Canada Post vs Purolator: which lane actually gets your CUSMA claim applied?

This comparison is usually framed as rates and transit times. For a brand whose goods qualify under CUSMA, the more consequential difference is whether the preference claim reaches the entry and gets applied.

The claim has to survive the whole chain

A CUSMA claim needs the goods to qualify, a complete certification of origin to accompany the shipment, and the entry to be filed claiming preference. All three have to happen.

The first is about your product data. The second is about how your label tool generates the commercial invoice. The third is about the lane — how the entry is made and by whom, and whether the certification data you transmitted is used.

Why lanes differ

Postal and courier channels make entries through different processes with different documentation handling. What is transmitted electronically, what is read from a paper invoice, and who files the entry are not the same across lanes.

The practical consequence is that identical goods with identical paperwork can get preferential treatment on one lane and be assessed duty on another. This is not a conspiracy; it is a documentation-handling difference. But it costs real money.

Measuring it instead of guessing

Take your duty invoices for the last quarter. For each shipment of goods you believe qualify, record the lane and whether duty was assessed.

You now have a success rate per lane, from your own data. Brands doing this for the first time commonly find one lane performing much worse than they assumed, and the gap is usually larger than any rate difference they were optimising for.

What to do with the finding

First, fix the documentation. If your commercial invoice is missing a required data element, the claim will fail on any lane and the lane comparison is moot.

Then, if a lane still underperforms with correct paperwork, route qualifying goods away from it or escalate with the carrier using your own numbers.

Finally, re-measure. Documentation handling changes, and a success rate from last year is not evidence about this quarter.