We Lost £3,200 in One Month. Here's the Mistake That Caused It

I want to tell you about the worst month we had in the first year of running Exact Solution.

Not the slowest month. Not the most stressful month. The month where we looked at the numbers at the end and realised we had actually gone backwards — where the revenue coming in was less than the money we had spent, lost to returns, and written off in inventory depreciation.

£3,200. Gone. In 30 days.

And the worst part? We had seen it coming for three weeks and hadn't acted fast enough.

The Setup

We were growing. Month six, things were moving in the right direction — orders were up, our listings were getting traffic, and we'd just expanded into two new product categories. We felt good about where things were heading.
That feeling is dangerous.

When things are going well, you stop scrutinising the numbers the way you do when things are hard. You start making decisions based on momentum rather than data. We bought more inventory than we needed because the previous month had been strong. We assumed the return rate would stay where it had been. We didn't look closely enough at which specific products were coming back and why.

That was the mistake. Not one dramatic error. A series of small assumptions that compounded into a £3,200 problem.

What Actually Happened — The Breakdown

When we finally sat down and pulled apart where the money went, it came out of three places:

Returns processing: £1,100

Each electronics return costs between £30–65 to process when you factor in testing, refurbishment, repackaging, and depreciation. We processed 31 returns that month. At an average of roughly £35 per return in processing costs, that's £1,085 before we've even accounted for the revenue that left with each refund.

The return rate that month was 19% — nearly double our normal rate. We hadn't noticed it climbing through the month because we were watching revenue, not return rate. By the time the month closed and we ran the numbers, the damage was done.

Inventory depreciation: £1,400

We had bought aggressively going into that month based on the previous month's sell-through. The sell-through didn't materialise at the same rate. Devices sat in storage longer than planned. Only 48% of returned items are resold at full price — the rest get marked down, relisted at a lower grade, or written off entirely. We had 14 devices that came back in worse condition than they left, requiring regrading to B or C condition and repricing accordingly. That markdown hit was £1,400 across the batch.

Shipping costs on returns: £700

We offer free returns. It's commercially necessary — shoppers are more likely to complete a purchase when they know they can easily return a product, and in the refurbished electronics space, a 30-day return policy is a baseline trust signal. But free returns aren't free for us. That month, 31 return shipments at an average of £22.50 each was £697.50 in outbound shipping costs — for products we were receiving back, not sending out.

Total: £3,197. We rounded it to £3,200 because the exact number was too depressing to look at.

What Caused the 19% Return Rate

We dug into every return reason from that month. The pattern was clear once we looked:

41% — Battery performance below expectation

We had a batch of devices where battery health was sitting between 76–82%. We were listing them as Grade B with battery health disclosed — but the disclosure was in a spec table at the bottom of the page. Customers were buying without reading it, receiving a device that didn't last the day, and returning it.

The battery health wasn't a surprise. The position of the information was the problem.

33% — Cosmetic condition mismatch

Our Grade B description at the time said "light signs of use." That phrase means something different to every person who reads it. A customer who interpreted "light signs of use" as barely noticeable received a device with visible edge wear and returned it. Not because it was defective. Because the description created a different expectation than the reality.

26% — Changed mind / found alternative

These we couldn't have prevented. Some returns are just the cost of operating with a 30-day policy.

The first two categories — 74% of our returns that month — were entirely preventable. They were information problems, not product problems.

What We Changed

Battery health moved to the top of every listing.

Not in a spec table. Not halfway down the page. Line three of every product description. Specific percentage. Context around what that means in real daily use. We also set a hard internal rule — any device with battery health below 80% gets a replacement before listing, not after a return.

Grade descriptions became specific and concrete.

"Light signs of use" became: "Minor cosmetic wear visible on casing — edge scratches may be present. Screen in good condition. Fully functional across all features." We added individual unit photography so customers could see exactly what they were buying before they clicked.

We built a weekly return rate alert.

Every Monday morning, the previous week's return rate appears in our internal dashboard. If it crosses 10%, we investigate immediately — not at the end of the month when the damage is done. The alert system cost us an afternoon to build. It has saved us from repeating that month multiple times since.

The Number That Still Bothers Me

£3,200 was the cost. The number that bothers me more is the 3 weeks.
For three weeks that month, the return rate was climbing. We had the data. We just weren't looking at the right metric at the right frequency. We were watching revenue — which looked fine on the surface because we were still doing volume — while the underlying return rate was quietly eating the margin on every sale.

Most teams fall down at the accounting reconciliation stage — if returns are not tracked in real time, the P&L overstates revenue and understates costs, which means the management accounts overstate margin. That is exactly what happened to us. The revenue line looked healthy. The margin line told a different story.

The lesson isn't complicated: watch return rate weekly, not monthly. In a physical products business where each return costs £30–65 to process, a two-week lag in catching a problem is a very expensive two weeks.

What the Month Cost Beyond the Money

The £3,200 was recoverable. What was harder to recover was the review tone that month.

Customers who received a device with a battery that didn't match their expectation left reviews. Not furious reviews. Quiet, disappointed ones. "Battery not as good as expected." "Condition was acceptable but not what I imagined." Those reviews sit on your listing and affect conversion for months after the underlying problem is fixed.

We've fixed the process. The reviews from that month are still there.
That's the cost nobody puts in the post-mortem.

Three Things I'd Tell Anyone Running a Physical Products Business

Watch return rate weekly. Revenue can look fine while return rate is destroying your margin. By the time you see it in the monthly P&L, the damage has already happened.

Information position matters as much as information accuracy. We weren't hiding the battery health. We were disclosing it in the wrong place. Customers weren't reading it. The result was identical to not disclosing it at all.
The cheapest return is the one that never happens. Every £35 you spend on a return processing cost is £35 that would have stayed in the business if the listing had been accurate enough that the customer knew what they were buying. Invest in listing quality before you invest in anything else.

The £3,200 month was the most expensive lesson we paid for in year one. It was also the most useful one.

Exact Solution sells professionally refurbished MacBooks, iPhones, and laptops across the UK, Poland, and Europe — tested, graded, warranty backed.
https://www.exactsolution.com/collections/laptops