Run an experiment¶

An experiment runs a panel of synthetic shoppers against two versions of one page — your current page and a new version you want to test — and files a verdict on which converts better. You supply both versions. Squoosh does not write the new version for you; it tests the real change you made.

This page walks through the wizard from start to launch.

Before you start¶

Two things must be true before Squoosh will launch a run:

• Your synthetic shoppers are calibrated. Calibration builds the panel of shoppers that runs your test. Until it's done, the launch button is disabled and the wizard shows a Set up shoppers link. See Calibrate synthetic shoppers.
• Your workspace has enough credits. Each run costs credits, and the wizard shows the exact cost before you launch. If your balance is short, the wizard says so, and you can top up. See Credits and cost.

You also need the URL of version B, the new page you want to test. Squoosh tests the page you point it to; it does not generate a variant.

Open the wizard¶

1. In the sidebar, click Experiments.
2. Click New experiment.

The wizard opens as a single sheet with numbered sections. Fill them top to bottom.

Target¶

Section 01 Target is where you name the experiment and enter the page you're testing against.

1. In Control page (A), enter the URL of your current page. A valid URL shows Looks valid.
2. In Experiment name (optional), type a name that describes what you're testing, for example PDP layout v2 vs current. Leave it blank and Squoosh names the experiment for you from the page and goal.

Control page (A) is your current, live version, the baseline every result is measured against.

Variant¶

Section 02 Variant is where you tell Squoosh how to reach version B. Squoosh does not generate the variant. You bring it, and Squoosh tests your real change.

Under How do you ship variants?, choose one mode:

For Two URLs:

1. Confirm the A · Control card shows the control URL from the Target section.
2. In Variant page (B), paste the URL of the new version. It must be a valid URL and must differ from the control. When it's accepted you'll see Looks valid · differs from control.

Goal and traffic¶

Section 03 Goal & traffic sets what you're optimizing and how shoppers arrive.

1. Under Primary goal, choose Add to Cart (default) or Checkout, the conversion the experiment measures.
2. Under Traffic profile, choose Organic / Direct (default) or Paid Ad (PDP), where shoppers enter from.

Traffic profile sets the source and intent of your shoppers. It is not a split between the two versions; there is no traffic-split control.

Choose the rigor¶

1. Drag the Rigor slider to a level. It snaps to Standard (the default), Medium, High, Extra High, Max and Ultra. The description under the slider says what each level is for.
2. Read the methodology note under the slider. It changes with the level and tells you which kind of lift the run reports:
   • Estimated lift (Standard and Medium). Squoosh's AI reviewer compares the two pages and grounds its read in what the shoppers did. Treat it as a directional read, not a statistically conclusive result.
   • Measured lift (High and above, Two URLs only). Squoosh counts verified conversions on each page across paired synthetic shoppers and reports the difference with a 95% sampling interval. The note shows the interval each level gives: it is wide at High (about ±27 percentage points) and narrows as rigor rises, so a lower level can still come back with no clear winner. It measures synthetic shoppers, not your customers.

Higher rigor runs a larger experiment, so it takes longer and uses more credits; the footer shows the exact cost. Standard and Medium run in real browsers on your live pages. High and above run on captured copies of your pages from the Site Crawler, so they appear only when Site Crawler is turned on in Settings → Features, and only in Two URLs mode. Extra High, Max and Ultra are very large runs: they can take hours or, at Ultra, days to finish, and Ultra uses an extremely large number of credits. A measured run also needs your shopper pool fully calibrated; if it isn't, the launch is refused before any credits are charged and asks you to re-sync your shoppers.

The VWO / Optimizely / AB Tasty / Convert and Feature flag modes offer Standard and Medium only and report an estimated lift, because a measured result needs a separately hosted variant. Switch to Two URLs to use the higher levels. Demo workspaces always report an estimate.

