Average order is a ratio, not a result
Average order value is revenue divided by the number of receipts. The fraction has two variables, and either can move. If receipts fall faster than revenue, average order rises even though the business is worse off.
That is why average order is one of the four multipliers of revenue, not a goal in itself. How it connects with traffic, conversion and frequency is covered in the article on the revenue formula.
Three traps
Small buyers left. An illustrative example: a shop sold 1,000 receipts a month at UAH 400, revenue UAH 400,000. Then a competitor opened nearby and 300 buyers with small purchases moved there. 700 receipts remained, with a larger share of big ones, and the average became UAH 460. A 15% rise. Revenue meanwhile: UAH 322,000, down 19.5%.
Prices rose, not the basket. Raise prices by 10% and average order rises by about the same without any change in buyer behaviour. That is not a sales achievement; it is inflation. The check is simple: look at items per receipt. If the number is the same, the basket hasn't changed.
The rise came from discounts. "Three for the price of two" lifts average order, but margin per receipt can fall — especially if the buyer would have taken the third item anyway. Watch margin per receipt, not just the amount.
Raising average order without cutting margin
One specific suggestion at the till. Not "anything else?", which gets "no", but a specific add-on: "with these boots — a leather care spray, UAH 180". One phrase the salesperson says to every buyer in that category. It costs nothing except training, and its effect is easy to measure.
A threshold. Free delivery or a gift above a certain amount. Set it slightly above the current average order, so the buyer is one inexpensive item short of it. A threshold set too high gets ignored.
A bundle with something to add. A bundle makes sense when it adds an item the buyer wouldn't have taken separately, and the margin on it covers the discount.
Placement. Add-ons next to the main product, not in a separate section. For a site, a "bought together" block on the product page and in the basket.
How to check what worked
Compare the period before and after a change on four numbers, not one:
- Number of receipts — did it hold?
- Average order — did it rise?
- Items per receipt — more? If not, prices did the lifting.
- Margin per receipt — did it hold?
If the change can be introduced partially — one shift uses the new phrase at the till and another doesn't, or one shop in a chain has the threshold and another doesn't — compare them against each other over the same period. That separates the effect of the change from the season.
Our position
Don't give sales staff a target on average order alone. Such a target rewards turning away small purchases and pushing expensive items on people who don't need them, which hits the number of receipts and returns. A target that pairs average order with the number of receipts doesn't create these distortions.
The position does not hold for a business where the order is almost fixed: a single-price service, a membership, a booked treatment. There average order barely moves, and the work is on the number of customers and their frequency.