Order value is the monetary amount of a single customer purchase, the total attached to one order. Which charges count toward that total, such as shipping, tax, discounts, and returns, is a reporting choice, so one order can carry different values in different systems. Averaging many orders produces a separate metric, covered in the average order value (AOV) entry, which also holds the benchmarks and the tactics for raising it.
This entry stays with the single order. The question it answers comes before any average, any dashboard, and any comparison: when someone says an order was worth a certain amount, which number are they using?
Order value describes one event. In e-commerce, a customer checks out, an order is created, and a value is attached to it. AOV describes a population of those events. It takes the value of many orders over a period and divides it by how many orders there were.
The difference sounds small, but it decides what each metric can answer. Order value answers questions about a specific transaction: what this customer paid, what this invoice says, how large this refund should be. AOV answers questions about a pattern, such as whether customers spend more per checkout this quarter than last, or whether one channel produces larger baskets than another.
The two are tied together in one direction. AOV can never be cleaner than the order values it is built from. If individual orders are recorded on mixed bases, some with shipping and some without, some adjusted for returns and some not, the average inherits that mix. Nothing in the division step fixes it. That is the practical reason to settle the definition at the order level first.
Sections that rest on reasoning rather than a published source are labeled as analysis.
Most orders are built from the same handful of parts. What changes from one business to the next is which parts are added, which are subtracted, and at what moment the number is frozen. The breakdown below is analysis, not a published standard. The research behind this entry did not surface an official, cross-industry definition of what a single order's value must include.
The starting point is the price of each item multiplied by its quantity. Then discounts come off. A coupon code, a percentage promotion, a bundle price, or a voucher can all lower what the customer actually pays.
Discounts are where the first fork appears. One system may record the order at list price and track the discount as a separate line. Another may record only the discounted amount. A third situation, typical of marketplaces, involves a discount funded by someone other than the seller, so the customer pays one amount while the seller is credited with another. All three can describe the same order accurately. They are not interchangeable.
Shipping charges, sales or value-added tax, and service or handling fees sit on top of the item total. The customer sees all of them in the checkout total. Whether the business counts them as part of the order's value is a separate decision, made by whoever designs the report.
The reasoning tends to run like this (analysis). Tax is usually collected on behalf of a government and passed on, so many reports leave it out. Shipping is less settled, because some businesses treat it as revenue and others treat it as a cost they simply pass through. Fees vary with the business model. The next major section sets a published company definition beside the rule used in this glossary's AOV entry, and the two do not draw the line in the same place.
An order's value can change after checkout. Items come back, orders are cancelled before they ship, and partial refunds are issued for damaged or missing goods. Each of these raises the same question. Does the recorded order value change, or does it stay at the checkout amount while the reversal is tracked somewhere else?
At least one public company answers this explicitly. Revolve Group builds its average order value from gross sales "prior to product returns." Put simply, an order is counted at what was sold, not at what was kept. That is a legitimate choice. For any business where returns are common, it will also produce a higher number than a net-of-returns approach, which is exactly why a reader needs to know which version a figure uses.
A component-by-component comparison of where a published company definition and this glossary's AOV rule draw the line, so a reader can see why figures built on each basis cannot be read against each other.
| Component | Revolve Group definition | Glossary AOV entry rule | Source / label |
|---|---|---|---|
| Value basis | Total gross sales from its sites in the period, so the value is gross, not net | Not framed as gross or net; specific charges are counted or excluded (rows below) | Revolve Group, Second Quarter 2026 Financial Results release, August 4, 2026, company definition, VERIFIED. URL: https://investors.revolve.com/news/news-details/2026/Revolve-Group-Announces-Second-Quarter-2026-Financial-Results/default.aspx AOV rule: glossary AOV entry |
| Product returns and refunds | Measured prior to product returns: an order counts at what was sold, not what was kept | Refunds are excluded | Revolve Group, Second Quarter 2026 Financial Results release, VERIFIED. URL: https://investors.revolve.com/news/news-details/2026/Revolve-Group-Announces-Second-Quarter-2026-Financial-Results/default.aspx AOV rule: glossary AOV entry |
| Shipping and handling | Not stated in the quoted definition | Counted | AOV rule: glossary AOV entry |
| Sales tax | Not stated in the quoted definition | Excluded | AOV rule: glossary AOV entry |
| Chargebacks | Not stated in the quoted definition | Excluded | AOV rule: glossary AOV entry |
| Denominator | Total orders placed in the period, not orders delivered or kept | Not stated in the Order Value entry | Revolve Group, Second Quarter 2026 Financial Results release, VERIFIED. URL: https://investors.revolve.com/news/news-details/2026/Revolve-Group-Announces-Second-Quarter-2026-Financial-Results/default.aspx |
| Reported output | AOV of $299 for Q2 2026, against $300 a year earlier (group-wide); shown only to pair a definition with its output, not as a benchmark | No figure; this is a counting rule, not a reported result | Revolve Group, Second Quarter 2026 Financial Results release, company-reported metric, VERIFIED. URL: https://investors.revolve.com/news/news-details/2026/Revolve-Group-Announces-Second-Quarter-2026-Financial-Results/default.aspx |
| Read-across | Internally consistent | Internally consistent | Analysis: the sharpest contrast is returns (measured before returns vs refunds removed), so a figure built one way cannot be read against a figure built the other way |
Put the components together and a single order can be described by several defensible totals. A gross figure takes the order as placed. A net figure subtracts discounts, returns, or both. Tax and shipping can sit inside or outside either version. None of these is wrong. Each is a different answer to the question of what the order was worth.