Capture both pages (High and above)¶

The higher levels run on Site Crawler captures of both versions for every device your shoppers use. Review & run shows which captures are ready and offers to capture a missing one; Run experiment stays disabled until every capture is ready.

Verify conversion targets (High and above)¶

When you choose High or above with Two URLs, the Review & run section shows a Conversion capture block. It is required for a measured result.

1. Squoosh verifies the conversion target (add to cart, checkout or the primary call to action) on both pages automatically as soon as both URLs are valid.
2. If a page has more than one possible target, Squoosh asks you to choose the target on each page. Pick them, then click Re-check targets.
3. For a Checkout goal, Squoosh verifies the checkout page from the Site Crawler's captured checkout view for each version's store — you don't enter a checkout URL. The block shows the captured view per version (when it was captured, which crawl, and the page). If a version has no captured checkout view yet, or the newest one isn't a usable checkout page, click Capture with Site Crawler (or Open Site Crawler in the footer). It opens the Site Crawler in a new tab with a checkout capture ready to start from that version's page — start the capture from a product page, so the crawler can add an item to the cart and open the checkout. Squoosh re-checks the targets on its own when the crawl finishes; you can also click Re-check targets.
4. Wait for Both conversion targets are verified. The check expires after 15 minutes; if it lapses before you launch, click Re-check targets again.

Run experiment stays disabled until the check passes. If Squoosh reports that conversion capture isn't supported on one of the pages, choose Standard or Medium rigor for an estimated result, or test different pages.

Review and run¶

Section 04 Review & run summarizes the run and lets you edit the shopper tasks.

1. Check the summary grid, including Type, Variant via, Goal, Rigor and Method. Method reads Estimated · AI review at Standard and Medium and Measured · conversions at High and above with Two URLs.
2. Review the Shopper tasks, the shopping objectives each shopper attempts, drafted for you while you filled the form. Edit any task inline, or remove ones you don't want. At least one task is required. The list you see is the list that runs.
3. Confirm Cost, the credits left after launch, and Execution: Priority in the footer.
4. Click Run experiment.

The selected level's shoppers run before Squoosh finalizes the result. A/B experiments do not currently stop when an interim confidence score crosses a threshold. See Experiment lifecycle and stopping.

If the launch button is disabled, one of the gates in Before you start isn't met: shoppers aren't calibrated, the variant URL is missing or matches the control, or your credit balance is short. At High and above it also stays disabled until the captures and the conversion-target check are ready (see Verify conversion targets).

After you launch¶

Squoosh reserves the panel, warms up both versions, and moves you to the live experiment page. The results fill in as the first shoppers enter the funnel, so you can watch the run build.

When the run finishes, read the verdict at the top of the experiment page. For an estimated run (Standard or Medium) that is the outcome, the lift, and the confidence. For a measured run (High and above) the page is labeled Village conversion lift and shows each version's conversion rate, the difference and its 95% sampling interval instead of a confidence meter; see Conversion-evidence results. "Keep the original" is a real outcome; a change that costs conversions is worth knowing before you ship it. There is no ship or deploy button. Rolling out the variant is your call.

See How Squoosh decides a winner and Reading lift and confidence for how to read the result.

Credits and cost¶

Your workspace shares one credit balance. An A/B test costs 30 credits for 15 paired shoppers, 100 credits for 50, or 200 credits for 100. The price is per session at every size; there is no separate charge for the method. The 100-shopper option is the one that reports a measured lift instead of an estimate. The wizard shows the selected cost and your remaining balance before launch. See Credits and tests for the sidebar estimate, Conversion Report pricing and monthly limits.

Site Crawler is a separate capture feature with its own reservation and refund rules; its saved screenshots do not yet replace experiment sessions.

Your plan caps how many tests you can run per month and how many properties you can have. Hitting a cap prompts you to upgrade. To add credits or change your plan, contact the Squoosh team at support@squoosh.ai; self-serve billing isn't available yet.