Black Friday does not create problems so much as reveal them. The broken redirect that cost you a trickle of traffic in July costs you real money in November, and the support question that used to arrive twice a day arrives forty times.
Everything below should be done well before the week itself. Changing storefront behaviour during peak traffic is how small problems become expensive ones.
Fix your broken URLs first
This is the least exciting item and the one with the clearest payback. Any 404 receiving traffic is a page you are already paying for and getting nothing from.
Pull your 404 report, map anything with meaningful traffic to a live page, and do it now rather than during the sale. Traffic multiplies during peak, and so does the cost of every dead end.
Decide your offer structure early
A flat site-wide percentage is the simplest option and usually the worst one — it discounts everything equally, including the products that would have sold anyway.
Volume tiers and bundles let you increase order value rather than just cutting margin. If you are going to discount, discount in a way that moves more units per order.
- Quantity breaks on your consumables and repeat purchases
- Bundles on complementary products that already sell together
- Keep the ladder short — two tiers, not five
Plan for selling out
You will sell out of something, and it will be the product you least expected. The question is whether that traffic is captured or lost.
Make sure back-in-stock capture is live before the sale, not added halfway through it. Demand recorded during peak is also the best possible input for January's reorder.
Prepare for the support spike
Question volume rises faster than order volume, because peak traffic includes a lot of people unfamiliar with your brand asking things your regulars already know.
Answer the top questions on the product page itself — shipping cutoffs, returns during the sale period, sizing — so the same question is not asked forty times. Then make the remaining ones easy to ask.
Freeze changes before the weekend
Set a cutoff after which nothing on the storefront changes: no theme edits, no new apps, no pricing experiments.
Every store that has a bad Black Friday for technical reasons made a change in the last seventy-two hours. The upside of a late tweak is small; the downside is your entire peak weekend.
The outcome
The spike lands on a store that is already working, instead of turning existing cracks into lost revenue.