Black Friday 2026 lands on 27 November.
If you're reading this in October and you haven't decided your offer yet, you've already given away the part that mattered.
Not the sale itself. The sale will be fine. Everyone's sale is fine. Shopify's own figures put the weekend at $14.6 billion globally in 2025, up 27% on the year before, and a rising tide carries most boats through four days.
The bit you've lost is the margin, the list health and the quality of the customers you pick up. Those get decided in September and October, by people who plan, and they get surrendered in November by people who don't.
Most brands treat Black Friday as a week of emails. It is a three month operation where the emails are the last and easiest part.
So here it is in order.
First, decide whether you should be doing it at all
Worth saying before the rest, because nobody selling you Black Friday services will say it.
Sell a considered, high ticket product with a long purchase cycle and Black Friday may be actively bad for you. You'll pull forward purchases that were coming anyway in January at full price, train your best customers to wait for November, and pick up a cohort of discount hunters who never buy again.
A brand doing eighty grand a month on a £400 AOV with a six month consideration window is not in the same business as a brand shifting £30 candles. Running the same November playbook in both is daft, and it happens constantly because the calendar says it's November.
The test is simple. Work out what proportion of your November revenue would have arrived in the following ninety days without a sale. If you genuinely don't know, that's your answer for this year: run a smaller offer, measure it properly, and decide with data next time.
Plenty of well known brands now sit the weekend out entirely and grow anyway. That's an option available to you too.
Run that test, get a yes, and the rest of this applies.
September: you are not selling, you are collecting
The single biggest lever on your Black Friday revenue is how many of the right people are on your list when the offer goes out. Not the offer itself.
Work it backwards. If your list converts at roughly 2% over the BFCM weekend and your AOV is £70, every extra thousand engaged subscribers is worth about £1,400. Getting a thousand people on your list in September costs a fraction of what acquiring them costs in the third week of November, when every DTC brand in your category is bidding against you.
Spend September acquiring subscribers, not customers.
Practically:
Turn your popup from a 10% off offer into something that captures more people. Quizzes, waitlists, early access lists. Early access converts better in this window than a discount does, because you're offering something that doesn't exist yet and costs you nothing.
Run a dedicated Black Friday waitlist and promote it hard from mid September. It gives you a segment of people who have self identified as ready to buy at a discount, which is the most valuable segment you'll have in November.
Push zero party data collection while you've got attention. Category preference, size, gift or self purchase. You'll use every field of it when you segment the sale.
Run paid social to a list building offer rather than a product page for a few weeks. CPMs in September are a fraction of what they'll be in November.
Now for the job nobody does until it's too late to help.
October: fix the plumbing before you need it
You're about to increase your sending volume by a factor of three or four in a fortnight. If your domain reputation is shaky, November is when you find out, and by then there's nothing to be done.
Google and Yahoo have required authentication, one click unsubscribe and a spam complaint rate under 0.3% for anyone sending 5,000 or more a day since February 2024. Since November 2025 Google has escalated persistent failures from temporary deferrals to permanent rejections. Microsoft went further in May 2025, rejecting outright on Outlook, Hotmail and Live.
That's not a warning about the future. That's the current state.
Do your sunsetting in October, not January.
It feels wrong to shrink the list right before your biggest sale. Do it anyway. The people you'd be cutting haven't opened anything in six or twelve months. They will not buy on Black Friday. What they will do is fail to engage in volume, right at the moment your complaint rate is under the most scrutiny, and drag your inbox placement down for everyone else on the list.
Cutting 20% of your list in October and reaching 90% of the remainder beats keeping everyone and reaching 60%.
Other October jobs. Warm any new sending domain now, because it takes weeks and cannot be rushed. Check your Postmaster Tools numbers and give yourself a month to fix anything ugly. Test your templates on mobile properly. Across BFCM datasets, roughly two thirds of orders come from phones, and a higher share of the traffic does.
The three numbers you need before anyone opens a calendar
Most Black Friday planning meetings start with creative. They should start with arithmetic.
Your contribution margin per order at full price. Revenue minus COGS, minus shipping, minus payment fees, minus your actual returns cost. Not gross margin. The number that's left when the order is genuinely finished.
Your discount floor. The depth at which that contribution figure hits zero. Below it, every additional order makes you poorer. Brands discover their floor is 34% having run 40% off the previous November more often than anyone wants to admit.
Your incremental cost per order over the weekend. Paid CPMs spike, support tickets multiply, warehouse overtime is real money. A weekend order costs more to fulfil than a Tuesday one.
Have those three and the offer discussion takes twenty minutes. Without them you're picking a number because a competitor picked a number, which is how the whole market ended up racing each other to the bottom while telling themselves it was strategy.
One more thing your finance director will ask that most agencies can't answer: what's the returns rate on discounted orders versus full price ones. It's usually higher. If you're running 30% off on a category that already returns at 25%, the real margin position is worse than the spreadsheet says.
