A/B testing for Shopify

Find the version of your store that makes more money.

Variantly tests content, images and prices on your storefront, reads the results in plain English, and applies the winner in one click.

No theme editsWorks on every Shopify planFree trial through Shopify

Example · product page test · day 18
A · OriginalB · New images

Variant B is ahead

94% chance to beat A · +12% revenue per visitor

Apply winner

What you can test

Test the changes that move revenue.

Pick a change, choose how much of your traffic sees it, and launch. No theme code, and one toggle to switch it all off.

Content and sections

Headlines, banners and whole page sections. Try the bolder pitch on half your visitors before you commit to it.

Product images

Lead with a different photo or reorder the gallery, and find out which image actually sells the product.

Prices

Try a new price and let Shopify’s own discount system enforce it at checkout. Judged on profit, not just orders.

Every plan, not just Plus

Results

Results you can read without a statistics degree.

No jargon and nothing to decode. Variantly tells you which version is winning, how sure it is, and what to do next.

VariantVisitorsConversionRevenue / visitorProfit / visitor
A · Original4,1121.9%£0.55£0.31
B · New images4,0872.2%£0.62£0.36
Variant B: 94% chance to beat the original.Expected lift between +4% and +19%Worked example
  • Chance to beat is how likely it is that B really is better, rather than lucky.
  • Profit per visitor counts your product costs and any test discounts, so a cheaper price cannot fool you.
  • Order matching ties every order to the version the shopper actually saw, using your store’s own order data.
  • Consistent shoppers keep one version for the whole test, on every visit.

Worth the bother?

See what a winner is worth to your store.

Set the sliders to your store and see the yearly value of a lift at the low end, middle and high end of the example above.

Straight arithmetic: revenue × lift × 12. Only a test tells you where your lift lands; finding it costs nothing on the Free plan.

Your store today: about £6,050 a month

Low end +4%£2,900a year
Middle +12%£8,710a year
High end +19%£13,800a year

The honest bit

Honest when your traffic is low.

Most tools will happily call a winner from a handful of orders. Variantly will not.

Verdicts wait for evidence

A stricter reading kicks in when traffic is low, and the verdict holds until the data has earned it.

“Not yet” is an answer

When a test cannot conclude anything, Variantly says so in plain words, along with what to do next.

Noise never wins

The low-traffic mode stops you calling a winner too soon and shipping a change built on luck.

How it works

From install to applied winner.

01

Install and switch on

Install Variantly, then enable the app embed in your theme settings. Two minutes, and no code.

02

Launch a test

The wizard walks you through what to change, who sees it, and how your traffic is split.

03

Apply the winner

One click keeps the winning version, with rollback and a full audit trail if you change your mind.

See the full walkthrough

Test ideas

Out of ideas? Steal these.

Six tests merchants run first. Each one takes minutes to set up, and every one has surprised somebody.

Product page

Lead with the lifestyle shot

A Gallery opens on the packshotB The in-use photo goes first
Product page

Shorten the pitch

A Full description above the foldB Three plain bullet points
Home page

Make the offer the headline

A Brand strapline in the bannerB “Free delivery over £40”
Price

One pound under the round number

A £30B £29
Price

Dare the raise

A £24, today's priceB £27, judged on profit
Product page

Reorder the gallery

A Current image orderB Best-selling angle first

Price testing

The bravest test, made safe.

A

£29

Today’s price

Conversion 2.0%

Revenue / visitor £0.58

Profit / visitor £0.34

B

£25

Test price, enforced at checkout

Conversion 2.3%

Revenue / visitor £0.58

Profit / visitor £0.30

Variant B sold more, matched A on revenue, and earned less. Profit per visitor caught what revenue hid. Variantly judges price tests on what reaches your bottom line.

How price testing works

Questions

Fair questions, straight answers.

Will it slow my storefront down?

Variantly adds one small script through Shopify's standard app embed. A shopper's version is decided once, then remembered, so nothing is re-decided on every page.

Do I need to touch my theme code?

No. You switch on the Variantly embed in your theme editor with one toggle. Switching it off again is the same toggle, and your theme is left exactly as it was.

Will shoppers see prices jumping around?

No. Each shopper keeps the same version, and the same price, for the whole test. The price they see is the price they pay, enforced at checkout.

Can I undo an applied winner?

Yes. Applying a winner is one click, rolling it back is one click, and every change is recorded in an audit trail.

What if my store does not get much traffic?

Variantly has a low-traffic mode that reads your results more strictly and never invents a verdict. It tells you honestly when there is not enough data yet, and what your options are.

Run your first test this week.

Install Variantly, switch on the embed, and launch a test on the page you care about most. Your store has a better version. Go and find it.

No theme editsFree trial through Shopify