What it means
Picking is the act of retrieving items from storage to fill an order, and a picker can select the wrong product, size or quantity; the error may be caught before packing or reach the customer. An order-level picking measure counts an order as accurate only when all its picked items are right, so you divide accurately picked orders by all picked orders and multiply by 100, as Shopify presents in one picking-accuracy formula.
A line-level measure instead counts correct picks among all pick lines, which can be useful for locating process problems but gives a different percentage, since a ten-line order with one error still fails the whole-order test. State which measure the dashboard uses, because comparing a line-level 99% result with an order-level 99% result as though they were identical is misleading.
Also state whether the check occurs before packing or after delivery. A fictional warehouse picks 100 orders and finds 4 with picking errors, so its order-level accuracy is 96%, and if each order has several lines the line-level result could be quite different.
Order accuracy is broader: the right items may be picked but the package could be mislabelled, damaged or sent to the wrong address, and Shopify's order-accuracy discussion covers errors across fulfilment and delivery. A return does not always prove a picking mistake, since a customer might change their mind or report a product defect, so code reasons carefully before using returns as the only error source.
Errors caught by a packing check still matter for picking performance because they caused rework, so record both detected picking errors and customer-visible errors, with every correction leaving a trace of the original pick, the error, the action taken and whether the customer was affected. Barcode scans, clear shelf labels and location controls can reduce wrong-item selection, and master data should be audited as well as worker behaviour.
Similar-looking products can be placed apart or marked clearly, size and colour variants deserve extra care, and a process change should be tested for both speed and accuracy. Warehouse managers often balance throughput with correctness, since a faster pick rate is not a win if it causes more rework and replacement shipping, so track labour productivity alongside the error rate.
An unexpected drop might reflect a new worker, a location move, stockouts or a change in the audit method, so investigate before assigning blame; the metric is a signal about a process. Small samples produce noisy percentages: two errors among twenty orders produce 90% accuracy, while the same two errors among two thousand produce 99.9%, so show counts and volume with the percentage.
An audit that checks only a small sample should report its coverage and selection method, because a selectively checked sample cannot be treated as a complete census without qualification. Use order-level counting to reveal customer impact and line-level counting to identify how frequently individual selection steps fail, and do not average their percentages together.
Segment by shift, product family, storage zone and order complexity, since a concentrated error pattern can point to a mislabelled bin or confusing catalogue entry that broad warehouse averages conceal. Set targets based on customer needs, costs and realistic process capability, avoid importing a claimed industry benchmark without matching methods, and remember that a perfectly picked order can still be late, so a broader perfect-order measure needs additional factors and a clear definition.
In practice
Real-world examples.
Example
A warehouse flags a wrong size selected from two adjacent bins before packing. The packer catches it with a scan check, so the customer never sees the error. The team still records it as a picking error because it caused rework.
Example
A manager reports 96 correct orders out of 100 picked, or 96% order-level accuracy. The dashboard states that the check happens before packing and that the unit is the order. Colleagues at another site can then compare like with like.
Example
A team compares line-level errors and whole-order errors after introducing scanning. Line-level accuracy improves slightly, while the number of failed orders falls more clearly. The comparison shows both customer impact and how often individual picks fail.
Formula
Calculation
Order-level picking accuracy = accurately picked orders / total orders picked x 100
Line-level picking accuracy = correct pick lines / total pick lines x 100
Worked example. A fictional warehouse picks 100 orders containing 400 pick lines. Four orders contain errors, and those four orders account for 5 wrong lines.
- Order-level accuracy = (100 - 4) / 100 x 100 = 96 / 100 x 100 = 96%.
- Line-level accuracy = (400 - 5) / 400 x 100 = 395 / 400 x 100 = 98.75%.
- The two results differ because one wrong line fails a whole order, so label which measure a dashboard reports.Case study
Seen in the real world.
In this entirely fictional case, Maple Warehouse sees errors concentrated in two similar product variants. Its audit finds four inaccurate orders in a sample of one hundred. The team improves bin labels and scans at pick time, then compares later order-level errors, pick speed and customer complaints. The fictional result is not a benchmark.
Maple also checked whether the improvement cost speed. Pick rates dipped for two weeks while staff adjusted to scanning, then recovered, and replacement shipments fell. The manager kept reporting both order-level and line-level accuracy so that later changes could be compared consistently.
Watch out
Common mistakes.
- Mixing order-level and line-level percentages.
- Using all returns as a proxy for picking errors.
- Celebrating faster picks without checking rework or customer impact.
Questions
People also ask.
What counts as an accurate order?
Under an order-level measure, every picked item and quantity must match the order.
Is it the same as delivered-order accuracy?
No. Shipping, damage and addressing errors can arise after picking.
Should caught errors be included?
Yes, when measuring picking quality, even if a later check prevents customer harm.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related