Deciding the offer
Most brands go wrong here, and rarely by being too cautious.
Adobe Analytics put peak 2025 discounting by category at around 31% off for electronics, 28% for toys, 25% for apparel and 22% for computers. Those are the deepest points of the season across the whole market, not the average.
So when you're sat there wondering whether 40% off is enough to compete, look at that list again. You are almost certainly not being outdiscounted as badly as you think you are. The pressure you're feeling is coming from the two loudest competitors in your feed, not from the market.
Before you pick a number, work out three figures.
Your floor. The discount at which contribution margin goes to zero after COGS, shipping, payment fees and returns. Not gross margin. Contribution.
Your ceiling. The deepest discount you can run and still make the weekend worth doing.
Your headline. What goes on the site and in the subject lines.
The headline should sit closer to the floor than you'd like, because the depth comes later and it comes selectively. If you lead with your deepest number, you've got nothing left to escalate with on Cyber Monday and nothing to give the people who need a bigger push.
Gating the depth by who they are
This is the difference between a mature programme and a course graduate blasting 30% off to everyone with a pulse.
Your list is not one audience. On Black Friday it's at least five, and they need different offers.
Your best customers don't need a discount to buy, they need early access. Give them the sale 24 or 48 hours before everyone else at the standard rate. You get revenue pulled forward into a quieter window, better inbox placement going into Friday because your first sends land on your most engaged people, and a loyalty signal that costs you nothing extra.
Recent buyers, people who bought in the last 30 days, need care rather than aggression. They may feel mugged off if last week's full price purchase is now 30% cheaper. Consider a credit or a gift rather than hammering them with the sale.
Engaged non buyers, subscribers who open and click and have never purchased, are your deepest discount audience. They've told you repeatedly that they want the product and that your normal price is the blocker. This is where your ceiling number goes, quietly, in a segmented send. Not on the homepage.
Lapsed customers get the winback angle rather than the sale angle. Same discount, different message.
Everyone else gets the headline offer.
Five segments. Five different depths and framings. Same sale.
The sending calendar
Volume goes up. This frightens people, and the fear costs them more than the sending does.
A workable shape for the week:
Early access to your top segment on the Wednesday or Thursday. Launch to the full list on Friday morning, then a second send Friday evening to non openers with a different subject line. Saturday and Sunday are quieter, one send each, ideally with a different angle rather than the same offer shouted louder. Cyber Monday gets two, one in the morning and a genuine last chance in the evening.
The last chance email on Monday evening is consistently one of the highest revenue sends of the entire year. Do not skip it because you feel like you've emailed too much.
Exclude purchasers from every subsequent send in real time. Nothing generates complaints faster than getting your fourth "don't miss out" after you've already ordered. A flow filter or a suppression segment updating live is not optional here.
A large and growing share of BFCM revenue now lands before Black Friday itself, which is why the early sends matter more than they used to. The weekend is no longer the whole event. It's the peak of a month, and if you're only planning four days you're planning the smaller half of it.
Cash, not just margin
A point that gets missed because email people think in revenue and business owners think in cash.
Black Friday converts inventory into cash quickly, which for a lot of brands is the actual reason to do it. Clearing slow stock at 40% to fund January's buy is a perfectly sound decision even if the margin looks poor in isolation.
That's a different objective from acquiring customers, and it needs a different offer. Stock clearance should be deep, specific and quiet. Segmented sends, not the homepage. Customer acquisition should be shallower and public.
Decide which one you're doing before you build the campaign. Brands that try to do both with one offer usually clear the wrong stock at the wrong depth and call it a record weekend because the top line went up.
The week after, which decides whether it was worth it
Black Friday hands you a large cohort of new customers acquired at a discount. Discount acquired cohorts repeat worse than full price ones. That is why your November cohort will look great this month and disappointing next April.
So the plan does not end on Cyber Monday.
The people who bought on Friday need a post purchase sequence built for them specifically, not the generic one. They bought at 30% off. They do not yet know why your product is worth full price. That's what December's job is.
Get the product education in. Get them using the thing. Get the second purchase happening in January at full price, because a discount buyer who repeats at full price becomes a normal customer, and one who doesn't was just a margin donation.
Segment your December sending so BFCM buyers get a different treatment from everyone else. It's a fortnight of work and it changes the value of the whole quarter.
What to actually do next
If it's September, you're building the list and fixing the plumbing.
If it's October, you're sunsetting, warming and locking the offer.
If it's November, you're segmenting and scheduling, and the outcome is already mostly determined.
If it's December, you're deciding whether the customers you just bought are worth keeping.
Whichever month you are reading this in, the next job is the same: work out your floor before anyone asks you for a number. Everything else on this list gets easier once you have it, and the brands who find out in February what the weekend cost them are the ones who never did.