Everything in this series is survivable at normal trading volume because there is slack in the system. A ten-day return-to-resale time is untidy in June. A back-in-stock alert sent to four hundred people for twelve units is a small waste in August. A variant that quietly goes to zero while advertising continues costs a manageable amount in a normal week.
In Q4 the same failures cost several times more, arrive simultaneously, and land on a team with no spare hours. Nothing new goes wrong. Everything that was already going wrong stops being absorbable.
This guide is a timeline, not a theory. The dates are for 2026.
The dates
- Black Friday — Friday 27 November 2026
- Cyber Monday — Monday 30 November 2026
- Christmas Day — Friday 25 December 2026
Carrier last-posting dates are published in the autumn. Do not guess them, and do not put a last-order date on your site until you have the real ones.
When is your last useful order date?
The most urgent calculation in this guide, and one almost nobody runs.
Stock needs to be live on the site roughly a week before Black Friday — call it Friday 20 November — so it can be photographed, merchandised, and included in campaigns built in advance.
Work backwards using your true lead time from Guide 08: purchase order raised, to stock sellable on site, including receiving. Not the supplier's quoted figure.
| Your true lead time | Order by |
|---|---|
| 4 weeks | 23 October |
| 6 weeks | 9 October |
| 8 weeks | 25 September |
| 10 weeks | 11 September |
| 12 weeks | 28 August |
If you are reading this in early September on a twelve-week lead time, that window has closed. Plan Q4 with the stock you have, and do not place an order in October believing it will arrive in time — it will land in December and become January's markdown.
If you are on ten weeks, you have days rather than weeks.
Add a week to every figure above if your supplier has ever been late. Peak is when suppliers are most stretched, and lead time variance widens exactly when you can least absorb it.
What to order: depth, not breadth
The instinct at peak is to widen the range. It is usually wrong.
New lines have no sales history, so you cannot forecast them, so you will guess, and you will guess at the moment when a wrong guess is most expensive. Meanwhile the products you know sell will go out of stock in their best sizes.
Order depth on proven lines, weighted to the variants that actually sell. Guide 03 covers the arithmetic; at peak, the variant point becomes the whole game. Ordering forty units evenly across three sizes when the middle size is sixty per cent of demand guarantees you sell out of the size everyone wants and discount the rest in January.
Apply your buffer generously on A-tier lines and accept thinner cover on the tail. Peak is the one time of year when the cost of a stockout on a bestseller clearly exceeds the cost of carrying it.
The freeze
From 1 November, stop changing things.
- No theme changes. A conversion bug introduced on 20 November is discovered on 28 November, by which point it has cost more than the change was worth.
- No new app installs. Every app writes to your storefront. Peak is not when to find out how.
- No SKU restructuring. Guide 05's work is a January job, not a November one.
- No new email platform, no migrations, no replatforming.
Anything that must ship should ship by the end of October, with a fortnight of live trading behind it before volume arrives.
The exception is stock. Stock keeps moving. Everything else holds still.
What changes about advertising at peak?
Two things happen at once: your spend rises and the cost of each impression rises with it, because every other advertiser is bidding for the same attention.
That changes the arithmetic in Guide 04. Spend running against an unavailable variant is wasteful in July. In late November it is wasteful at a substantially higher rate per hour, during the period when your total spend is at its annual maximum. The same failure, on the same variant, costs a multiple of what it costs the rest of the year.
Three things to do before November:
- Tag every product currently in paid advertising with
paid-ads, and build the Flow workflow that alerts you when one of those variants hits zero. The Flow recipes companion has the configuration. It will not pause the spend, but it means someone knows within minutes rather than at the weekly review. - Separate your top-spend products into their own campaigns or ad sets so you can pause selectively. Doing this in October costs nothing; doing it on Black Friday is not possible.
- Check your feed settings now — availability updates, strict availability, and inventory tracking enabled on every variant. Guide 04 covers why each one matters.
And the creative problem gets worse at peak, because campaign creative is built weeks in advance and the hero product is the one most likely to sell out. Have a replacement creative ready for each of your top three products before you need it.
What changes about messaging at peak?
