Do price framing A/B tests work in E-commerce?

In line with both baselines. Price framing tests win 12% in E-commerce, against 13% for price framing tests generally and 14% for E-commerce generally. Neither the category nor the vertical is moving this much.

The strongest change type we can measure in E-commerce is CTA copy at 17% on 50+ tests.

Small winner set. Few of these tests cleared the bar. The 95% interval on that win rate runs 7% to 22%, so read the figure as directional.

The numbers

	E-commerce	all industries
Tests analysed	50+	100+
Beat control	12%	13%
No measurable difference	77%	—
Lost to control	11%	—
Median lift when it won	not reported	—

Median traffic per variant was 1877 visitors. Most ran on product pages (37), homepages (25), category pages (8).

Every change type we can measure in E-commerce

change type	win rate in E-commerce	tests
CTA copy	17%	50+
Layout	16%	100+
Styling	16%	500+
Hero image	15%	50+
Body copy	14%	100+
Price framing (this page)	12%	50+
Split URL	11%	250+
Headline	8%	100+
Social proof	8%	50+

The industry overview is at E-commerce benchmarks; rates across all industries are on the price framing benchmark.

What counts as a price framing test

The number stays the same; the presentation changes. Per-day instead of per-month, annual instead of monthly, fee broken out instead of rolled in, anchor shown or hidden.

Control and variant wireframe for a price framing A/B test — Control on the left, variant on the right. Only the changed element is highlighted.

Price framing tests that won in E-commerce

CTA/price copy offering a plan priced with a discount noted
Hides the default price, and adds price displays to one-time and subscription purchase labels, showing a 10% discounted price with struck-through original price for the subscription option
Adds a Save X discount label next to the sale price and restyles the price box layout and sizes
Hides the base price, adds original price plus a 10%-discounted subscription price with strikethrough next to purchase option labels, and relocates the pack-size selector into the add-to-cart aside.

Price framing tests that did not

In the cart drawer, hides the original price element and strips the original total label while showing the discounted total with strikethrough styling
In cart drawer, hides the original pre-discount price element, shows the compare-at price with the original total label removed and strikethrough styling applied
Adds VAT labels next to prices and relocates/renames the wishlist link to Save for later, plus adds USP text for a specific product

These are individual tests, not rules. Each ran on one site, with one audience, against one page we are not showing you. They are picked to be illustrative rather than sampled at random, and a change that won here can lose on your page for reasons none of this captures. Read them as prompts for what to test, not as findings to copy.

Who ran these tests

This is every Mida account that ran a readable test — in-house marketers, founders, product teams, and agencies working on client sites. Nothing here is filtered by who ran the experiment or how experienced they are.

Low win rates are normal in experimentation, including at the top end. Microsoft's experimentation team, reporting on its own platform, found that only about one third of ideas improve the metric they were designed to improve — and that roughly another third actively hurt it. That is a dedicated experimentation organisation with research, prioritisation and review behind every test.

A mixed population like this one runs below that. The gap is roughly what disciplined practice buys you: ideas grounded in research rather than opinion, one variable at a time, and tests built so the result can actually be read.

So treat these as a general base rate, not a target. If you would rather not close that gap the slow way, find a CRO agency in the United States, the United Kingdom, the Netherlands or Australia — or browse every region.

A win is a variant that beat its control on that test's primary goal with a statistically significant result. Tests that never got enough traffic to say anything either way are excluded. The methodology has the full detail, including what these numbers cannot tell you.