Most BFCM strategy advice is written for stores that sell things once. Pick a discount, build the creative, watch the orders, count the revenue on Monday. For a subscription brand, that last step is where the advice stops being useful — because the number on Monday tells you almost nothing about whether the weekend worked.
Shopify merchants generated $14.6 billion over BFCM 2025, up 27% year over year. Black Friday 2026 falls on 27 November and Cyber Monday on 30 November. If you sell subscriptions, there's a third date that matters more than either: 27 December, when the subscribers you acquired at 40% off are charged full price for the first time.
The single decision that determines what happens on that date is whether your BFCM discount recurs.
A BFCM strategy is the plan governing your offers, timing, channels and creative across Black Friday Cyber Monday. For a subscription brand it has one component no other store needs: a decision about what happens to the discount after the first order.
That's the difference between selling a product at a discount and selling a price at a discount. A one-time store gives away margin on one transaction. A subscription store can give it away on every transaction for as long as the customer stays — which, if the structure is wrong, is a liability that grows the longer the subscriber sticks around.
Generally, no. The strongest structure is a discount that opens deep and steps down.
Three options, and what each does to the economics:
Protects margin, but you compete with 40% sitewide offers
The flat structure is the most common and the most expensive. It looks generous on Black Friday and it means a subscriber who stays two years costs you more the longer they stay. You acquired them at a loss and then locked in that loss as a standing arrangement.
A tiered discount fixes it without weakening the headline offer. The customer still sees 40% off at the moment of decision. Your margin returns at the second charge. In Loop, tiered discounts are configured on the selling plan — you set the BFCM rate on order one and the standard rate from order two, and after the season you edit one field rather than rebuilding anything.
Two things make the tiered version work rather than feel like a bait and switch: print both prices on the widget so the renewal rate is visible before purchase, and say plainly when the rate changes.
Two constraints should set the number, and neither is what your competitors are doing.
Your COGS at that depth. Not on order one alone — across the first three cycles, since a subscription spike is a standing order rather than a sale. The December cycle draws on inventory, January's cycle is that minus cancellations and failed payments plus winbacks, and February's is the surviving cohort. That third number is the one your margin has to clear.
What your existing cohorts already tell you. If you're running more than one selling plan or discount level today, you have a live experiment. Filter cohorts by selling plan and delivery frequency in subscriber analytics, and rank plans by retention at order three or four rather than conversion at checkout. Your best converter is rarely your best retainer.
Whatever depth you land on, set a minimum number of billing cycles on the BFCM selling plan. Loop's selling plan settings include this specifically to prevent misuse of subscription discounts, and it's the most direct control you have against someone subscribing purely to take the offer.
Raising the number every year is the default move and it degrades in two directions.
It trains customers to wait. Cherene Aubert, who has worked with over 100 ecommerce brands, describes this as offer innovation stagnation — the same sitewide sale at increasing depth, year over year, until the discount is expected rather than exciting, and customers who would have bought at full price simply hold out until November.
It punishes the people who already bought. Aubert's sharper point: if you're at 20% off and go to 25% during BFCM weekend, you've just angered every customer who bought at 20%, and they'll contact support asking for the difference. The reconversion you wanted comes at the cost of a support queue and a worse relationship with the customers most likely to buy again.
The alternative is a differentiated offer rather than a deeper one. A gift with purchase, a bundle tier, a limited-edition variant — something that gives an existing customer a new reason to buy without repricing what they already own.
Subscription bundles work particularly well here because the saving scales with box size rather than with the recurring price — a four-item box saves 10%, an eight-item box saves 30%, and the largest tier carries the best-value badge.
Earlier than the weekend, and not as one long sale.
MacCoy Merkley, CMO of Portland Leather Goods, treats BFCM as a 55-day season running from 1 November to 25 December — but with staggered, varied offers rather than one sitewide discount left running. His framing is a "promotional rolodex": a library of offers built in advance, each matched to a specific customer segment and a specific week.
The trap on the other side is real too. A month-long sitewide sale dilutes the brand, forces you to beat an inflated number next year, and pulls demand forward out of January.
For a subscription brand there's an additional constraint most calendars miss: every extra day of the offer is another day of renewal dates landing in the same December window. A sale running 1–30 November spreads first renewals across the whole of December. That's better for fulfilment and worse for measurement, and it's worth deciding deliberately rather than discovering in December.
The sequence that works, given subscribers are a channel most brands don't use:
Loop's campaign scheduling supports this directly — you set the audience and the exact start and end time, schedule it up to a year ahead, and the banner auto-hides on expiry so there's no stale offer to clean up. An exact end time is urgency you can prove.
This is the part a one-time store never has to solve. A subscriber renewing at your standard 15% subscribe-and-save rate on 28 November is looking at a 40% sitewide discount they can't have.
Give everyone the same deal and your subscribers cancel in order to take it. Give them nothing and they notice. The early-access window above is the cheapest answer — it costs no margin, it arrives before the public sale, and it frames the relationship as loyalty rather than pricing.
Four checkpoints, and only the last two mean anything.
If second-order retention for the BFCM cohort sits materially below a September or October control, the discount was too deep or the offer attracted the wrong buyer. If CAC payback never arrives, go one tier shallower next year.
Loop's cohort analysis guide covers building the view, and the subscription KPIs guide covers the surrounding metrics.
Should my Black Friday discount apply to every renewal?
Generally no. A discount held at the same rate forever means acquisition cost never pays back, because every renewal comes at reduced margin. A tiered structure — the BFCM rate on order one, your standard rate from order two — keeps the headline offer intact and gives payback a date.
How deep should a subscription BFCM discount be?
Deep enough to compete, shallow enough that your COGS clears across the first three cycles rather than just the first order. Use your existing cohort data: rank selling plans by retention at order three, not by checkout conversion.
Is it bad to increase my discount during BFCM weekend?
It creates a support problem. Customers who bought at the earlier rate will ask for the difference back. A differentiated offer — a gift with purchase, a bundle tier — gets you reconversion without repricing what people already bought.
When should a subscription brand start its BFCM campaign?
Most brands now run offers across November rather than a single weekend, but as staggered offers rather than one continuous sitewide sale. For subscription brands, note that a longer offer window spreads first renewals across more of December.
What should existing subscribers get during BFCM?
Early access, before the public sale, ideally on terms at least as good. If new customers get a better deal than your existing subscribers with no acknowledgement, some of those subscribers will cancel and resubscribe to take it.
When do I know if my BFCM worked?
Late January for second-order retention, and March for cohort LTV at order four against blended CAC. Revenue on 30 November is a sales number, not a retention one.