Black Friday is November 27. What to change on your store before then, and what to leave alone

Between now and Black Friday on November 27, change only what is cheap to roll back and quick to check: offers, shipping thresholds, cart messaging, paid landing pages and post-purchase offers. Leave the theme, the app stack and checkout code alone. Structural changes made in November tend to break in the one week a store cannot afford to be down.

What the week is worth this year

Adobe forecasts $275.1 billion in US online holiday spending between November 1 and December 31, up 6.7%. Cyber Week alone, Thanksgiving through Cyber Monday, is forecast at $47.5 billion. Black Friday is expected to bring $12.9 billion, up 9.2% and growing faster than Thanksgiving or Cyber Monday, and Cyber Monday on November 30 about $15.1 billion.

Two more numbers from the same forecast shape what to fix. Mobile is expected to carry 57.4% of the season's online spending. And discounts are forecast to peak at up to 30% off list during Cyber Week.

That second number matters more than it looks. When most competitors land on similar discounts, price stops separating you from them. What separates you is how easily a phone shopper understands the offer, trusts it and pays.

Change now, or leave alone

Wallets now carry a third of the orders

Shopify's own figures for last year show why the wallet row sits near the top of that table. Merchants on the platform took $14.6 billion over the 2025 Black Friday Cyber Monday weekend, up 27%, and 32% of those orders were placed with Shop Pay.

The shipping threshold row earns its place for a similar reason. Baymard's research into why shoppers abandon carts keeps finding extra costs shown late as the leading cause. A cart message that says exactly how far the shopper is from free shipping turns that cost into a reason to add one more item.

A third of buyers choosing the one-tap option tells you how many people do not want to type a card number on a phone. If your express buttons only appear on the final checkout step, move them up to the cart and the product page. It is a settings change, it is reversible, and it removes the slowest part of a mobile purchase.

Most A/B tests are now too late to finish

A test needs enough orders to separate a real lift from noise, and it should run in whole weeks. Three weeks out, only very high-traffic pages can still reach a reliable read before the freeze.

So for most stores the right move in November is not to test but to ship well-understood fixes and watch them closely: a clear threshold message, wallets where thumbs can reach them, delivery cut-off dates stated plainly. Save the uncertain ideas for January, when traffic is cheaper and the result can be read properly.

The post-purchase offer is the safest lever left

A post-purchase offer appears after the order is paid, so it cannot lose the sale it follows. That makes it the one meaningful revenue change you can still make in late November without risking conversion.

In a supplement programme I led for Optimize Minerals, post-purchase upsells added about 10% to average order value. Pick one complementary product, price it to feel like an add-on rather than a second decision, and make the decline button as easy to find as the accept button.

AI shoppers arrive with the question already asked

Adobe also expects traffic to retail sites from AI assistants to rise 130% this season. Those visitors often land on a product page having asked an assistant about price, delivery dates and returns before they ever see your site.

Give them the answers as plain text on the page, not inside images or accordions: the sale price, the order-by date for delivery before the holidays, and the return window. Text an assistant can quote is text a shopper can trust.

The freeze plan

Pick a code freeze date, ideally the Tuesday before Thanksgiving, and hold it: no deploys, no app installs, no theme edits. Test the rollback once before the freeze, not during an outage.

Then watch checkout conversion by device every hour on Black Friday and Cyber Monday. A sudden drop on mobile is almost always a broken element, not a change in demand, and the hour you spot it is worth more than any test you could have run.

Plan the watch around the peaks, not office hours. Shopify recorded its busiest minute of last year's weekend at 12:01 p.m. Eastern on Black Friday, when merchants processed $5.1 million in sixty seconds. Have someone who can roll back a change awake and watching through the middle of that day, with the phone numbers for your platform, payment provider and developer written down before the morning starts.

How we run the Black Friday conversion sprint, including the freeze and watch plan, is set out on the service page.