The risk starts when a number travels without its basis attached. Put a gross order value next to a net one and the difference looks like a performance gap. It is a definition gap.
Revolve Group, a publicly listed online retailer, publishes its definition alongside the metric. In its second quarter 2026 results release, the company defines average order value as "the sum of the total gross sales from our sites in a given period, prior to product returns, divided by the total orders placed in that period."
Three choices are visible in that one sentence. The value is gross, not net. It is measured before returns. And the denominator is orders placed, not orders delivered or kept. The same release reports AOV of $299 for the second quarter of 2026, against $300 a year earlier (company-reported metric, group-wide, VERIFIED against the release). The figure appears here only to show a definition and its output side by side. It is not a benchmark for any other business.
The AOV entry in this glossary follows a different rule. It counts shipping and handling, and it excludes sales tax, refunds, and chargebacks. Set beside the Revolve definition, the contrast is sharpest on returns: one approach measures before returns, the other removes refunds. Both are internally consistent. A figure built one way still cannot be read against a figure built the other way.
Public companies report AOV for their own investors, using their own definitions. The RealReal, a luxury resale company, reported AOV of $629 for the third quarter of 2026, up 8% from the same quarter of 2025 (company-reported metric, company-wide, VERIFIED against the release). Its definition basis is not quoted in this entry.
Placed beside Revolve's $299, the RealReal figure invites a quick conclusion: customers spend about twice as much per order there. Part of that may well be true, given how different the two product ranges are, but the numbers alone cannot carry the claim. Without both definitions in hand, there is no way to tell how much of the gap comes from what customers buy and how much comes from how each company counts. That is the test this entry argues for (analysis). Before comparing two order values, check the basis of each.
The same caution applies to any AOV figure published by a single company, an industry report, or a platform. A company's reported number describes that company's transactions under that company's rules. It is evidence about the company, not a market average.
This section is conceptual analysis. It describes how the value of one order commonly moves through a business, without making claims about any specific platform's fields, settings, or defaults.
Web and app analytics tend to capture an order at the moment of purchase. The value recorded is whatever the site sends at checkout, so the definition is effectively set by whoever implemented the tracking. If returns or cancellations happen later, the analytics record may never change unless someone sends a correction. The result usually sits closer to a gross, at-checkout figure, regardless of what the finance team uses.
Analytics teams often pair order value with conversion rate to estimate how much revenue a channel or page produces. That pairing is only as reliable as the order value inside it. A conversion rate measured carefully and multiplied by a loosely defined order value still produces a loosely defined result.
When a purchase value is passed to an ad platform, it usually becomes the conversion value for that sale, and conversion value is the numerator in return on ad spend (ROAS). This is where the basis question gets expensive. If the value sent includes tax and shipping, or ignores returns, ROAS will look stronger than the margin behind it.
The effect compounds when bidding is set to pursue value. A system steering spend toward high reported conversion value will favor orders that look large at checkout, whether or not they stay large after returns.
Finance teams care about what the business keeps. That pulls their version of order value toward net: after discounts the business funded, after returns and refunds, and usually without tax collected on someone else's behalf. The finance figure also tends to arrive later than the analytics or advertising figure, because returns and refunds take time to settle.
So one order can be worth three different amounts in three systems. That is normal. The trouble begins when those figures are compared or swapped without anyone noting which is which.
