What changed in agent pricing between last year and this one

Four recorded fee-model changes and their effect at two order values. We collected it on 2026-10-05, from 96 entries, and we state the limits as plainly as the finding.

Across the eleven agents we checked, fee models fall into three shapes — percentage, tiered and flat. Which one is cheapest depends on your order size, and the difference at $150 is larger than the difference between lanes.

Free storage windows run from 30 to 90 days, and the clock start differs: warehouse check-in at some agents, purchase at others. That difference is worth about a week, which matters more than the window length at the margin.

Published fee table

Whether the model is readable without opening a ticket.

Storage clock start

Check-in or purchase — a week of difference.

Consolidation policy

Whether it is offered, and whether it restarts the packing clock.

Claims documentation

Whether compensation terms are published rather than quoted on request.

Three source labels separate what we know from what we assume. Item prices are published figures; freight, duty and fee ranges are modelled from published terms; the transit figures are measured.

Measured

Stage timestamps and weights from the 96-entry sample, collected on 2026-10-05.

Community-reported

Patterns that recur in buyer discussions and that we could not reproduce ourselves.

Published

Read from carrier tariffs, agent fee pages and destination thresholds.

Not measured

Anything we could not source. Written as such rather than estimated.

The known biases are listed rather than hidden. Weight bands are assigned from category baselines rather than weighed individually — that is the largest source of error in the census and it is stated on every page that uses it.

Sample sizes are given for every range on this site. A range without a sample size is not a measurement, it is an opinion.

Most estimates fail because they treat the item price as the cost. Across the 96 entries we hold, the non-item segments on a hoodies/sweaters order run from $7.50 to $74.40 — a range wide enough to change the decision about whether to order at all.

The practical consequence is that optimising the item price is a different activity from optimising the order. A 10% cheaper item saves 10% of one segment; a packing change can move two bands of another.

A US express order reaches the door in a median of 11 days and a p90 of 21. The tail is not a slower version of the median; it has its own causes, and they are listed rather than left to chance.

Last-mile delivery accounts for the largest single block — five of 11 median days. The two customs steps, which buyers worry about most, account for a combined two to five days in the median case.

Holiday closures

The warehouse closes but the seller clock keeps running, which adds two to three days at the front.

Battery or liquid declaration

A different export route, adding two to four days.

Address failure

A failed delivery attempt adds two days and sometimes a redelivery fee.

Peak season priority loss

The economy lane loses queue position from late October.

Duty and the compensation ceiling are both anchored to the same declared number. Lowering it lowers both, by the same mechanism, at the same moment.

The parcels in our sample with a declared value near a threshold showed the widest estimate error, because the duty term moves non-linearly there. If your figure sits within 10% of a threshold, budget the upper bound.

Declare accurately and keep the order record. A plausible figure with evidence survives a claim; a precise figure without evidence does not.

The checklist is deliberately short. Eighteen items across five groups is enough to catch the defects that show up in practice, and short enough that it gets used on every order rather than only on expensive ones.

The verdict rule: one fail is a return; three queries, or any query on shoes and jackets, is a query; everything else passes. The lower threshold on shoes and jackets reflects how hard those defects are to remedy after shipping.

A verdict you wrote down before looking at the next photo set is worth more than one you reconstructed afterwards.

Deviation between estimate and final invoice has six identifiable causes. Only one of them — random inspection — has no preventive action.

A well-run order deviates by under 5%. Beyond that, one of the six causes above is usually identifiable after the fact — which is the point of listing them rather than treating deviation as noise.

The volumetric step is the largest single cause and the least disclosed. If your estimate came without packed dimensions, it is a lower bound.

Seasonality hits transit harder than it hits freight. The surcharge is the visible part; the queue is the part that costs you a week.

The economy lane often avoids the surcharge entirely and loses priority instead, so the saving shows up as time rather than money. That trade is worth taking in September and questionable in November.

Ownership is clear at seven of the eight handoffs. The exception is the QC verdict, where the warehouse produces evidence and the buyer produces the judgement.

Payment to seller

Agent owns the transaction; you own the accuracy of the buy request.

Seller to warehouse

Seller owns dispatch; agent owns follow-up.

Check-in

Warehouse owns registration; the storage clock starts here at some agents.

QC

Shared and therefore ambiguous — the only genuinely unowned step.

Packing

Warehouse owns dimensions; you own the consolidation instruction.

Export

Agent owns documentation; you own supplying complete details.

Clearance

Carrier owns the filing; you own the declaration.

Last mile

Carrier owns delivery; you own address accuracy.

The three longest tails sit at handoffs two, seven and eight. All three are addressable before payment: a 72-hour rule, an accurate declaration, and a postcode check.