClassifyIQ produces a structured draft with proposed HTS codes, reasoning, alternatives, and flags for things to verify. You review against the source material, make your call, and export a clean report for your client or your file.
A note from the founder
Hi — I'm Deepak. I built ClassifyIQ after watching enough classification work happen up close to notice the same pattern: careful brokers spending most of their time on research scaffolding rather than the actual judgement call. The challenge was often gathering enough structured information to make a defensible classification decision efficiently. Getting the picture clear enough to decide on the code was the hard part.
ClassifyIQ is, honestly, a research assistant. That's the pitch. It reads the product details, drafts the research, lays out the reasoning, and gives you something to push against. The expertise stays with the broker. The license stays with the broker. The accountability stays with the broker. The tool just clears the desk a bit faster.
If it saves you an hour on a tricky line item, it's done its job. If it helps a small brokerage compete with a larger one on turnaround, even better. I'd rather build something quietly useful than something loudly marketed.
— Deepak R, Sydney, Australia
Join the Beta
We're working hand-in-hand with independent brokers to shape ClassifyIQ.
Join the BetaYour shipment data is treated as confidential and is not used to train public AI models.