Business-to-business orders follow the same logic with extra layers. This section is brief analysis, because the research behind this entry found no source material on B2B order value specifically.
A B2B order often starts as a purchase order, a document the buyer issues listing agreed items, quantities, and prices. The value on that purchase order, the value invoiced, and the value finally paid can all differ, because of partial deliveries, negotiated adjustments, or contract terms. Suppliers may also set a minimum order value, a floor below which they will not process an order or will add a charge. In that setting, order value is an operational threshold as well as a reporting metric.
The basis question still applies. A team comparing order values across accounts or periods needs to know whether it is reading purchase order value, invoiced value, or collected value.
Six common reporting mistakes that come from the basis problem, what each one does to the numbers, and the check that keeps order value figures readable.
| Mistake | What goes wrong | Safeguard | Source / label |
|---|---|---|---|
| Mixing bases in one report | Gross figures from one channel sit beside net figures from another, creating differences that look like performance but are really definition | Settle the definition at the order level first, since the averaging step does not fix a mixed basis | Analysis, Order Value entry |
| Measuring against another company's published number | A listed company's AOV follows that company's rules and product mix, so the comparison has no meaning without the definition | Check the basis of each figure before comparing two order values | Analysis, Order Value entry |
| Ignoring when returns are counted | A value recorded at checkout and one recorded after the return window closes describe different things, even for the same order | State when the value was measured | Analysis, Order Value entry |
| Letting tax and shipping drift in and out | If treatment changes when a checkout or tracking setup is rebuilt, the trend line breaks and nobody notices | Keep a stated note of what the value includes | Analysis, Order Value entry |
| Sending inflated values to ad platforms | Values that include pass-through charges or ignore returns make advertising look more profitable than it is, and value-based bidding favors orders that only look large at checkout | Know which basis feeds conversion value before reading ROAS | Analysis, Order Value entry |
| Publishing a figure without its definition | The number travels without its basis attached, so readers cannot tell which version it is | Add a one-line note stating what the value includes and when it was measured | Analysis, Order Value entry |
The mistakes below follow from the basis problem described in this entry. They are analysis, not findings from a study.
No. Order value is the amount of a single order. Average order value divides the total value of many orders by the number of orders. Because AOV is built from individual order values, it carries whatever definition those values use.
It depends on the reporting system, and practices differ. The AOV entry in this glossary counts shipping and excludes sales tax, while other systems draw the line elsewhere. Any order value figure should say whether shipping and tax are inside it.
Both versions are in use. Revolve Group, for example, measures gross sales prior to product returns for its AOV, while a net-of-returns figure reflects what the business kept. Neither is wrong, but every figure should state which one it is.
Only when both definitions are known and match. Company-reported figures, such as Revolve's $299 for the second quarter of 2026 or The RealReal's $629 for the third quarter of 2026, reflect each company's own counting rules and product range. The gap between them is not a measure of performance.
A minimum order value is a threshold a seller sets, below which an order is not accepted or carries an extra charge. It turns order value into an operational rule rather than only a metric, and it appears most often in wholesale and B2B arrangements.
The two usually record different things at different times. Analytics tends to capture the value sent at checkout, while finance tends to work from what was kept once discounts, returns, and refunds settled. Both can be accurate, but they should not be swapped for each other.
1. Revolve Group, Inc. "Revolve Group Announces Second Quarter 2026 Financial Results." Investor relations release, August 4, 2026. Used for: company-reported AOV ($299 vs $300) and the company's AOV definition. Confidence: VERIFIED. URL: https://investors.revolve.com/news/news-details/2026/Revolve-Group-Announces-Second-Quarter-2026-Financial-Results/default.aspx
2. The RealReal, Inc. "The RealReal Announces Third Quarter 2026 Results." Investor relations release, October 29, 2026. Used for: company-reported AOV ($629, up 8% year over year). Confidence: VERIFIED. URL: https://investor.therealreal.com/news/news-details/2026/THE-REALREAL-ANNOUNCES-THIRD-QUARTER-2026-RESULTS/default.aspx
We use cookies to ensure the website runs optimally and to help us understand how you use our services. You can choose which categories to allow. Read our Privacy Policy.
Required for basic website functionality. Cannot be disabled.
Help us understand how visitors interact with the website. Data used anonymously.
Used to display relevant ads and measure campaign effectiveness.
Enables live chat, social media integrations, and language preferences.