Clause 1: which event starts the clock
Every agent publishes a free storage term, and the 3 most common versions start the clock at a different event. Arrival at the warehouse is the obvious candidate, but a QC pass or a notification to the buyer can just as easily be the stated start.
That difference changes what you owe without changing 1 word you agreed to. If the clock starts at arrival and QC takes several days, part of the free window can be consumed before you have seen the photographs that decide whether you keep the item at all.
Search the terms for the event rather than the duration. Finding the phrase that names the start event takes about 2 minutes and tells you more than the number of free days, because a duration means nothing until you know when it begins.
Terms are also published in more than 1 place. A summary page, a help-centre article and the account terms can each describe the window differently, and where they disagree the version you accepted when you opened the account is the 1 that governs.
Find the event that starts the clock before reading the duration. The two only mean something read together.
Clause 2: per item, per parcel or per account
Storage terms are written either per item or per parcel, and the 2 readings diverge as soon as you order more than 1 thing. An item-level clock gives every warehouse line its own deadline; a parcel-level clock gives the box a single one.
The effect is largest for buyers who order in batches. Where 5 items arrive across 2 weeks and the clock runs per item, the first line can fall due while the fifth has not been checked yet, and the [merge and pack step](/steps/merge-parcel/) turns into a race rather than a plan.
- Per item: each warehouse line carries its own deadline, and lines can fall due on different days.
- Per parcel: the merged box carries 1 deadline, and merging early does not reset it.
- Per account: everything held shares a single balance, which is the reading least often spelled out.
There is a third reading, less common but real: some terms run the clock per account, treating everything held as 1 balance. That reading favours batch buyers and surprises single-item buyers, which is why the exact wording matters more than any summary an agent gives you in chat.
The arrival date is the figure worth writing down first, because every other reading depends on it. A buyer who can see that 4 lines landed on 2 different dates knows immediately which reading the account is likely to apply.
Ask which reading applies in writing, and ask it before the second item arrives. A 1 line answer at that point is free; the same answer requested later arrives alongside a charge you are already disputing.
Clause 3: what the wording says closing actually triggers
Terms describe the end of a free window in 1 of 3 ways, and the 3 are not equivalent. The first says a charge begins; the second says the item may be disposed of; the third says the warehouse may act at its discretion.
Disposal clauses deserve a 2nd reading. A term permitting disposal is usually paired with a notice requirement, and that requirement is where a buyer has any protection at all, so the question to ask is what notice the warehouse promises before it acts.
A charge clause is milder, and it is still the 1 that costs money first. Where a daily charge begins the day a window closes, the cost of delay grows with every day spent comparing shipping routes, and a slow decision quietly becomes an expensive one.
None of this is a reason to read a whole terms page. Search it for 3 words instead, namely charge, dispose and notice, then read the sentences those 3 sit in; they are usually 2 paragraphs apart and they answer the question between them.
Reading those sentences in 1 pass takes about 5 minutes and produces a clear picture of the day the window closes, which is 1 of the few genuinely useful 5 minutes in an agent order.
The first day after the window closes
The day after a free window closes is a billing event rather than a shipping event, and it is the same day whether 1 item is sitting in the warehouse or 5. The goods stay exactly where they are, nothing is packed, and the only thing that changes is the price of the next day.
That makes the day after a poor moment to begin planning, and the reason fits in 1 sentence: the decisions that matter, which are whether to ship, merge or return, are all cheaper to take on the last day of the window than on the first day after it, because on that last day you still have a choice rather than a bill.
2 things are worth doing on that day regardless. Ask for the current balance on the item in writing so the figure is dated, and confirm whether the item can still go back to the seller, because a return route can close on a different clock from the storage window.
The same day is also the moment to stop comparing shipping options, and there are usually 2 routes still on the table at that point. Once a charge is running, every extra day of comparison carries a price of its own, and the cheapest route on paper can stop being the cheapest route in practice.
- Ask for a dated statement of what is owed, rather than reading a dashboard figure on screen.
- Ask whether the return route to the seller is still open, which is a separate question with a separate answer.
- Ask what each next decision costs, so that shipping, merging and returning can be compared as 3 prices rather than 3 feelings.
Four misreadings that quietly cost money
Most of the money lost to storage terms is lost through 4 misreadings, and none of them involves a hidden fee.
- Reading a window as a deadline to ship, when it is usually a deadline to act on the item.
- Assuming a merge resets the clock, when a merged box usually inherits the oldest line date rather than a fresh one.
- Assuming a warning will arrive, when many terms make a reminder optional and place the tracking duty on the account holder.
- Treating the storage window and the return window as 1 clock, when they can be owned by different parties.
The second misreading is the expensive one. Where 3 items arrive on 3 dates and are merged, the combined parcel usually carries the deadline of the earliest line, so a merge can shorten the time you believed you had.
Misreading 3 is quieter because it depends on a message that may never come. A buyer who treats a reminder as a promise stops tracking dates, and the first notice they read is the one that already includes a charge.
All 4 misreadings share 1 root, which is treating a summary as a term. Summaries are written for readability while terms are written for enforceability, and the gap between the two shows up the first time a charge is actually applied.
Treat a reminder as a courtesy rather than a term. Track the dates yourself, or ask the agent to flag the earliest line.
How to pin the version that applies to your order
Terms change, so a post from 1 year ago describes 1 year ago. The version governing your order is the one published when you accepted it, which makes copying the useful habit and remembering the unreliable one; our own [term snapshots](/drift/) exist for the same reason.
3 pieces of evidence settle almost every later disagreement: the terms page as it read on the day you accepted it, the date your item was recorded as arrived, and any message in which an agent described the window in their own words.
- Save the terms page with a date in the file name, since a screenshot without a date proves less than a file with one.
- Note the start event you found in the wording next to the recorded arrival date of the item.
- Keep the agent reply that states the window, because a written statement is easier to rely on than a recollection.
Where the wording is ambiguous, ask rather than interpret. A message reading please confirm the date my storage charge begins, and which event starts it takes about 1 minute to write and produces an answer that is very hard to argue with later.
Finally, do the arithmetic on the whole decision rather than on the storage line alone. A charge that starts after a free window is 1 line in a total that also carries domestic delivery, merge work and international freight, so the cheapest day to act is a comparison rather than a rule.
The [consolidation topic](/table/consolidation/) keeps those moving parts in one place. Two habits finish the job here: save the terms with a date, and write the start event beside the arrival date, because those 2 lines turn a later question about charges into a reading exercise rather than an argument.
Flow dependency: the storage clock and the merge decision belong to different owners, so the first day after a free window closes is a billing event rather than a shipping event — the item does not move and only the price of the next day changes.
This note explains the mechanics; the list is where the items are.