How Much Cash Should You Hold for Customer Returns?

A typical situation looks like this. A clothing shop owner has finished a good sales month and wants to place a $20,000 stock order. Her bookkeeper has updated the cash forecast, and the payment seems to fit.

“The money’s already in the bank,” the owner says. “Why would we wait?”

The bookkeeper asks how much of last month’s sales customers can still return. Some orders have only just arrived at customers’ homes. The shop’s return window is still open, and the owner hasn’t put those possible refunds into the forecast.

For this hypothetical shop, the question is how much of the money already collected needs to stay available. A sale can be paid, delivered, and still send cash back out of the business a few weeks later.

The owner doesn’t need to freeze every dollar until the last return date. She does need an estimate grounded in the shop’s experience before committing to another purchase. The bookkeeper’s next job is to connect those earlier sales to the withdrawals they could still create.

Spending rose. That doesn’t settle what’s safe to spend.

The Bureau of Economic Analysis reported on October 29 that consumer spending rose 0.5% in September, measured in current dollars. After adjusting for prices, the increase was less than 0.1%. Those are national estimates across goods and services, subject to revision.

The obvious takeaway is that customers spent more dollars. An advisor still has to ask what happens to the money after this particular client’s sales reach the bank. The release doesn’t measure the shop’s return rate or establish that refunds are rising.

That distinction matters when an owner wants to use a strong sales month to fund the next order. Completed customer payments support a cash plan, but the plan also needs the refunds those sales may produce.

Start with the issue in front of you: the forecast has left out a possible withdrawal. Pull the shop’s actual return policy and transaction history before choosing a percentage. Look at comparable products and selling periods, especially if this month included a promotion or a different mix of merchandise.

Ask the owner which recent sales are still eligible for a refund. That gives the review a useful boundary instead of applying last year’s average to everything in the bank.

Match the refund to the sale that caused it

A monthly refund total can mislead you. Money returned in October might relate to purchases made in September. Dividing October refunds by October sales mixes two different groups of customers.

Follow a completed group of orders from the original sale through the end of its return window. Compare the dollars actually refunded with the original sales dollars on the same basis. Exclude canceled orders that never produced a receipt, and keep exchanges separate from cash refunds.

For a current group, subtract refunds already processed and estimate what remains. If recent sales are too new to reveal a pattern, use an older, comparable group whose return window has closed. Record why that comparison fits and what could make it unreliable.

Suppose the opening shop has $100,000 of collected sales still exposed to returns. Its history supports an estimate of another 8% coming back as cash refunds. That puts $8,000 of future withdrawals into the plan.

These are illustrative assumptions, not retail benchmarks or BEA findings. A percentage based on returned units won’t produce the same answer if expensive items come back more often than cheaper ones.

Check the amount the customer actually receives, including any refundable shipping or tax. The point here is the bank withdrawal; financial reporting and tax treatment need their own review. Keep amounts already refunded out of the remaining estimate.

Put the withdrawals where the bank will feel them

Use the 13-week cash forecast to see whether those refunds change the stock decision. Estimate payment weeks from the shop’s observed return timing and processing delay. Already approved refunds should use known payment dates where available.

For this example, assume the forecast’s lowest balance is $45,000 before the proposed purchase and remaining refunds. All ordinary receipts and payments are already included. The owner wants to keep at least $15,000 available for operations.

Assume the purchase and refunds both fall before that low point. Paying $20,000 for stock and $8,000 in refunds leaves $17,000. The order fits the stated floor with only $2,000 to spare.

Now test a remaining refund rate of 12%, perhaps because the current product mix has less history. Refunds would total $12,000, leaving $13,000 after the same purchase. That falls $2,000 below the owner’s floor.

The larger percentage is a stress case, not a prediction. Its value is showing how little variation the decision can absorb. Ask what evidence supports the lower estimate before calling the full order affordable.

Check how refunds move through the payment processor, too. If the forecast already uses deposits net of refunds, subtracting those same amounts again will understate cash. Reconcile the settlement report with the bank so each refund changes the forecast once.

Decide what can wait, then update the estimate

Under the stressed assumptions, reducing the purchase to $18,000 leaves $15,000 at the modeled low point. That gives the owner a specific alternative to discuss with the supplier. It works only if the smaller order is available and the rest of the weekly balances also hold.

Waiting may mean missing sales or paying more for a later shipment. Compare those consequences with the cash exposure rather than assuming a delay is free. The advisor explains the trade-off; the owner chooses the commitment.

Don’t count returned merchandise as an immediate replacement for the refund. Inspect it first and establish whether it can be sold again. Put a supported resale receipt into the forecast when expected, without assuming the next customer appears the day the first one gets paid back.

Keep a short record of the sales group, remaining refund estimate, and expected withdrawal dates. Review actual refunds each week, remove payments already made, and revise what remains. Release the cushion as evidence changes, not because the owner wants the balance to look larger.

Back at the opening conversation, the owner can still buy stock. The review has shown why the full $20,000 order leaves little room and what a smaller commitment would change. Money already collected can still have a return trip to make.

Clear Path To Cash Advisor helps advisors apply this kind of judgment in client conversations. The system supports the work of finding the missing cash assumption, comparing actions, and helping an owner choose what to commit.