Selling food online is a logistics promise with a storefront attached

For anything chilled, frozen or freshly made, the delivery slot is a condition of the sale, not a shipping option. The date and window get chosen before checkout completes. Capacity per slot has to be enforced, so you cannot sell more deliveries than you can make. And cut-off times must be visible, never discovered: 6pm the day before is a fact somebody can plan around, while order early is not. Stores that treat this as an afterthought generate cancellations and refunds that no marketing spend can offset.

• Slot selection before payment, not after
• Capacity limits per slot, enforced automatically
• Cut-off times stated clearly on the product page
• ZIP code or zone checks before the cart is built

Ingredient lists, allergen declarations, nutritional information and storage instructions have to be accurate, present and easy to find, because customers make health decisions from them. Structure this data as fields on the product, never as free text buried in a description, so it displays consistently, can be filtered on, and can be updated without editing prose. In the US the FDA requires nine major allergens to be declared (sesame joined the list in 2023), and an allergen filter is only trustworthy if every product carries the field. Recipe or batch changes need a process that updates the product data at the same time as the label, not weeks later.

Available means something different when a product has a use-by date. Stock levels need to reflect what can actually be dispatched within its shelf life, and products with short lives may need to disappear from sale before they physically run out. Batch and date tracking are worth the effort in this category. The alternative is shipping something that arrives too close to expiry, and that produces refunds and reviews that follow you for months. Inventory sync covers the general mechanics; the perishable case simply raises the stakes.

Food and drink is a repeat category by construction, because people finish what they bought, and most of that revenue depends on how easy it is to buy the same cart again. Order history that becomes a one-tap reorder, subscriptions for staples, and a saved-favorites list do more for annual revenue than most acquisition work. WooCommerce already ships Order again; the job is putting it where a returning customer lands. The subscription model fits this category unusually well, provided the self-service side (skip, pause, change date) works without contacting support.

Many food businesses sell within a delivery radius, ship nationally only for shelf-stable lines, or face restrictions on what can cross state lines. Alcohol is the hard case. Its licensing and age-verification requirements differ across all 50 US states, and a handful prohibit direct-to-consumer shipment outright. The storefront needs to know which products can go where before the customer builds a cart, and say so plainly. Otherwise the failure happens at checkout. This is general information and not legal advice, and the specific rules for your products and destinations are a question for a specialist.

Products, pricing, stock, orders, customers, tax and payments remain in WooCommerce, along with your slot and route logic if it already runs there. The storefront reads it over the Store API (WooCommerce's built-in cart and checkout interface) like any other cart data and reimplements nothing. What changes is the storefront layer: how quickly the catalog loads on a phone in a kitchen, how the slot chooser presents capacity, how ingredient data is displayed, and how few taps a repeat order takes. Those are exactly the areas where a heavy theme costs a food store money.

If you sell shelf-stable products nationally with no slot logic, standard ecommerce advice applies and none of this complexity is needed. The work here earns its cost when delivery windows or perishability are central to your operation, when allergen data is being maintained by hand across two systems that disagree, or when reorder friction is visibly capping the repeat revenue the category should be giving you.