What Is Price Testing? Why Fewer Orders Can Win

Quick Definition

Price testing is trying different prices for the same product to see which one earns more. You can split visitors into groups that see different prices at the same time, or change the price for a while and compare it with the weeks before. Judge the result by profit per visitor, not by the number of orders.

Two Ways to Run a Price Test

In a split test, visitors are sorted at random into groups. Each group sees a different price for the same product at the same time. Since they shop in the same week, price is the main thing that differs. It's one kind of A/B testing.

Or run a before-and-after test: charge $30 for two weeks, then $36 for the next two, and compare. Everyone sees the same price on any given day. But anything else that changed in those weeks, like a holiday or a new ad, gets mixed into the result.

Either way, two things pull against each other. A higher price earns more per sale, but it usually means fewer sales. The test tells you which one wins for your product and shoppers.

Fewer Orders, More Profit

Say you sell a candle that costs you $12. You test $30 against $36 and show each price to 5,000 visitors. Every order is one candle.

	$30 price	$36 price
Visitors	5,000	5,000
Conversion rate	2.4%	2.0%
Orders	120	100
Revenue	$3,600	$3,600
Profit per candle	$18	$24
Gross profit	$2,160	$2,400
Profit per visitor	$0.43	$0.48

Look only at conversion rate and you'd pick $30. On revenue, it's a tie. Profit is what settles it: $36 earns $240 more ($2,400 − $2,160).

$36 stays ahead while its conversion rate is above 1.8%. At 1.8% you'd get 90 orders, and 90 × $24 = $2,160, which just ties the $30 price.

Still, a gap of 20 orders could be luck. Check for statistical significance before you change the price for good.

Where Price Tests Go Wrong

Not enough orders. With only a handful of sales a week, you may wait months before the result means anything.
Shoppers talk. In a split test, two friends can see two prices for the same item on the same day. If they notice, both may trust you less, and in some places it can raise legal questions.
Discount codes muddy the price. When lots of buyers use a code, they aren't paying the price you're testing. Look at what people actually paid.
Ads show one price. Someone who clicks an ad or shopping feed listing may land on the other one.
Later effects haven't shown up yet. Two weeks of data covers first orders, not whether a higher price changes repeat buying months later.

Price Testing on a Shopify Store

In Shopify, every variant has its own price, and you can add a "Cost per item" to each. Fill that in, or you'll see sales but not profit. A split test also needs a setup that keeps each visitor on one price and charges it at checkout.

A lower-risk route is to keep the base price and test how it's shown. Try price anchoring with a "was" price, an $89 versus $90 ending, or the size of a discount. Our Shopify A/B testing guide for offers and prices covers what to test first and how long to run it.

Growth Suite, our app, has an A/B testing module that compares offers with different discount depths or urgency. You choose the goal that picks the winner: conversion rate, AOV or total revenue. It changes the discount, not the regular price. To test the regular price before and after, the app's Product Price Editor can raise or lower prices. It keeps Google Shopping and Meta Ads in sync, and one click puts the old price back.

A Sale Is Not a Price Test

A 20%-off week can look like a free price test, but it isn't one. A sale brings emails, a deadline and a rush of deal hunters, and any of that can push orders up by itself. So a sale tells you how the lower price sells with all that help, not on a normal Tuesday.

Price testing isn't dynamic pricing either. A test ends with one price you keep. Dynamic pricing keeps moving with demand, time or competitor prices.