What a chargeback actually is.
A chargeback is a forced reversal of a card transaction initiated by the cardholder’s bank, not by you. The customer disputed the charge with their bank, the bank pulled the funds back, and Shopify (or your gateway) deducted the disputed amount plus a chargeback fee from your next payout. You can fight the dispute, but the funds are gone until resolution.
There are three lifecycle stages: dispute opened (funds held), evidence submitted (you submit proof of delivery, customer authorization, etc.), and resolution (you win, funds return; you lose, funds stay gone). Each stage has its own accounting consequence. Booking it as a single bad debt at open is wrong because the outcome is unknown.
The journal at dispute open.
Disputed amount moves from cash (Shopify clearing) to a Disputed funds receivable account. The chargeback fee is non-refundable, so it expenses immediately. On a $124 disputed transaction with a $15 chargeback fee, the journal is debit Disputed funds receivable $124, debit Chargeback fees expense $15, credit Shopify clearing $139.
The original sale and COGS stay on the books. They will only reverse if you lose. Booking the chargeback as a sales reversal at open is too eager and creates a journal correction when you win. The disputed funds receivable is the cleanest holding account because it parallels the actual cash flow: held until resolution.
The four-stage lifecycle.
A chargeback is not a single event — it is a state machine. Treating it as one event (bad debt write-off at open) creates a sea of journal corrections later. Treating it as four states means each transition has one clean entry, and your books always match the actual cash position.
Resolution: you win.
When you win the dispute, the funds return on the next payout. Reverse the receivable: debit Shopify clearing $124, credit Disputed funds receivable $124. The chargeback fee stays expensed; gateway providers do not refund the fee even on a win.
Net effect on a winning fight: $15 expense for the fee, no impact on revenue or COGS. The original sale stays clean. Time spent fighting is operational overhead, not directly bookable. If you want to track win rate impact on contribution margin, run a separate non-GL metric.
Resolution: you lose.
When you lose, the funds stay gone. The receivable becomes a write-off. Journal: debit Bad debt expense $124, credit Disputed funds receivable $124. The original sale and COGS reverse only if the merchandise was returned (rare in chargeback scenarios) or if the dispute was a friendly fraud where the customer kept the goods (most common, no return).
The most common chargeback scenario in cross-border DTC is friendly fraud: a customer in one country places an order, receives the goods, then disputes the charge with their bank. You lose the dispute because cardholder presence at delivery is hard to prove. The cost: revenue gone, goods gone, fee paid. Books should reflect all three.
Documentation that wins disputes.
You win disputes with documentation. Proof of authorization (3DS verification), proof of delivery (signed POD or carrier confirmation), proof of customer engagement (account login, IP match), and proof of policy disclosure (terms accepted at checkout) are the four pillars. Most disputes are won or lost in the first 7 days based on the strength of this evidence.
For merchants shipping internationally, the proof of delivery is critical. DHL, FedEx, UPS, Royal Mail, Australia Post all provide signed POD or tracking with delivery confirmation. Without it, even valid transactions lose. In Nonari you can attach the delivery proof to the order’s invoice when it ships, so when the dispute opens the evidence is already on file.
- Proof of authorization: 3DS verification or AVS match
- Proof of delivery: signed POD or carrier-confirmed delivery
- Proof of customer engagement: account activity, IP, browser fingerprint
- Proof of policy: timestamped acceptance of terms and refund policy
Chargeback ratios and consequences.
Card networks (Visa, Mastercard) monitor chargeback ratios. Above 1% disputes-to-transactions, you risk being placed in a monitoring program. Above 1.5%, you face fines and potential gateway termination. Merchants accepting international cards need to track this ratio per gateway and per card type.
Track chargebacks-to-orders, not just dollar ratios, because card networks use both. Track by card type (Visa vs Mastercard vs Amex have different rules). Track by storefront, because a high-fraud international storefront can pull down your overall ratio without you noticing. Your gateway’s dispute report has the card-type split; in Nonari each store sits on its own branch, so the per-storefront view is a branch filter on your chargeback accounts.
Reserves and provisions.
If your chargeback rate is consistently above 1%, consider posting a chargeback reserve. Estimate expected chargebacks as a percentage of revenue, expense it monthly: debit Chargeback expense, credit Chargeback reserve liability. When actual chargebacks land, they hit the reserve instead of the P&L: debit Chargeback reserve, credit Disputed funds receivable.
This smooths chargeback expenses across periods and forces conscious provisioning. If actual chargebacks consistently undershoot the reserve, dial it down. If they overshoot, dial up. Most Shopify merchants do not need a reserve until cross-border GMV is meaningful, but the moment you cross $50k/month international, it is worth setting up.
Where Nonari fits.
What reaches Nonari is the money: Shopify Payments deducts a chargeback from a payout, and once the connection’s accounts are mapped that deduction posts to the adjustments account you chose and shows on the Shopify Payouts report. From there, the lifecycle above is a pair of journal entries you post — reclassify the chargeback into Disputed funds receivable when the dispute opens, then clear it to the payout if you win or to bad debt if you lose — each one recorded in the audit log.
Because files can be attached to the order’s invoice, the delivery proof and the 3DS record can sit on the order before any dispute exists, so fighting one is an evidence-attach action, not an evidence hunt. The accounting and the fight share the same record, so you spend the dispute window arguing facts instead of assembling them.