What Your Refund and Void Rate Is Telling You

At close, the day's total looks healthy, and yet the drawer doesn't quite reconcile. Most of the time the gap isn't theft or a counting error — it's two numbers almost nobody reads back: refunds and voids. Every return handed to a customer and every sale wiped before it was rung reduce what the day actually earned, but they vanish into the net total. You see what's left, never what walked out. Until you look at them on their own, they keep nibbling at margin without ever showing up as a line worth watching.

Both numbers are already sitting in your Clover data. Every refund and every void carries a timestamp, an item, an amount, and often the employee who entered it. The trouble isn't collecting them — Clover does that — it's reading them as something other than one-off events. A single refund tells you nothing; the same item refunded week after week tells you plenty. The right lens is the rate: how many of your sales come back or get wiped, measured against how many you make. That ratio is what turns an ordinary incident into a useful signal.

Why the rate matters more than the count

A refund on its own looks like a simple courtesy: an unhappy customer, money back, move on. But add them up over a month and set them against sales, and those courtesies form a pattern. A product that comes back on two percent of its sales is a very different animal from one that comes back on one in six — and the second costs you twice, once in the margin lost on the reversed sale, and again in a customer who leaves with one less reason to return. The raw refund count won't tell you where you stand, because a high-volume seller refunded now and then is normal, while a small item refunded constantly is not.

Voids tell a different story, one less about the product than about what happens behind the counter. Cancelling a sale before it's tendered is often perfectly legitimate — the customer changes their mind, an item scans twice, someone fixes a keying error. But when voids cluster on one register, one shift, or one employee, they usually point to a process worth revisiting: a workaround learned wrong, a register set up badly, or, less often, something that deserves a real look. The rate is what separates the noise from the pattern.

The signs worth catching

The first sign is a sales total climbing while the cash in the drawer doesn't keep pace. That's the classic symptom of refunds you aren't tracking: the gross looks strong, the net says otherwise. Another is the same item coming back to the counter more often than the rest — you often feel it before you measure it, but the impression stays vague until it's written down as a number. On the void side, be wary of one register or one time slot producing noticeably more than the others at similar volume. It isn't necessarily a problem, but it's always a question worth asking.

How to read refunds and voids in Clover

You're looking at the same data two ways: refunds by item, and voids by register or employee. The by-item view tells you which products disappoint enough to come back; the by-register view tells you where, in the flow of a sale, things get wiped. A month of ordinary trading is enough to see the shape — just avoid a month marked by a big event or a product recall, or read that stretch separately.

1. Pull last month's refunds and group them by item, most refunded to least.
2. Set each item's refund total against that item's sales — the rate is what matters, not the count.
3. Pull the same month's voids and group them by register, employee, or hour.
4. Find the register or slot voiding noticeably more than the rest at similar volume, and ask the question before you conclude.
5. Note the two or three items with the highest return rate — those are the ones that deserve a decision, not a general worry.

You don't need all of it at once. Ranking refunds by item and spotting your single heaviest-voiding register is usually enough to surface one concrete action. The goal isn't a perfect audit of every dollar handed back, but knowing which product and which process cost you most, so you act there instead of everywhere.

The mistake: judging these numbers by total

The most common mistake is glancing at the month's refund total, deciding it's "not that big," and closing the report. The total lumps a high-volume seller refunded normally together with a small item refunded abnormally, then buries them in a single figure that calls for no action. The other half of the mistake is treating refunds and voids as the same thing. One is mostly about your products, the other mostly about your process; blending them means answering a dress-sizing problem and a register problem with the same fix. Separate them, set them against sales, and the pattern the total was hiding comes into view.

One thing to do this week

Pull last month's refunds and rank them by item. Take the top two or three and, for each, set the refunded total against the item's sales — anything coming back on more than one in ten purchases deserves a decision: a sizing note, a supplier change, a clearer description, or dropping the reorder. Then make one move on the void side: check whether a register or an employee produces far more than the rest, and ask the question rather than jumping to a verdict. A pattern you correct usually pays back faster than a month of refunds you simply absorb.

How CornerPilot helps

Reading refunds and voids by hand once is doable; watching how they move — an item whose return rate climbs after a supplier switch, a register that starts voiding again — is the tedious part in raw Clover reports, and most merchants do it once and then forget. CornerPilot connects to your Clover data on a scheduled sync and lays out your refunds by item and your voids by register over time, so the pattern and its drift are there to read instead of rebuilt each month. It works from the refunds and voids Clover already records — it won't guess why a customer returned something or calculate margin for you; it shows clearly what's coming back, and where.

If you run more than one location, the same view compares rates store to store — an item that comes back a lot in one shop and not the other usually points to a difference in display or floor advice rather than the product itself. See how it works on the features page, or view pricing when you're ready.

What actually changes

Once you read these two numbers as rates, they stop being an end-of-day fact of life. You catch the product that disappoints before you reorder a case of it, you fix the register habit that inflates voids, and you stop restocking what comes back more than it earns. The net total will still be the figure you check at close. Your refunds and voids are what tell you what that total let slip along the way — and, once read, where to win it back.