What is AOV?

By the HonestTag team · Published July 12, 2026

AOV gets treated as a vanity metric: a number that drifts up and to the right in a monthly deck while the decisions happen elsewhere. That is a mistake. AOV is structural. It bounds your acquisition math, and when it is computed sloppily, everything downstream of it (acquisition ceilings, payback windows, budget decisions) inherits the error.

The formula, and the split that matters

AOV = net revenue / order count. Net means what it says: exclude tax, subtract discounts.

One AOV is not enough, though. You need two:

• Blended AOV averages all orders, first purchases and repeat purchases together.
• First-order AOV averages only new customers' first purchases.

They differ whenever repeat buyers behave differently from first-time buyers, which is almost always. Repeat buyers know the catalog, trust the brand, and buy without hesitation; first-time buyers hedge with smaller baskets. Only first-order AOV belongs anywhere near acquisition decisions.

Why it matters: the contribution ceiling

A merchant may compute first-order AOV x gross margin percent as the contribution ceiling on a first order: the hard bound on what you can pay to acquire a customer before repeat purchases enter the story. A store with a $90 first-order AOV and 55% gross margin earns about $49 of contribution on a first order. Pay more than that in NCAC and every new customer starts underwater, repayable only by orders that have not happened yet.

HonestTag does not compute first-order AOV or use it for this suggestion. When a store has set no explicit NCAC target, the suggested NCAC ceiling is derived as blended AOV x gross margin percent x an LTV multiple, and every verdict that uses the derived ceiling says so. A derived number presented as if you chose it would be dishonest; a derived number labeled as derived is just a starting point you can override.

AOV moves and their traps

AOV responds to levers, and the levers are well known: free-shipping thresholds, bundles, and post-purchase offers all raise it. The traps are quieter.

Heavier discounting raises order count while cutting net AOV. Run that for a quarter and MER can look flat while the machine underneath changed completely: more orders, thinner orders, same top-line ratio. The average hid a structural shift.

The second trap is reading AOV alone. Watch it next to NCAC. Rising acquisition cost alongside rising first-order AOV can be healthy: you are buying bigger customers. Rising cost with flat AOV is not; you are just paying more for the same customer.

Computing it honestly

Three rules, each with a reason:

• Exclude tax. You do not keep it. It is not order value; it is money you collect for someone else.
• Subtract discounts. You never received that revenue. A $100 order with a 20% code is an $80 order.
• Net out refunds. Money returned is not order value, no matter how good the week looked before the return.

Each shortcut inflates AOV, and an inflated AOV inflates everything downstream of it: the contribution ceiling, the suggested NCAC bound, the payback math, and every budget decision resting on them. Small definitional cheats compound into large strategic ones.

The dishonest versions

• AOV on gross revenue, tax included. The average rises; nothing about the business improved.
• Refund-blind AOV. Returns vanish from the math and order value stays flattered.
• Blended AOV in acquisition math. The most damaging one. Repeat buyers pull the average up, the contribution ceiling looks higher than it is, and NCAC looks affordable when it is not.

What each version drives

1. First-order AOV is the acquisition-facing number. It feeds acquisition math and moves for different reasons than the blended number, so the two are not interchangeable.
2. An NCAC ceiling depends on current margin and AOV. A ceiling computed from last year's margin is a fiction with a formula attached. The ceiling shifts whenever margin or AOV does.
3. An AOV number read alone can mislead. A free-shipping threshold that lifts AOV but costs margin and conversion can lose money while the average climbs, so the average and the contribution behind it are different things.

How HonestTag computes it

HonestTag works from your store's own Shopify order data and computes blended AOV as net revenue across all new and returning orders divided by all orders, net of discounts and refunds and excluding tax. HonestTag does not compute first-order AOV; merchants may calculate that separate concept from first purchases. Google Ads and Microsoft Advertising receive the available retraction or restatement for a refund, and Klaviyo and Google Analytics 4 receive a refund event. HonestTag does not send Meta, TikTok, Pinterest, Snapchat, Reddit or OpenAI Ads a refund or retraction event; the refund is recorded in order proof. New customers attribute to the first ad click, each attribution stores a proof record (click id, timestamps, match method), and every threshold and formula is published in numbers on the in-app methodology page. Definitions for every term here live in the glossary.

Frequently asked questions