A BFCM Discount Calendar for Shopify Stores (2026)

Most BFCM calendars are lists of dates with tasks attached. The problem with that format is that it does not tell you what depends on what, so teams work on the visible items and discover the structural decisions late.

This one is ordered by dependency. Each week's work only requires decisions made in earlier weeks, which is what stops a team from building emails against an offer that changes twice.

The anchors: Black Friday falls on Friday 27 November 2026, Cyber Monday on 30 November, and Cyber Week runs to 3 December.

Early October: decide the offer

Nothing else can start until this is settled, and it is the week most commonly skipped.

Model two or three candidate offer structures against contribution margin on real product-level data — not gross margin. Shipping subsidies, payment fees and elevated return rates all move against you during BFCM, and an offer that looks comfortable at gross margin can land near break-even once they are included.

Decide what each offer is for. Clearing specific stock, raising average order value, acquiring customers, or defending against a competitor. These point at different mechanics, and an offer asked to do all four does none well.

Choose the mechanic, not just the number. A flat sitewide percentage is the easiest to build and the most expensive, because it discounts your best sellers at the same rate as the stock you actually wanted to move.

Mid October: lock the mechanics and build

Lock the offer. Late changes to discount logic are the single largest source of checkout bugs, and every week of delay here compresses the testing window at the other end.

Build in staging. Every offer, configured as it will actually run.

Decide combination behaviour for every pair of offers that could be simultaneously active — including discount codes still circulating from earlier campaigns that nobody has expired.

Start the landing pages. Building them now gives you real URLs to test the discount logic against, and lets them be indexed before the period rather than launching on the busiest day of the year as pages Google has never seen.

Late October: test and audit

Walk every customer path. Every offer combination a real shopper could reach, on a real device, including guest checkout and at least one accelerated checkout method completed end to end.

Audit apps. Anything not currently earning its place adds page weight and a failure mode. Remove it now — not in November, when there is no time to observe the consequences.

Check inventory against the offers. Gifts, bundle components and buy-X-get-Y items are drawn down by demand for something else, and they are the stock nobody is watching. Set alerts and decide in advance what each offer does when its component runs out.

Early November: warm the list

Build the audience before you need it. Email remains the highest-return BFCM channel, and its return depends on a list that has heard from you recently. The offer should not be the first message a subscriber has received in three months.

Send early access to your best customers. This shifts some demand off the peak days, which helps fulfilment, and it rewards the people most likely to buy anyway with something better than a public discount.

Finalise creative — ads, emails, landing page copy. Every headline should repeat its offer's promise word for word, because shoppers scan for confirmation they are in the right place and a paraphrase costs you an expensive click.

Mid November: freeze

The freeze date is the most valuable line in this calendar. After it, no changes ship to checkout, cart or discount logic.

This is not caution for its own sake. A change needs live traffic afterwards to reveal its problems, and a change made two days before Black Friday has no observation window at all. The reason a late change feels safe is that you have not yet seen it fail.

Creative and content can still change. Logic cannot.

Agree the rollback plan. Know how to disable each offer quickly, and who has access. During peak, the correct response to a misbehaving discount is to turn it off, not to debug it live.

Cyber Week: watch, do not build

27 November to 3 December. The work now is observation.

Watch order-level margin, not revenue. Revenue looks good during BFCM almost regardless. Margin per order is what reveals a discount applying that should not be.

Watch the proportion of orders receiving each offer. A rate that moves sharply overnight means something changed and nobody was told.

Have a person placing test orders, twice a day. The failures that hurt most are silent ones no alert is configured for.

Write down what breaks, as it happens. Peak generates a dozen small discoveries everyone is certain they will remember and nobody does. This note is the entire basis of next year's preparation.

The week after Cyber Monday

The most under-planned week in the calendar, and the one with the best return available.

You have just acquired a large number of first-time buyers, at the highest acquisition cost of the year. Most merchants exhale and let them go quiet, which means the period bought revenue rather than customers.

Send a genuine post-purchase sequence. Not a receipt. Delivery expectations, how to use the product, what to do if something is wrong. BFCM buyers frequently know nothing about your brand beyond one discounted transaction.

Give them a reason to come back that is not a discount. Training buyers that you are only worth buying from in November is how BFCM revenue stops being incremental.

Handle returns well. Return rates are elevated on discounted purchases. A good return experience is one of the few things that converts a discount-driven buyer into a repeat one.

What to do if you are starting late

Most merchants read a calendar like this in November, not October. Skipping straight to the current week is the wrong response, because the earlier weeks contain the decisions everything else depends on.

Compress instead. The minimum viable version, in priority order:

One day: model the offer against contribution margin. Even a rough calculation on your top twenty products will tell you whether the discount you were about to run is survivable. This is the single highest-value hour available to a late starter, because it is the decision that cannot be corrected afterwards.

One day: build and test the offer, including combinations. Do not skip the combination testing, even under time pressure. It is the failure that costs the most and it is invisible until a customer finds it.

Half a day: expire old codes and audit dead apps. Fast, and removes two of the most common peak failures.

Then stop adding. A late starter's instinct is to compensate with more — more offers, more emails, more landing pages. The opposite is correct. One well-tested offer beats three that nobody had time to walk through, and complexity is what breaks under load.

What to genuinely drop if time is short: additional offer mechanics beyond the first, extensive landing page work, and anything requiring development. What never to drop: combination testing, an accelerated checkout test, and inventory alerts on gift and bundle components.

Mid December onward: account for it honestly

Measure incrementality, not gross revenue. A share of BFCM sales would have happened anyway at full price. If you cannot estimate that share, you cannot say whether the promotion made money — and a rough estimate from last year's baseline beats the implicit assumption of zero.

Review the note you kept. Decide what to repeat and what to drop while it is fresh.

Expire everything. Codes, automatic discounts, banner messaging. Sale messaging left visible in January is a common and damaging error.

For offer structure and margin modelling, see the BFCM 2026 series. Checkout verification is covered in checkout and payments, and the discount mechanics in promotions and offers.