How to increase average order value: what 42,511 orders in 12 Israeli stores show

What average order value is, and what we measured

Average order value (AOV) is the average amount a customer pays per order: total revenue in a period, divided by the number of orders in that period.

It matters because the customer is already paid for. The click that brought them cost the same whether they bought one item or three, a price we measured for Google Ads and for Facebook and Instagram ads. A bigger basket is growth without buying another click: same media spend, same orders, more revenue, and a higher real ROAS.

Ask how to increase average order value and you usually get a list of tactics. We looked at the basket first: 12 Israeli ecommerce stores, 1 January to 28 September 2026, 42,511 orders. Order value is what the customer paid after discounts. Amounts are in Israeli shekels (₪) and include VAT and shipping.

Average order value in ecommerce: the range

The average order value at the median store was ₪350. The smallest basket was ₪160, the largest ₪840.

The gap reflects what each store sells and at what price, not how well it is run. Compare your basket with your own, before a change and after it.

The basket does not grow by itself

Across nine months, holidays and sale events included, the median index stayed between 95 and 103: a swing of a few percent, with no trend.

A single store can move more, but the conclusion holds: the basket does not grow by waiting. It takes a different offer. Five levers follow, each with what the data shows and what it cannot.

1. Know your median order, not only your average

In all 12 stores the median order was lower than the average order, by 18% at the median store.

The median order is the one in the middle: half the orders are smaller, half larger. A few very large orders pull the average up, so your typical customer spends less than your average order value suggests.

The typical customer spends less than the average order. Set the threshold by that customer, not by the average.

What the data shows: a threshold set from the average is aimed at a customer who spends more than most of your customers do. What it cannot show: where your own median sits. Your order export will tell you.

2. The second item is where the money is

At the median store, 46% of orders contained a single item: one product, one unit. Between stores that share ran from 3% to 85%.

When a second item goes in, the order does not double. An order with two units was worth 34% more than an order with one, at the median across stores. Three units or more: 1.95 times a single-unit order.

The likely reason: the second item is usually cheaper than the first, or it carries the discount. That is an explanation, not a measurement: we measured order value, not which item went in first.

What the data shows: a second item added between 16% and 70% in every store we measured. What it cannot show: what happens if you offer a second item to everyone. People who buy two arrived with a different intention.

3. A discount that asks for something in return

At the median store 51% of orders carried a discount, automatic or by code. Between stores: 8% to 88%.

In 8 of the 12 stores, orders with a discount were larger than orders without one: by 10% at the median, measured after the discount came off. Not everywhere. At one end discounted orders were 34% larger, at the other 41% smaller.

What you can act on is the structure. A flat discount across the store also goes to people who would have bought anyway. A tiered discount, where a larger basket earns a higher percentage, and a second-item offer both ask for something in return.

Then check that the margin survives. A discount on the second item gives up margin only on the added item. A discount on the whole basket gives it up on the first item too, which the customer was buying anyway.

4. A free shipping threshold just above the median order

In all 10 stores with enough orders of both kinds, orders with free shipping were larger than orders that paid for it: by 85% at the median, even with the fee counted inside the orders that paid.

That is not a lift caused by free shipping. It usually starts above a basket threshold, so large orders are the ones that get it. The threshold sorts orders. Whether it also moves them depends on where it sits relative to the typical order.

It can miss both ways. Too low, and most orders clear it anyway: at the median store 66% of orders paid nothing for shipping, through a threshold, a promotion or pickup. Too high, and one more item will not get the typical customer there.

Take a store that sits on our medians: an average order value of ₪350 and a median order 18% below it, about ₪290.

Set from the average, the threshold is ₪420. The typical customer has to add ₪130: more than a second item adds at the median store. It will mostly reward people who were buying a lot anyway.

Set from the median, it is ₪330. The distance is ₪40: less than a second item added in every store we measured. One more item, even a cheap one, clears it.

What the data cannot show: that moving the threshold will grow your basket. Treat it as a test, not a promise: compare the median order and the share of single-item orders before and after, and weigh the shipping you absorb against the margin on the added item.

5. Bundles and sets

A bundle turns several units into one decision: the customer chooses one product that already contains the second and third item.

What the data shows: an order with three units or more was worth 1.95 times a single-unit order. More, but less than double. What it cannot show: anything about bundles alone. We did not separate them from ordinary multi-item orders.

Price a bundle like a second item: the discount on the set has to be smaller than the margin the extra units bring.