In short
Short answer
- Only clients on Shopify Plus can test content inside the information, shipping and payment steps of checkout. That is Shopify's rule for every app.
- Clients on other plans (except Starter) can still test the Thank you and Order status pages, and price, offer and shipping tests change what checkout charges on every plan.
- Checkout tests count only shoppers who reach checkout. Most of them already intend to buy, so expect small changes and plan the sample on checkout traffic, not store traffic.
- One-click post-purchase upsells depend on Shopify's approval for each app, so check the status before you promise one to a client for BFCM.
- Late October is late for a checkout test that has to finish before the freeze. Be honest with clients about what can be settled by November 15.
Who this is for: agency strategists deciding which checkout and after-payment tests to run for each client before BFCM.
Timeline for 2026
| Date | Step |
|---|---|
| Wednesday, October 28 | Sort clients by plan; check checkout traffic per client |
| By Sunday, November 1 | Latest launch for a two-week test |
| By Sunday, November 15 | Read and decide; checkout blocks for the sale agreed |
| Monday, November 16 | Freeze: no checkout editor changes |
| November 26 to 30 | Sale week: watch payment errors and support contacts |
Checkout changes during the freeze are the riskiest changes a client can make. Whatever goes into checkout for the sale should be in place, and tested, by November 15.
Step 1: Sort clients by plan
Shopify's developer documentation says checkout UI extensions for the information, shipping and payment steps "are available only to stores on a Shopify Plus plan" (read October 12, 2026). Development stores can use them too, which is useful for building and checking a test before a live Plus store runs it.
| Client plan | Inside checkout | Thank you and Order status pages | Price, offer and shipping tests |
|---|---|---|---|
| Shopify Plus | Yes | Yes | Yes |
| Basic, Shopify, Advanced | No | Yes | Yes |
| Starter | No | No | Check the app's requirements |
Confirm the plan in each client's Shopify admin rather than from memory. A client who moved plans this year may not have told you.
Step 2: Pick checkout tests that change a decision
Checkout is where shoppers hesitate over trust, delivery and the final total. Good candidates for Plus clients:
- Trust near the payment step: secure payment, returns and a guarantee, against the checkout as it is. Mechanism: reassurance where the card number is typed.
- A delivery promise under the shipping methods, only where the warehouse already meets it. Mechanism: removing doubt about arrival before the holidays.
- A low-priced add-on that suits most carts. Mechanism: order value, which can cost conversion if it distracts. Judge it on revenue or gross profit per shopper who reached checkout, not on how many added it.
Avoid testing anything that changes which shipping rates or payment methods appear in BFCM week. Those are configuration decisions with a support cost, and a bad arm hurts the client's busiest days.
Step 3: Size the test on checkout traffic
A checkout test only counts shoppers who reach checkout. That means fewer visitors than a page test, but a much higher baseline rate.
(an illustrative Plus client, not a real store)
The client has about 800 shoppers a day reaching checkout, and 45% of them place an order. With two arms (the current checkout and one change), 95% confidence and 80% power:
| Smallest change worth detecting (relative) | Shoppers per arm | Days at 800 a day |
|---|---|---|
| 5% (45% to 47.25%) | 7,705 | 20 days |
| 3% (45% to 46.35%) | 21,371 | 54 days |
A test launched on October 28 has 18 days until November 15. It can't settle a 5% change in time, and a 3% change belongs to January. The honest options are to start the 5% test now and read it after BFCM, or to run it from January and leave checkout as it is for the sale.
Sizes come from the same formula Benchmyrk uses to judge results; plan each client's numbers with the sample size calculator, and see sample size and test duration for the reasoning. Run whole weeks, at least two.
Step 4: Non-Plus clients: test after payment
For clients not on Plus, the Thank you page is the best after-payment surface. The order that showed the page is already paid, so it never counts for the test. What counts is what the buyer does next.
Useful tests:
- A next-order code against no offer, judged on repeat orders within 30 days and the revenue they bring after the discount.
- A referral prompt against a returns or care note, judged on referral clicks first and repeat orders later.
- A one-question survey ("How did you hear about us?") in one arm, to check it doesn't pull attention from the code.
A repeat-purchase goal needs its window to pass. A test judged on repeat orders within 30 days can't be read until 30 days after the last buyers joined, so a Thank you page test started now reports in January. Set that expectation with the client up front.
Step 5: One-click post-purchase upsells
Shopify's post-purchase page, between payment and the Thank you page, lets a buyer add a product to the order they just placed with one click. Shopify treats it as a beta: an app can use it freely on development stores, but needs Shopify's approval for live stores, and each store picks a single post-purchase app. It is available on every plan except Starter, and Shopify skips the page for some payments, such as wallets and installment providers (Shopify developer documentation, read October 12, 2026).
Before you put a post-purchase upsell in a client's BFCM plan:
- Check which post-purchase app the client already uses. There can be only one.
- Check that the app you plan to use has Shopify's approval for live stores.
- Judge it on upsell revenue net of refunds and on profit per order, not on take rate alone.
Mistakes to avoid
- Promising checkout tests to a non-Plus client. Shopify doesn't allow app blocks inside checkout on their plan.
- Sizing on store traffic. Only shoppers who reach checkout count.
- Testing shipping or payment options in sale week. Make those decisions before the freeze.
- Reading a repeat-order test early. The window has to pass for every buyer in the test.
- Counting the order that showed the Thank you page. It was paid before the test could change anything.
Checklist per client
- [ ] Shopify plan confirmed in the client's admin
- [ ] Plus: one checkout test, sized on checkout traffic, with a realistic end date
- [ ] Non-Plus: a Thank you page test with a repeat-order window the client understands
- [ ] Post-purchase app and Shopify approval status checked before promising an upsell
- [ ] Primary goal agreed: orders per shopper who reached checkout, or revenue per shopper
- [ ] Checkout editor changes finished by November 15
Planning checkout for one brand? Start from the BFCM hub. The general timeline is in the guide to testing before Black Friday.