Back-in-stock volumes rise sharply in Q4, and the batching problem from Guide 06 scales with them. Alerting six hundred people to a restock of twenty units on the first weekend of December does not produce twenty grateful customers; it produces twenty sales you would have made anyway and five hundred and eighty people who now distrust your emails during the period when your email list is your most valuable asset.
Set batch sizes before the season, not during it.
Add an inventory condition to every flow that features a product, if you have not already. At peak the interval between an item selling out and a scheduled email promoting it shrinks to hours.
The January returns wave
The part that is planned for least and lands hardest.
Most stores extend their returns window over Christmas — returns accepted until the end of January, so gifts bought in early December can be returned by whoever received them. This is a sensible commercial decision and it is worth understanding what it commits you to.
It is a goodwill extension, not a legal requirement. The statutory cancellation window under the Consumer Contracts Regulations runs 14 days from delivery, so an item delivered on 3 December carries statutory rights that expire in mid-December. Everything beyond that is your policy, and you can set different terms on the extended portion — this is not legal advice, and it is worth confirming your wording. What you cannot do is apply restocking fees within the statutory window, which Guide 09 covers.
The operational consequence is what matters. A large share of December's revenue reverses in January, arriving as a concentrated wave, at the point in the year when you have the least staffing and the most pressure to clear stock. Your January cash planning is probably built on December's gross figures.
Three preparations:
- Estimate the reversal. Take last January's return volume as a proportion of December's sales. If you have not tracked it, that is this year's job.
- Staff for the wave, not the average. It is a fortnight of concentrated work, not a level increase.
- Prioritise the queue by stock position. Guide 09's free fix matters most here: returned units of variants currently at zero go first, because January is when you are trying to trade out of overstock and every recovered unit is one you do not have to buy.
After peak
Do this in the first week of January, while the evidence is fresh.
- Record actual lead times for every Q4 order. This is your most accurate lead time data of the year, because it was collected under stress.
- Pull sell-through by variant and note which sizes and colourways ran out first. That is your buying brief for next year, and it is worth more than any forecast.
- Note every stockout. Which variant, which days, what was running against it.
- Measure return-to-resale time during the January wave. It is your worst-case figure and therefore the honest one.
- Then unfreeze. January and February are when the SKU work in Guide 05 and the app audit in Guide 02 should happen.
The App Ceiling
This section appears in every one of The Operational Guides. It describes where the advice above stops working.
Everything above holds to roughly £1m–£1.5m in annual revenue. Past that point the constraint changes, and it changes in a way that adding further apps does not address.
Apps are built to be sold to many stores. That is what makes them affordable, and it is also what limits them: an app can only act on the data inside its own boundary. Your forecasting tool does not know what your ad platform is spending. Your ad platform does not know what is out of stock. Your back-in-stock tool does not know what your supplier lead times are. Each app is correct within its own scope and blind outside it.
Below roughly £1m, a person bridges those gaps. Someone looks at the stock report, notices a line is running low, and adjusts. The bridging is invisible because it is absorbed into the founder's day.
Above roughly £1m, three things happen at once. SKU count rises, so there is more to bridge. Order volume rises, so the consequence of missing something rises with it. And the founder's time is now spent on growth rather than operations, so the bridging stops happening reliably. Nothing breaks visibly. Revenue keeps climbing. What changes is that a percentage of it begins to leak in places nobody is looking — advertising spend running against unavailable variants, stock accumulating in the wrong location, returns sitting unprocessed for a week and a half while the item they contain is out of stock and being advertised.
That is not an app problem, and no app solves it, because the solution has to sit between systems rather than inside one. It requires something built for your specific stack, your specific SKU structure, and your specific supplier terms.
Peak is the honest test of where you sit against that ceiling. The manual bridging that works all year is exactly what fails in November, because it depends on a person having spare attention and in November nobody does. If your Q4 was rescued by someone working through the last weekend of the month, the system did not work — a person did, and that is not a system you can grow on.
Where to start
Work out your last useful order date for each supplier today. Not this week. If you are on a long lead time, the decision is already being made whether you make it or not.
Then put the freeze date in the calendar and tell whoever needs to know.