> ## Documentation Index
> Fetch the complete documentation index at: https://developers.perform.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Checkout Widget A/B Testing

> Compare your current delivery date display against the Perform.AI date on live traffic and measure the difference in conversion rate for your storefront.

**Available on request.**

## Summary

Checkout Widget A/B Testing compares two delivery date displays on your own live traffic and reports the difference in conversion rate. One group sees the Perform.AI estimated delivery date; the control group sees whatever you show today, measured through an invisible widget that changes nothing on the page.

## Who it’s for

E-commerce leads and conversion owners who need to know what the delivery date is worth before committing to a rollout, and who cannot get that answer from a before-and-after comparison, a vendor benchmark or an experimentation platform that cannot produce a delivery date. It also serves teams that want to measure their current conversion rate before anything on the storefront changes.

## Key benefits

* **A baseline without changing the storefront.** The invisible widget counts sessions and conversions on your current setup while shoppers see nothing different.
* **The comparison is made on your own traffic.** Control and test groups run at the same time, over the same period, on the same storefront, measured the same way, so the delivery date is the only thing that differs.
* **Test the date, the wording or the placement.** Variants can differ in how ambitious the date is, in format (a weekday, a calendar date, days remaining) and in which pages carry the widget and where.
* **Read the result where the date is produced.** Conversion rate and funnel drop-off per group come from [Checkout Analytics](https://developers.perform.ai/products/checkout/checkout-analytics.md), with the same session definition, coverage rule and reporting window.
* **Roll out a share of traffic rather than all of it.** A controlled share of shoppers sees the Perform.AI date while the rest carries on unchanged.

## How it works

Variants are widget variants of the [Checkout EDD Widget](https://developers.perform.ai/products/checkout/checkout-edd-widget.md), and the baseline is the invisible widget placed on a pre-checkout page and the order confirmation page. Variants, the traffic share and the baseline are configured by Perform.AI. The conversion rate and funnel drop-off a test is judged on are the Checkout Analytics numbers, so a test needs a widget on the order confirmation page, uses the fixed rolling 14-day window, and refreshes nightly. The portal reports the rates; judging whether a difference is statistically significant is done outside it, and a conversion is an order confirmation widget load, not a verified order. Widgets implemented through the Shopify app have no analytics coverage, so those stages cannot be tested.

Contact your account team to enable this feature.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.