Black Friday is won in September. By November you are only executing decisions you already made.
Every year peak season arrives faster than the plan does, and you end up buying the most expensive traffic of the year with untested creative.
4 min read
Peak trading concentrates a year's competitive intensity into a few weeks. Reported CPM increases during the season range widely by platform and category — some datasets show a modest single-digit rise, others show 25% to 35% above the annual mean, and Black Friday week itself frequently runs at two to three times the annual baseline.
The number matters less than what it implies: the traffic you buy in late November is the most expensive traffic you will buy all year, so it is the worst possible moment to be testing creative, warming up cold audiences or discovering that your tracking broke.
Everything below is about moving work earlier.
What actually happens to costs
2–3×
Typical Black Friday week CPM against the annual baseline in reported ecommerce datasets
Benly Meta ads seasonal campaign analysis
8–10 weeks
Commonly recommended lead time for building audiences and testing creative before costs climb
Triple Whale BFCM advertising guide
Two things drive it. Every advertiser raises budgets at once, which raises auction prices for everybody. And consumer attention is fragmented across more offers, which lowers response rates at the same moment costs rise.
The brands that do well are not the ones who bid hardest. They are the ones who arrived with a warm audience, proven creative and working tracking, so their money buys conversion rather than discovery.
A working timeline
Ten to eight weeks out
- Run the tracking health audit properly. A break discovered in November is a catastrophe; the same break found in September is a Tuesday.
- Confirm inventory and margin by product, so you know what you can afford to promote and what you must not.
- Decide the offer. Not the creative — the offer. What it is, who gets it, what it costs you at the volume you expect.
Eight to five weeks out
- Test creative concepts at normal-season prices. This is the single highest-value thing in the whole plan: learning what works while learning is cheap.
- Build the audiences. Every visitor, add-to-cart, engager and email subscriber you gather now is someone you do not have to buy in November.
- Grow the owned list deliberately, because email, WhatsApp and SMS are the only channels whose cost does not double in December.
Four to two weeks out
- Scale winning creative gradually, using proper budget increases so nothing is in a learning phase during peak.
- Build and test the full messaging sequence end to end, including the send that goes out at 6am on the day.
- Check site performance under load, and complete a real purchase on a phone.
- Set the floor: the minimum return at which you will keep spending, decided calmly now rather than at 11pm on the day.
Peak week
- Change as little as possible. Every edit resets learning at the most expensive moment of the year.
- Watch delivery and inventory, not ROAS hour by hour.
- Lean on owned channels, where the marginal cost of another message is close to zero.
The four expensive mistakes
- Launching new creative in peak week. You are paying peak prices to find out whether an idea works, and the answer arrives too late to use.
- Big budget jumps on the day. A 5× increase resets learning and buys exploration at the year's highest prices.
- Discounting deeper than the margin supports, because a competitor did. Their cost base is not yours.
- Ignoring existing customers. They are the cheapest revenue available and the least contested, and every hour spent on cold prospecting during peak is expensive by definition.
The week after, which nobody plans
Two things happen in early December that are worth preparing for.
Costs fall as advertisers exhaust their budgets, which makes the following fortnight some of the cheapest efficient traffic of the quarter. Most brands are too tired to use it.
And you now hold a large cohort of first-time, discount-acquired customers who will repeat at a lower rate than your usual customers unless something is done about it. The post-purchase sequence that runs in December decides how much of peak season revenue turns into a customer base rather than a spike.
Common questions
How much do ad costs rise during Black Friday?
Reported figures vary widely by platform and category — from single-digit season-wide rises to 25% to 35% above the annual mean, with Black Friday week itself often running at two to three times the annual baseline.
When should I start preparing for peak season?
Eight to ten weeks out. Creative testing and audience building must happen while traffic is cheap, because doing them in November means paying peak prices for discovery.
Should I change campaigns during peak week?
As little as possible. Every material edit resets learning at the most expensive moment of the year, so peak week should be executing decisions made earlier rather than making new ones.
What is the biggest peak season mistake?
Launching untested creative in peak week. You are paying the year's highest prices to learn something you could have learned in September for a fraction of the cost.
What should I do the week after peak?
Two things: use the cheaper traffic while other advertisers are exhausted, and run a real post-purchase sequence for the large cohort of discount-acquired first-time customers you just gained.
How Glimmio handles this
Glimmio monitors tracking health continuously, which is what turns the pre-peak audit from a task somebody remembers to do into a signal that arrives when something breaks.
During peak, recommendations still wait for approval and new campaigns are still created paused — which matters most in the week when an accidental change is the most expensive it will ever be.
- Manual approval by default — nothing runs unattended
- New campaigns and ads are always created paused
- 7-day recovery window on eligible changes
- 48 permissions across 13 roles, scoped per client account
Go deeper on this
The product pages and setup guides that cover what this article describes.
Searches this answers
- black friday ad strategy ecommerce
- bfcm cpm increase planning
- q4 paid media plan d2c
- peak season advertising costs
- holiday ad campaign preparation shopify
Read next
How to Scale a Meta Budget Without Breaking It
You found something that works, increased the budget, and it immediately stopped working.
ReadThe 20-Minute Tracking Health Audit for Shopify Stores
Performance dropped three weeks ago and you still do not know whether it was the market, the creative, or a tracking break.
ReadWhat Discounting Actually Costs You
Revenue is holding up but profit is not, and the discount calendar has quietly become the marketing plan.
Read