What is average order value?

AOV answers how much revenue an order represents on average, rather than how much each customer spends over their lifetime. One customer may place several orders. Define the period, currency, included orders and revenue treatment before calculating or comparing the average.

A larger order is not automatically more profitable

AOV can help you understand the order mix or examine a basket change, but it does not subtract every business cost. Discounts, product costs and fulfillment can change profitability even when the average order grows. Read it alongside completed orders, conversion rate and the relevant margin measure instead of using AOV alone as an acquisition-spend ceiling.

Specify the numerator and denominator

The general formula is AOV = included order revenue / included orders. Shopify’s sales report specifically uses gross sales minus discounts, divided by orders, excluding post-order adjustments. Its definition reflects product revenue when the order was placed, excluding taxes and shipping. Do not substitute Shopify’s total sales metric, which includes additional components, without relabeling your calculation.

Shopify: Sales report definitions.

Different report definitions need different labels

Revenue averages and what they measure
AverageRevenue basis or denominatorInterpretation
Shopify AOVProduct gross sales minus discounts, excluding post-order adjustments; divided by orders.Order value when placed, not all collected amounts.
Custom refund-adjusted order averageYour documented revenue after the chosen adjustments, divided by the defined order set.Label the refund timing and denominator; not automatically the platform’s AOV.
GA4 average purchase revenueAverage purchase revenue in the transaction group of events.Depends on recorded analytics events and their values.
Revenue per userA user denominator rather than an order denominator.Answers a different question when customers place multiple orders.

Shopify: Sales report definitions. GA4: API metric definitions.

Worked hypothetical AOV

Suppose a hypothetical store has USD 1,200 in eligible product gross sales, USD 200 in discounts and 20 included orders. With a Shopify-style order-placement basis, AOV is (1,200 − 200) / 20 = USD 50 per order. Taxes and shipping are excluded, and later adjustments are outside this example. The result is a labeled calculation, not a customer outcome or a benchmark.

gross_sales_usd = 1200
discounts_usd = 200
orders = 20
aov_usd_per_order = (gross_sales_usd - discounts_usd) / orders
# Result: 50 USD per order

Compare consistent populations

Keep the revenue basis, currency and order inclusion rules fixed across a comparison. If the average changes, inspect product and channel mix before attributing it to a new offer. For combined segments, divide their combined revenue by their combined orders; averaging segment AOVs without weighting each segment by orders can give the wrong overall value.

Avoid mismatched counts and missing events

Revenue and orders must describe the same population and period. A refund processed today can relate to an earlier order, so an adjustment-based report needs an explicit timing rule. Analytics purchase averages also depend on recorded event data: compare their definitions and collection coverage with the order ledger before assuming a discrepancy is real customer behavior.

Shopify: Sales report definitions. GA4: API metric definitions.

Questions about average order value (AOV)

What is a good AOV?

This definition establishes no universal benchmark. Compare consistent periods and comparable order populations, then assess the result alongside conversion and margin. A product-mix change can alter AOV without an improvement in profitability.

Does AOV include refunds, discounts, tax and shipping?

The answer depends on the report. Shopify’s documented AOV subtracts discounts from product gross sales and excludes post-order adjustments, taxes and shipping. A custom refund-adjusted average needs its own explicit label and rules.

How is AOV different from ARPU?

AOV divides a defined revenue total by orders. Average revenue per user uses a defined user count. If customers place multiple orders, the two averages answer different questions.

What if there are no orders?

With a zero order count, the division is undefined. Show no calculable AOV for that period rather than inventing a zero average.

Average Revenue Per User (ARPU) · Conversion Rate · Customer Lifetime Value · Cross-Sell · Upsell

Sources