Picking accuracy
A warehouse picks 10,000 order lines in a week and 40 of them are wrong: the wrong item, the wrong quantity or the wrong batch number. That leaves 9,960 correct lines out of 10,000, which is 99.6 per cent. The share of picked lines that match the order exactly is called picking accuracy.
How picking accuracy is calculated
The sum is the number of lines picked correctly, divided by the total number of lines picked, multiplied by one hundred. The arithmetic is the easy part. The hard part is how you find out that a line was wrong, because that decides what you are measuring.
An error comes to light in three places: at the packing bench, where a check weighing or a second scan catches it; in your own goods-in, when a customer sends the wrong item back; or in a complaint, where you hear about it and the item never comes back. Most warehouses see only the first two. Write down which sources you count, and expect the figure to fall the first time you add one.
There is no good number
A fasteners distributor with four hundred nearly identical parts in neighbouring bins is working against the physics of the job. A book wholesaler, where every title carries its own barcode and looks nothing like its neighbour, is not. Both can be run tightly, and their figures will not match.
The comparison that gets you somewhere is the one with yourself: this quarter against the last, and one zone against another in the same building. A zone that slips while the rest holds steady is a far more useful signal than the site average.
Read the figure beside picks per hour. Accuracy and speed pull against each other, and a number that improves while the other collapses has not improved anything.
Worked example
Of the 10,000 lines picked in the week, 40 were wrong. 9,960 divided by 10,000 is 0.996, so picking accuracy was 99.6 per cent.
Now split those 40. Twenty-eight were found at the packing bench and corrected before the parcel left the building. Twelve reached the customer. Measured at the pick face, accuracy is 99.6 per cent; measured as the customer experienced it, 9,988 of 10,000 lines were right, which is 99.88 per cent. Both are true, and the two error rates differ by a factor of three. A report that does not say which figure it means says nothing at all. These figures are an example.
What moves the number
- Scanning at the pick face. Confirming every line against a barcode catches the wrong item before it travels. Picking from a paper list and signing off the whole order in one step does not.
- Where the look-alikes sit. Two sizes of the same part in neighbouring locations will be swapped sooner or later. Moving them apart costs less than training people not to mix them up.
- Labelling and light. A location label that needs a picker to bend down, or a dark aisle at the end of the racking, produces errors that look like carelessness.
- Two batches in one location. Two batch numbers in the same bin turn the right pick into a question of which box the hand reaches for first. See stock accuracy.
- End-of-day pressure. The last hour before the carrier collection produces most of the week's errors. A wave plan that finishes earlier is an accuracy measure.
- New and temporary staff. Peak season brings in people who do not know the building. Errors cluster around them, and the answer lies in how they are rostered and what they are allowed to pick, not in the number.
Frequently asked questions
Do you count lines or orders?
Lines, for this figure. One wrong line in an order of ten lines is one error in ten picks, not one order in one. If you count whole orders you arrive at the perfect order rate, which is a different and stricter figure.
What counts as an error?
The wrong item, the wrong quantity, the wrong variant, or the wrong batch or serial number where that matters. Damage caused during picking usually counts too, because the result for the customer is the same.
Should errors caught at the packing bench count?
Yes, in this figure. An error caught at packing cost someone time, and it tells you what happens at the pick face. Keep a second figure for errors that reached the customer, and say clearly which of the two you are quoting.
Why is our figure worse since we started scanning?
Almost always because you now find errors that you used to ship. A figure that falls as the measurement improves is a change in the measurement, not a decline in the work.
Ready to see BizBloqs on your own process?
Book a demo and we will walk your warehouse and order flow end to end — inbound, storage, picking, shipping, returns — and tell you honestly what BizBloqs would change.
Two questions about your own operation
How many of your articles sit in a pick location they empty less than once a month?
Part one: Laying out a warehouse does not start with the racking
After your last integration, how many systems hold a stock quantity for the same item? Name them.