Four offer decisions set your Black Friday margin: how deep the discount goes, where the free shipping threshold sits, whether you bundle or discount, and what happens after the purchase. Traffic and creative get the attention. These four decide whether the weekend is profitable.
Discount depth is a margin decision, not a competitive one
The instinct is to match the loudest brand in your category. That brand usually has a different cost base, a different average order value and a different repeat rate, which means matching their percentage is matching a number that means something else in their business.
Look at what the market actually does. The most common tier is 20% off, used by roughly 31% of stores, and only about 8% go beyond 40%. The brands going deepest are usually clearing inventory or buying first orders they intend to monetise on repeat, and both of those are strategies rather than reflexes.
Run the arithmetic before picking a number. At a 60% gross margin, a 20% discount costs you a third of your gross profit per order. To stay level you need order volume to rise by roughly 50%. A 40% discount at the same margin needs volume to roughly triple. Write those two numbers down and ask whether your traffic plan supports them.
The threshold does more work than the percentage
A free shipping threshold changes what goes in the basket rather than what each unit costs, which is why it is usually the more profitable lever.
Peak makes this sharper. Shopify’s average cart during BFCM 2025 was $114.70 against an $85 annual average. A threshold set for normal trading sits below what shoppers would have spent anyway, which means you are giving away shipping on orders that were already clearing it.
Review the threshold against your peak average order value rather than your annual one. We cover the calculation properly in our guide to setting a free shipping threshold, and the short version is that it belongs just above the dense cluster of your order value distribution, not above the mean.
The largest single result we have measured came from this lever. At a DTC supplements brand, one shipping threshold test produced two million dollars in profit. It changed nothing about how many people bought. It changed what they put in the basket, which is exactly the mechanism peak amplifies.
Bundle instead of discounting, where the catalogue allows
A bundle raises perceived value without cutting the unit price of anything. A discount cuts the unit price of everything.
The practical rule is to build the bundle from your actual order data rather than from what you want to move. Pull the products customers already buy together and package those. Brands usually start from slow stock, which produces a bundle nobody wanted assembled from things nobody was buying.
| Offer type | What it changes | Margin effect |
|---|---|---|
| Sitewide percentage off | Unit price on everything | Worst. Applies to orders that needed no incentive. |
| Threshold-based free shipping | Basket composition | Strong, if the threshold is set correctly |
| Bundle at a blended price | What gets bought together | Strong. No single item is discounted. |
| Gift with purchase above X | Basket composition | Good, cost is fixed and known |
| Tiered discount by spend | Basket composition | Good, rewards the behaviour you want |
| First-order-only discount | Acquisition cost | Depends entirely on repeat rate |
The post-purchase decision nobody plans
Everything above happens before payment. The offer that runs after payment is the only one that cannot cost you the order you just took, which makes it structurally different from everything else on the list.
Across the accounts we run, post-purchase upsells add around 10% to average order value, and that figure has held across a personal care brand under strict brand guidelines and a supplements brand with a completely different customer. On peak weekend volume that is a meaningful number attached to a change that carries almost no downside.
Set it up before the freeze rather than during the weekend.
What the peak numbers actually tell you
Two figures worth holding in mind when you plan.
Conversion runs far above baseline. Average Black Friday conversion reaches 5.86% against an annual ecommerce average near 2.8%. Intent is doing a great deal of that work, which means your offer does not need to compensate for a funnel problem it cannot fix in nine weeks.
And engagement is higher than usual. Bounce rate during BFCM falls to roughly 42% against a 55% ecommerce average. People arrive ready. The failure mode at peak is rarely persuasion. It is friction, cost surprises and a basket that stays smaller than it needed to be.
Decide these four before November
Discount depth, with the volume increase it requires written next to it. Threshold value, set against peak average order value rather than annual. Bundle composition, built from real co-purchase data. And the post-purchase offer, live and tested before the freeze.
Everything else is execution, and the execution list is covered in our BFCM CRO checklist, which handles the technical and structural fixes this post deliberately leaves alone.
Frequently asked questions
How deep should a Black Friday discount be?
The most common tier is 20% off, used by around 31% of stores, with only about 8% going beyond 40%. Depth is a margin decision: at a 60% gross margin a 20% discount needs roughly 50% more order volume just to hold gross profit level.
Should I raise my free shipping threshold for Black Friday?
Usually yes. Shopify’s BFCM average cart was $114.70 against $85 annually, so a threshold set for normal trading often sits below what shoppers would spend anyway.
Is bundling better than discounting?
For margin, usually. A bundle raises perceived value without cutting any single unit price, while a percentage discount applies to every order including the ones that needed no incentive.
When should these decisions be finalised?
Early November at the latest, so the offers are live and tested before the pre-peak code freeze rather than being configured during the weekend itself.
If you want the offer structure tested rather than guessed, that is what our ecommerce CRO programme does, and the sequencing sits inside the wider CRO Program.