# Setting a Free Shipping Threshold From Your AOV Gap, Not a Round Number

Source: https://edgecommercejournal.com/blog/2026-11-17-free-shipping-threshold-aov-gap.html

November 17, 2026 - Edge Commerce Journal


November 17, 2026

*A merchant we work with set free shipping at $75 because two competitors used $75. Their average order value was $61. The threshold was not an incentive. It was a wall almost nobody climbed, and the shipping charge at $61 did the actual damage on every order.*

## The gap is the whole calculation

A threshold works when it sits a reachable distance above what shoppers already spend. The working range most operators land on is 15 to 30 percent above current average order value: close enough that adding one item closes the gap, far enough that the lift covers the shipping cost on orders that would have shipped paid. At $61 AOV, that puts the threshold near $70 to $79, which is why the copied $75 was nearly right by accident and still wrong in practice, because nobody checked the basket distribution underneath the average.

Averages hide the shape. If most orders cluster at $45 and a few wholesale-sized orders drag the average to $61, a $75 threshold asks the typical shopper to add $30, not $14. Pull the median basket and the share of orders sitting within $10 and $20 of the candidate threshold. That share is your realistic climb rate, and it is the number the decision should hang on.

## Run the margin math per order, not per month

Free shipping is a discount equal to your real shipping cost, and it should be priced like one. Take the orders that would newly qualify, multiply by your average outbound shipping cost after carrier rates and packaging, and set that against the gross margin on the extra items shoppers add to qualify. If your margin on an added item is thinner than the shipping you absorb, the threshold is buying revenue at a loss, which some stores accept for a season and none should accept by accident.

Watch the second effect too. A threshold set too low trains full-price shoppers to expect free shipping on baskets that were profitable with paid shipping. The test that catches it is contribution per order, split by new versus returning, over at least a few hundred orders. Revenue per session will flatter almost any threshold change. Contribution per order will not.

## Change one variable, then leave it alone for a full cycle

Threshold tests fail in a familiar way: the store changes the threshold, the banner copy, and the product page shipping message in the same week, then argues about which one worked. Change the number only. Keep the progress bar, the cart message, and the announcement placement identical, and run it long enough to cover your normal buying cycle, including a payday weekend if your category has one.

And write down what you expected before you start: the climb rate, the shipping cost absorbed, the contribution per order you would accept. A threshold decision made against numbers written down in advance survives the meeting where someone proposes $100 because it sounds premium.

Related reading on this site: [Your Product Page LCP Is One Image. Find It Before You Optimize Anything](2026-11-16-product-page-lcp-element-audit.md), [Holiday Returns Are a Margin Line, Not a Service Gesture](2026-11-16-holiday-returns-window-margin-math.md), [Your Free Shipping Threshold Is Probably Set From a Rule of Thumb. Run the Break-Even Instead](2026-11-15-free-shipping-threshold-break-even-math.md).

Pull your median basket and the share of orders within $15 of your current threshold. If that share is under a tenth of orders, your threshold is decoration, and the shipping fee is doing the talking.

## Frequently asked questions

### How far above average order value should a free shipping threshold sit?

Most operators land 15 to 30 percent above current average order value. Check the basket distribution as well as the average: if most orders cluster well below the average, set the threshold against the median basket instead.

### How do I know if free shipping is losing money?

Compare the gross margin on items shoppers add to qualify against the real shipping cost you absorb on newly qualifying orders, and track contribution per order rather than revenue. If shipping absorbed exceeds added margin, the threshold is buying revenue at a loss.

### Should the threshold be a round number?

Round numbers are easy to communicate, and that is their only virtue. Set the number from your AOV gap and margin first, then round it to the nearest clean figure inside the range that works. A round number outside the working range is just a wall.

### How long should I test a new shipping threshold?

Long enough to cover your normal buying cycle and a few hundred orders at minimum, with no other shipping or banner changes running at the same time. Short tests mostly measure novelty and weekend mix, not the threshold.
