Skip to content
SquareUp
Storefront experiments

You are guessing what actually sells.

Opinions pick your headline, not data.

Test it, and keep the winner.

Test any part of your store, keep what sells more.

Try a new headline, a new section order, or a new product order against your live one. Half your visitors see each, and you keep the version that sells more.

Join the waitlistShop the live demo
  • More than the headline
  • Design a variant by describing it
  • We never call a winner early
Sage, Salesperson

Your AI team

Run by Sage, salesperson. Works with Muse, Ace, Rave.

Why it matters

Test more than one line

The headline, the call to action, your section order, your product order, or whether a section shows at all. If a shopper sees it, you can test it.

Design a variant by describing it

Tell the Storefront Designer the change you want and it draws the variant. Your live store stays exactly as it is while you build it.

An honest read, never a lucky day

Results show in plain words, and we never call a winner early off one good day. You keep a version once it is really selling more.

Keep the winner in one tap

When a version is clearly ahead, make it your store with one tap. Nothing changes until you do.

How it works

How a test runs

One candle store, one better headline.

  1. Step 1 of 4

    Pick what to test

    • Choose any part of your store: a headline, the call to action, a section's order, your product order.
    • Your live store is Version A. You add the challenger as Version B.
    • A fixed, fair split: half your visitors see each, kept steady the whole time.

  2. Step 2 of 4

    Design the challenger

    • Describe the change and the Storefront Designer draws it.
    • Preview the exact version a shopper in the test would see.
    • Your live store never changes while you build the variant.

  3. Step 3 of 4

    Let it gather real data

    • Both versions run at the same time, on the same traffic.
    • Results read in plain words: added to cart, reached checkout, bought.
    • We never call a winner early off one lucky day.

  4. Step 4 of 4

    Keep the winner

    • When a version is really selling more, promote it in one tap.
    • It becomes your live store; the test is put away.
    • Not sure yet? Leave it running until the answer is clear.

A result you can trust, not a lucky day.

Both versions run on the same traffic at the same time, and we keep checking as visitors arrive, so a good first day can never be mistaken for a real win. You see the lift as a range, we watch that the split stays fair, and the number we show is the honest one. What you keep is what actually sold more.

Everything in the box

Test more than the headline

Hero headline, subtext, and imageCall-to-action wording and placementYour homepage section orderYour product grid orderHide or show a section

Design without code

Describe it, it gets drawnPreview the exact shopper versionLive store untouched while you buildSaves to the test, not live

An honest read

Plain-word results, cart to boughtOnly calls a win that holdsLift shown as a rangeFair split, watched throughoutGoal locked before the testThe honest lift, never a spikeOnly clean visits count

You stay in charge

Nothing changes until you promotePromote the winner in one tapStop a test any timeOne test per part

SquareUp

What you stop paying for

The A/B testing apps you rent per test.

Intelligems price testingVisually.io experimentsShoplift A/B testsNeat A/B Testing

Live in three steps

1

Pick what to test

Choose the part of your store to try a new version of. Your live store is Version A.

2

Add a challenger

Type the change, or design it in the Storefront Designer. It saves to the test, not your store.

3

Keep the winner

Let it gather real data, then promote the version that sells more in one tap.

Goes well with

All features

The seller

Your best salesperson, for every visitor

Learn more

Price negotiation

Haggles to a floor you set, never below

Learn more

Living storefront

A homepage that composes itself per shopper

Learn more

Smart discovery

Search by vibe, never a dead end

Learn more

Fair questions

What can I actually test?
Any part of your storefront a shopper sees: the hero headline, subtext, or image, the call to action, the order of your homepage sections, the product order in your grid, or whether a section shows at all.
Does my live store change while I build a test?
No. Your live store stays exactly as it is. A challenger version is built and previewed on the side, and only becomes your store if you choose to promote it.
How do I know the winner is real and not a lucky day?
Both versions run at the same time on the same visitors, and we keep checking as more people arrive, so a strong first day can never be mistaken for a real win. You see the lift as a range so you know how sure we are, we watch that the split of visitors stays fair, and you set the goal before the test starts so it can't be changed later to look good. You only keep a version once it is clearly selling more.
Do I need to know how to code or design?
No. Describe the change you want and the Storefront Designer draws the variant for you, then saves it to the test. You preview it before anyone sees it.
How many tests can I run at once?
You can run tests across your store, but only one runs on a given part at a time, so results never get tangled. Start another the moment one finishes.
Sells more

Stop guessing what sells. Test it, and keep the winner.

Clicks don't convert

Stop guessing what sells on your store.

Test the headline, your section order, or your product order, and keep the version that sells more.

We only use this to reach out when your spot opens.

We use your email and phone only to talk to you about getting in. See our privacy notice.