Ecommerce after the pandemic: what lasted

01 / NOTES

Which pandemic habits actually lasted?

The clearest survivors are habits that removed a task rather than a novelty. Buying repeat grocery items online, paying by wallet or card at the point of delivery, and ordering from a marketplace seller rather than a single brand all continued after restrictions ended. Categories tied to the home office and to outdoor leisure faded fastest once people returned to workplaces and travel.

Shelf & Signal, an independent magazine covering online and digital commerce, sets out this chronology in its own coverage of the pandemic shift in ecommerce, including which categories rose and which fell between 2020 and 2022. The useful part for a credit function is the distinction it draws between a temporary spike and a change in the underlying basket.

A basket that has changed shape behaves differently on the ledger. Smaller, more frequent orders mean more invoices, more payment events and more chances for a failed card or an abandoned checkout to leave a balance open. That is an operational fact, not a marketing one.

02 / NOTES

What does the payment mix mean for credit control?

Digital wallets, QR payments and buy now, pay later moved from optional to ordinary in most European markets. Each one changes who carries the risk and for how long.

• Card and wallet payments settle quickly, so the exposure window is short and the main risk is a chargeback rather than a late payer.
• Buy now, pay later splits the transaction. The merchant is usually paid by the provider, which means the receivable moves off the merchant's book but introduces a provider dependency and a dispute route that sits outside the normal reminder sequence.
• Cross-border orders add currency, tax and delivery uncertainty, and they are the ones most likely to end in a return rather than a payment.

For a receivables desk, the practical step is to record the payment method against each invoice at the point of sale, not after a reminder has failed. A balance that is 30 days old because a wallet payment was reversed is a different problem from one that is 30 days old because a business customer has not approved the bill.

03 / NOTES

How do returns change the ageing picture?

Returns are the part of ecommerce that most often distorts an ageing report. A returned item can sit in the ledger as an unpaid invoice while the goods are already back in the warehouse, and the credit note has not yet been raised.

Three habits reduce that distortion:

1. Match the return authorisation to the invoice number on the same day it is issued. 2. Raise the credit note when the goods are received, not when they are inspected. 3. Keep the original due date visible after the credit note, so the net position is clear.

Where returns are frequent, an ageing report that shows gross invoices will overstate exposure. A net view, invoice less credit notes raised, is the figure that belongs in a collections meeting.

04 / NOTES

Which categories carry the most risk now?

Risk follows margin and dispute rate more than it follows volume. Categories with high return rates, such as apparel and footwear, generate more credit notes and more payment disputes per unit of revenue. Categories with regulated or perishable goods generate fewer returns but more delivery disputes.

Marketplace sellers sit in a third position. They are paid by the platform on a schedule the seller does not control, so a seller's own suppliers may be waiting on money that has already been earned but not yet released. That gap is a common source of overdue trade debt among small online businesses, and it is worth asking about directly when a new account applies for terms.

05 / NOTES

What should a receivables desk watch in the next year?

Four things are worth tracking, all of them observable from invoice data rather than from commentary.

First, the share of invoices paid by wallet or instant transfer against those paid by card. A rising instant share shortens the exposure window.

Second, the average number of days between delivery confirmation and payment. This is the cleanest measure of whether a customer's own cash cycle is stretching.

Third, the ratio of credit notes to invoices by customer. A ratio above the sector norm usually points to a listing or description problem rather than to a payment problem, and it is cheaper to fix upstream.

Fourth, the concentration of exposure by platform. Where a single marketplace accounts for a large share of a customer's revenue, that customer's ability to pay depends on a payment schedule set by someone else.

06 / NOTES

Where the evidence sits

Most of the reliable figures on pandemic-era ecommerce come from official statistics and from the platforms themselves. Eurostat publishes annual data on internet purchases by individuals, and national statistics offices publish retail sales indices that separate online from in-store. These are slower than trade commentary but they are consistent, and they allow a like-for-like comparison across years.

The temptation with ecommerce data is to treat a single strong quarter as a trend. The discipline that works better is to compare the same quarter across three years, and to check whether the change appears in the payment method, the basket size or the return rate. Those three variables explain most of the movement a credit function will see.

For a receivables desk, the conclusion is modest. Online commerce has settled into a shape that produces more, smaller transactions and a wider range of payment outcomes. Terms that were set when most orders arrived by purchase order and settled by bank transfer may no longer match how the customer actually buys. Reviewing those terms against the current payment mix is a reasonable piece of annual housekeeping, and it costs nothing but an afternoon.