What to Lock and When: The Peak Season Freeze Calendar

Every part of a peak trading setup has a date after which changing it costs more than leaving it alone. Landing pages, tracking, offer mechanics, campaign structure and site code each have their own freeze point, and they do not all fall on the same day.

With Black Friday on 27 November 2026, this post sets out what locks when, why each date sits where it does, and what actually breaks when a freeze is ignored.

Why freeze anything at all?

Because changes made during peak carry a cost that the same change does not carry in October. A structural campaign edit resets learning while impressions are at their most expensive. A site deploy risks breaking checkout on the highest traffic day of the year. A tracking change mid season splits your reporting into before and after.

The freeze is a way of front loading decisions into the weeks when mistakes are cheap. It is not caution for its own sake, it is moving risk to where it costs less.

What locks by mid October?

Campaign structure, tracking, and landing page requirements. The week commencing 12 October is six weeks out from Black Friday, and it is the point where the account should be in its final shape.

By that date, audiences should be built, budget scenarios modelled, and campaigns live and learning. Tracking should be verified end to end, with pixel and server side events confirmed firing and baselines recorded so you have something to compare November against. Landing page requirements should be locked and feed health checked. Peak creative production should already be in flight rather than being briefed.

The single decision to make in this window is which budget scenario you will actually run, and what triggers a move between scenarios. Agreeing that now is what stops the mid November argument about whether performance is good.

What locks by late October?

Landing page tests and offer mechanics. Four weeks out is the week commencing 26 October.

Landing page tests should be concluding, so that only proven pages carry peak traffic. Campaign specific pages beat default collection pages when competition is at its highest, but only if they have been validated first, and a test that is still running on 20 November is not a test, it is a risk.

Offer mechanics should be settled in the same window: discount depth modelled against contribution margin, stacking rules decided, bundle and threshold mechanics tested. Once an offer is in an email subject line it is a commitment, so the margin conversation has to happen before that point rather than in January.

What locks two weeks out?

Site code, creative, and the promotion sequence. Two weeks before Black Friday is the week commencing 9 November.

Freeze non essential site changes and verify load speed, mobile experience and checkout under realistic conditions. Winning creative concepts should be locked, with contingency creative ready and fatigue replacements queued rather than commissioned. The promotion sequence should be final: early access, main event, extensions, with the rules for changing it mid flight agreed in advance.

The measurement job in this window is escalation rules. Daily reporting live, and a named person who decides what happens when a number moves, with an agreed threshold for acting. Peak rewards decisions made in hours, and hours are not enough time to agree who decides.

What stays unfrozen?

Budget within the pacing rule, creative rotation inside existing ad sets, and promotional messaging through catalogue overlays.

That last one matters more than it sounds. Switching offers on product ads through catalogue overlays changes the message without resetting learning, which is what makes the phase to phase pivots of the season possible. Setting the overlays up in October is what buys you that flexibility in November.

Merchandising also stays live, and should. Monitor conversion rate and stock in real time and merchandise to what is actually selling rather than to what the plan said would sell.

What breaks when a freeze is missed?

Three failure modes, in rough order of frequency.

The most common is a structural campaign change made in mid November that resets learning during Cyber 5, so the account spends its peak budget relearning. The second is a site deploy in peak week that breaks something small and expensive, usually in checkout or on mobile, discovered hours later. The third is quieter and worse: a tracking change mid season that makes November incomparable to October, so nobody can tell afterwards what actually worked.

None of these show up as a single dramatic failure. They show up as a quarter that underperformed for reasons the team cannot reconstruct in January.

Where these dates come from

These freeze points are drawn from the Peak Trading Calendar, which runs seven phases from six weeks before Black Friday through to January. Each phase names the media, creative, offer, website, measurement and decision for that week, and it is the section most teams keep open from October onward.

It sits in the Q4 Profit Playbook, free, alongside the readiness scorecard that scores seven areas red, amber or green before the season scores them for you. The playbook is built for founder led DTC brands at $5M to $30M on Shopify.

The wider pre peak audit, covering feed health, site speed, creative volume and contingencies, is in peak season readiness, and Webtopia runs this as one system for founder led brands as an ecommerce marketing agency.

Frequently asked questions

When should I stop making changes to my ad account before Black Friday?

Structural changes should be finished by the week commencing 12 October, six weeks before Black Friday 2026, so campaigns have time to exit the learning phase before budgets step up.

When should I freeze my website before peak?

Two weeks out, the week commencing 9 November. Freeze non essential changes and verify load, mobile and checkout before traffic arrives rather than during it.

Can I still change my offer during peak?

The message can change through catalogue overlays without resetting learning. The underlying economics, meaning discount depth and stacking rules, should be settled by late October.

What should be verified before the tracking freeze?

Pixel and server side events firing end to end, dashboards built, and baselines recorded so November has something to be measured against.

Is a code freeze really necessary for a small brand?

The smaller the team, the more it matters, because there is less capacity to diagnose a broken checkout at 9pm on Black Friday. The freeze is what protects the day you cannot repeat.