Get your Shopify store holiday-ready

Black Friday 2026 is 27 November, Cyber Monday 30 November. The useful version of this checklist is ordered by when things have to happen, because half of what people do in November should have been done in September, and the other half should not be done in November at all.

Most of it applies to any ecommerce store. The few Shopify-specific items say so.

Now — the things with a lead time

Fix how fast the store loads, on a phone, on bad network

Not on your laptop on office wifi. Open your own product page on mobile data and count. Every second before the page is usable costs orders, and it costs them worst on the traffic you will be paying for.

The usual culprits are ordinary: an enormous hero image, a carousel nobody scrolls, a font loading from a third domain, and apps you stopped using but never uninstalled. The last one is free to fix and often the biggest.

Audit your installed apps, properly

Open the apps list and justify each one out loud. Anything you cannot justify is loading scripts on every page view for no reason.

Uninstall rather than disable. A disabled app can still leave code in your theme, and holiday traffic is the wrong time to find out which.

Count your stock honestly, then decide the sold-out story

Reconcile what the system thinks you have against what is on the shelf. Then decide in advance what happens when a hero product sells out — continue selling on backorder with a clear despatch date, or go out of stock cleanly with a notify-me. Both are fine. Deciding at 9pm on Black Friday is not.

Publish your shipping cutoffs before anyone asks

Get the carrier deadlines for the destinations you actually ship to, then put them somewhere a shopper sees before checkout, not only in a policy page. “Order by 18 December for delivery before Christmas” on the product page converts, and it also prevents the support load that follows from not saying it.

Extend the returns window, and say so

Gifts bought in November are opened in December. A returns window that expires before the recipient has seen the thing generates disputes rather than returns. Most stores extend to mid-January; the important part is publishing it, since an unpublished generous policy converts nobody.

Next — pricing and offers

Decide the discount structure, not just the discount

“20% off everything” is the default because it is easy to decide, not because it is best. It discounts the customers who were going to buy anyway at exactly the same rate as the ones who needed persuading.

Worth weighing against it: spend thresholds (free shipping over £50), bundles for cross-sell, and quantity breaks where buying more of the same thing is plausible — consumables, refills, anything restocked. Each one asks the shopper to do something specific in exchange for the discount, which is the entire difference.

🔴 Whatever you pick, set it up and test it in October. Installing a pricing mechanism you have never run, in the week of the sale, is how stores discover at the worst possible moment that two discounts stack and a line item is below cost. If you are going to try quantity breaks — ours, TierUp, is free up to 10 rules, and so are several alternatives — do it now, with a test order, not on the day.

Work out what the deepest discount actually costs

For each offer, calculate the margin on a realistic basket, not on one unit at full price. Include the shipping you are giving away and the payment fees. Stores routinely discover in January that their best day was their least profitable.

Decide stacking on purpose

Sitewide sale, plus a welcome code, plus a loyalty discount, plus free shipping is four decisions that were each made alone. Write down what may combine with what, then configure it that way, then put a worst-case basket through checkout and look at the total.

Then — the operational side

Test checkout on a real phone

Complete an actual order. Check accelerated checkout still appears — Shop Pay, Apple Pay — with a discount applied, because some discount mechanisms remove it and the symptom is silent. Why that happens is worth ten minutes if you have ever seen it.

Confirm tracking survives a discounted order

Place a test order with the discount applied and check it appears correctly in your analytics and ad platform. If revenue tracking breaks only on discounted orders, you will discover it while trying to read your Black Friday results.

Write the customer-service answers in advance

Where is my order, can I change my address, does the discount apply to this, when will it arrive. Draft them now while you are calm. Volume triples and your patience does not.

Finally — stop changing things

Freeze the theme about a week out

Duplicate your live theme as a backup and stop editing. Most BFCM disasters are self-inflicted: a last-minute tweak, on a Thursday evening, by someone who has been working since six.

Know how to roll back, and have tried it

Know which theme to revert to and how long it takes. Knowing you have a backup is not the same as having restored one.

Leave the ad account alone

Campaigns need learning time. Rebuilding them the week before is throwing away the data that makes them work, at the moment it is worth most.

The thing most people skip

Decide now what you will measure in December. Not revenue — revenue always looks good on Black Friday. Margin after discounts and shipping, new customers versus returning, and how many of November’s buyers come back in February. That last number is the one that tells you whether you ran a sale or bought a crowd.