Orders Shipped
How many separate orders actually left the warehouse on a given day.
What it means
Orders Shipped is the number of orders dispatched to customers on a date. It counts orders, not items and not euros — an order containing 5 of something adds 1. It's dated on dispatch rather than on purchase, so an order placed on the last day of a month can land in the next one.
Show the math
Formula and a worked example
An order counts once, on the day it shipped, whatever was inside it. That makes it the parcel count rather than the demand count — Units OrderedUnits OrderedThe volume side of demand, before value. on the same table is the demand side, dated when the customer bought.
Two ratios do the work here, and neither is a column you can read directly.
Worked example. A week ships 1,200 orders and 1,860 units on Shipped Sales of €54,000. That's 1,860 ÷ 1,200 = 1.55 units per order and €54,000 ÷ 1,200 = €45 an order.
The next week ships 1,200 orders again — an identical headline — but 1,400 units on €36,000. Now it's 1.17 units and €30 an order. Same number of parcels, same picking and packing, a third less money.
It answers the question
How many parcels did you actually put out? It's the figure to plan fulfilment capacity and packaging against, because a parcel takes the same handling whether it holds 1 item or 5.
Why it matters
Parcels are what cost you time. Every order shipped is a pick, a pack, a label and a handover, no matter what's inside it, so a month that ships 20% more orders for the same revenue is a month that got busier and no richer.
It's also the honest way to read the ratio nobody prints. Divide Units ShippedUnits ShippedUnits shipped on this date by this and you get items per parcel. When that number drifts down, the same revenue is arriving in more boxes, and the cost of the boxes doesn't drift down with it.
What good looks like
There's no number to aim at — this is a count, so a good figure for a 20-SKU catalogue is a poor one for 200. Judge it against your own recent weeks and against the same weeks last year. The more useful readings are the two ratios above: items per parcel, and Shipped Sales per parcel. Both are stable in a healthy month and both move before the totals do.
How to improve it
| Lever | What you do | Expect | How long | Watch out for |
|---|---|---|---|---|
| Fast Clear the stockouts on your top movers | Reorder the SKUs that hit zero and raise their reorder points so they stop hitting it | Orders return to the pre-stockout run rate within a cycle | 2–4 weeks | More cash sits in inventory, and the slower lines start picking up long-term storage charges while they wait. |
| Fast Win back the listings that lost the Buy Box | Check Buy Box % on the same table and fix the price, stock or seller issue behind the drop | Orders recover on those SKUs within days | 1–2 weeks | Winning it back usually means matching a price. Orders come back on a thinner margin, which shows up in Net Proceeds rather than here. |
| Slow Convert the traffic you already have | Improve the main image, bullets and reviews on listings with sessions but weak Unit Session % | More orders from the same page views | 4–8 weeks | Nothing moves for weeks, and when it does the gain is easy to credit to the wrong change. |
| Slow Add variants to listings that already sell | Extend sizes and colours on proven products instead of launching cold ones | More orders from traffic you already have | 1–2 quarters | Every variant needs its own stock, and the ones that don't sell sit in Amazon's warehouse charging storage against orders they never produce. |
Every lever costs something somewhere. The last column is the one to read twice.
Read it with
Parcels and items are two different volumes. Read them together and you get the size of the average order without anyone printing it.
Volume growing on both counts
More parcels and more items inside them. If units grew faster than orders the baskets got fuller, which is the cheapest kind of growth you can ship.
Busier for less
More parcels carrying fewer items each. Every one of those extra orders still costs a pick, a box and a handover, and the revenue per parcel has fallen.
Fewer, fuller orders
Multipacks or bundles landing. Good for fulfilment cost per unit, worth checking that the drop in orders isn't a supply problem wearing a flattering ratio.
Both falling away
Fewer parcels and fewer items. This almost always has a supply cause before it has a demand one, and the demand column is right there to rule it out.
One is dated when the customer bought, the other when the parcel left. That gap is your fulfilment lag, and it's the only way to tell a demand problem from a warehouse one. Units Ordered holding while Orders Shipped falls is a backlog, a late inbound shipment or a carrier delay — nobody changed their mind, the boxes just haven't gone yet. Both falling together is demand.
Divide one by the other and you have what an average parcel is worth. It's the number that catches a mix shift the totals hide: revenue and orders can both rise while the value per order slides, which means you're shipping more work for each euro than you were last month.
Common misreads
It's dated on dispatch. An order placed on the 31st and shipped on the 1st lands in the next month, which is why this column and Units Ordered disagree at every month boundary.
One counts parcels, the other counts items, so any order with 2 things in it pushes units above orders. The more multipacks you sell, the further apart they sit — and the gap between them is the useful number.
A stockout, a late shipment into the warehouse or a lost Buy Box %Buy Box %Higher is better; losing Buy Box usually hurts sales. produces an identical-looking drop with demand untouched. Check the order-dated columns before you change anything about the listing.
Also called
Shipped orders · orders dispatched · dispatch count · shipment count
See yoursYour orders shipped by date, next to the units and the sales that went out with them.
Open Amazon Summary →