You can compare Shopify checkout offers without an A/B test by running one offer, turning it off, running the next, and counting orders yourself. SK Checkout Customizer does not include an A/B test. The catalog also does not include per-widget scheduling, translations, or revenue analytics. Two headlines are still comparable. The comparison happens in Shopify’s orders, not in a switch inside this app.
That limit is the whole method. There is no hidden experiment, no automatic winner, and no chart of conversion rate to quote. If a landing card mentions A/B testing, it is ahead of the catalog. Follow the catalog. The app is free either way. Free does not add a test tool the product does not have.
What you will not find in the app
An A/B test would split traffic between two versions at the same time and declare a winner. This app does not do that. A scheduler would publish and unpublish a block on a calendar. The catalog does not include one. Translations of the offer text are not in the catalog either, so write the headline in the language your buyers read. Revenue analytics would attribute money to the block. The catalog does not report impressions, clicks, or attributed revenue for these widgets.
- No A/B test: you cannot run two checkout offers on the same shoppers at once inside this app.
- No scheduler: you turn a block on and off yourself.
- No translation set: the words you type are the words shoppers see.
- No revenue report: Shopify admin is where you count the orders.
Checkout-step blocks still depend on your Shopify plan. Shopify’s help on checkout configurations documents app blocks on the information, shipping, and payment steps as Plus, and thank-you and order-status app blocks on Basic and higher. Compare offers only on a page the editor actually allows. A test that never rendered is not a failed headline.
Run one offer, then the other
Pick a single change. A different product, a different headline, or a different discount banner is enough. Changing all three at once means you will not know which change mattered. Write the dates down before you start. One full week against the next full week is easier to read than a Tuesday against a Saturday, because your traffic already swings by weekday. Use the same page both times: both on the thank-you page, or both on the Plus checkout step. Do not compare a payment-step bump with a thank-you row and call that a headline test. The moment of the offer changed, not only the words.
Between the two runs, remove the first block or replace its copy. Leave a note in your calendar so nobody “improves” the offer halfway through the second week. The app will not freeze the configuration for you.
Keep the offered product in stock for both periods. A sell-out in week two looks like a weak headline. If you cannot hold stock, count only the days both products were available, and write those dates on the sheet. The same rule applies to a discount code: if the code hits its usage cap halfway through, the period is over. The app will not warn you. Shopify’s discount screen will.
US, UK, and Canadian stores can use the same sequence. If you sell in more than one country, keep the country mix similar across the two periods, or the result will describe traffic, not the offer. The catalog will not segment that for you.
What you have to count outside the app
Decide the unit before the first order comes in. For an order bump, count orders that contain the add-on product. For a thank-you upsell, count orders placed after the visit that include the offered product, and be honest that a later purchase is not the same as a click on the thank-you page. For a discount banner, count uses of that code in Shopify discounts. Export the orders or read the report you already trust. Write the two totals next to the dates.
Also write down what you will not claim. A higher add-on count in week two can come from a sale, a email blast, or a stockout of the main product. If a promotion overlapped one week, throw the comparison out or note it beside the totals. The empty chart in your notes is intentional. Fill it with your figures. Do not borrow a percentage from an app that does not calculate one.
- Record the start and end date of each offer.
- Record the page: thank-you, order status, or a Plus checkout step.
- Record the product or the discount code you will count.
- Record any other campaign that ran in the same days.
How to choose what stays
Keep the offer that produced more of the outcome you defined, if the two periods were otherwise comparable. If they were not, run the pair again. There is no statistical calculator in this app, and this article will not invent a sample size that “proves” a winner. A quiet store may need a longer pair of periods than a busy one. You can see that from your own order volume.
If you need a simultaneous split, this is the wrong product for that job. Say so before someone on the team hunts for a variant setting that is not there. The offers you can configure, the order bump, the checkbox upsell, the thank-you product list, are still available. They are simply not an experiment platform.
Questions about comparing checkout offers
How is a winner chosen?
You choose it from order data you export or read in Shopify. The app does not pick a winner.
Can I test on the thank-you page only?
Yes. Run one thank-you offer, then replace it with the next, and count outside the app. You still will not have a simultaneous A/B split.
Does a free plan add the test later?
The app is free and the catalog has no A/B test on any tier, because there is no paid tier.
Change one offer at a time on SK Checkout Customizer, then read the orders in Shopify. Install SK Checkout Customizer for the page your plan allows, and keep the comparison in your own sheet.