Zero-inventory commerce means a storefront lists a distributor’s catalogue through an API, buys each unit at the moment the customer pays, and delivers the code seconds later. Nothing sits in stock, so a two-person team can offer five thousand products across gift cards, game credit and top-ups without buying a single code in advance.
The model is why so many gift card storefronts appeared in the last few years that look far bigger than the companies behind them. The catalogue is rented, not owned.
How on-demand fulfilment works
The storefront syncs the distributor’s catalogue (products, prices, regions, availability) and presents whatever slice fits its audience. At checkout, the customer pays the storefront; the storefront’s backend places an order against its supplier balance; the code comes back through the API and lands in the customer’s inbox. Elapsed time: seconds. The storefront’s take is the spread between the retail price it charged and the trade price it paid, earned per unit, at the moment of sale.
Structurally this is retail built on an aggregator or distributor: one integration and one commercial relationship stand in for dozens of brand contracts a small storefront could never sign.
What the model removes
Stock risk. No pre-purchased batch can expire, get stolen, or turn out to be the wrong region for your audience. You buy exactly what customers have already paid for.
Dead inventory. In a five-thousand-product catalogue, most SKUs sell rarely. Holding stock forces you to guess which ones; on-demand fulfilment makes the slow movers free to list.
Up-front catalogue investment. Testing a new vertical (say, adding game top-ups next to gift cards) costs a catalogue filter change, not a purchasing decision. Failed experiments cost nothing but attention.
What it keeps
Fraud exposure. You still accept card payments for instantly-delivered bearer goods, and a chargeback on a redeemed code is your loss, not the distributor’s. Zero inventory does not mean zero risk; it moves the whole risk budget to payments.
Support. The customer bought from you. Failed deliveries, redemption questions and already-redeemed claims land on your desk regardless of who held the stock.
Price volatility. Live trade prices move with issuer pricing and currency, and a checkout quoting yesterday’s price can sell below what you will pay a second later. Sync discipline, covered in catalogue sync and price drift, is the tax the model charges.
The float question
Zero inventory is not zero capital. Most distributors work on prepaid balance, so money still sits with the supplier before sales happen; the difference is scale and shape. A stocked reseller ties up capital in specific SKUs bought on a guess; an on-demand storefront holds a general-purpose float sized to a week of orders, spendable on whatever sells. Sizing and monitoring that float is its own discipline, covered in managing a prepaid balance.
When holding stock still wins
On-demand loses in two narrow cases. First, guaranteed availability: if a promotion will sell two thousand units of one SKU on one evening, live availability is a promise the API does not make, and a pre-purchased batch is insurance. Second, locked pricing: a campaign printed at a fixed price for weeks cannot float on a live trade price, so buying the batch fixes the economics in advance. Both are planned exceptions inside an on-demand operation, not arguments against it.
Who the model suits
Startups testing whether an audience buys at all, before any capital commitment. Apps and communities adding a commerce tab beside their real product. Niche storefronts whose edge is curation and audience trust rather than purchasing power; the creator-flavoured version of the same play is covered in selling game top-ups to your audience. The common thread: the value is the audience and the storefront experience, and fulfilment is rented from someone whose whole business is fulfilment.
Frequently asked questions
How can a store sell gift cards without inventory?
By integrating a distributor’s catalogue API. The store lists the products, and when a customer pays, the backend buys that single unit from the distributor and delivers the code within seconds. The store never holds stock; it holds a prepaid balance with the distributor and earns the spread between retail and trade price on each sale.
What is on-demand code fulfilment?
On-demand fulfilment means each code is purchased from the supplier at the moment of the customer’s payment rather than from a pre-bought batch. Delivery still happens in seconds because the supplier’s API issues codes in real time. The seller carries no stock risk and pays only for units already sold, at the trade price current at that moment.
When is holding stock better than on-demand?
In two cases: guaranteed availability and locked pricing. A promotion that must not sell out justifies pre-buying the batch, because live availability can change mid-campaign. A price printed in advance justifies locking the buy side too, since live trade prices float. Outside those, pre-purchased stock mostly adds capital lock-up and expiry risk.
How much capital does a zero-inventory storefront need?
Less than a stocked reseller, but not zero. Most distributors work on prepaid balance, so a float sized to roughly a week of orders has to sit with the supplier before anything sells. Unlike stock bought on a guess, that float is general purpose: it funds whatever customers actually order. The remaining capital needs are ordinary ones, payments reserves and working cash, not catalogue investment.
What are the main risks of the zero-inventory model?
Payment fraud and price drift, not stock. Codes are bearer goods delivered in seconds, so a chargeback on a redeemed code is the storefront’s loss, and the whole risk budget shifts to payment screening. Separately, live trade prices move, and a checkout quoting a stale price can sell below the price you pay moments later, which makes catalogue sync discipline part of the job rather than a nice-to-have.