If you are a TikTok Shop US seller who has budgeted for the 11.11 to 11.30 window and is about to put all of it into the four days of Black Friday and Cyber Monday, the slot you wanted no longer exists — and the useful question is not how to get it back. It is what the remaining supply actually costs per usable piece of content, and whether that cost survives the margin. A test costs the same number of impressions whenever you run it, but impressions do not cost the same in every week of November. The same test that costs six hundred dollars in the first week of the window costs about fourteen hundred in the last four days, which means the budget you saved for the peak buys eight tests instead of twenty. That is the whole argument for spending early, and it has nothing to do with brand building.
Take a $12,000 creator budget for the twenty days, a $42 average order and a $600 cost per usable test. In the warm-up week a thousand impressions cost about $7.50, so $600 buys 80,000 impressions and one clean read on one piece of creative. In the peak four days the same thousand impressions cost about $18, so $600 buys 33,000 impressions, which is not enough audience to conclude anything. Now run the two plans. Spread 20/30/50 and you get 20 tests in warm-up, 3 tested winners scaled in build, and $6,000 of volume at peak on creative you already know works. Put all $12,000 into the peak and you get eight tests, no time to read them, and a $42 product whose best hook is still a guess — on the four days when a guess costs the most. Same money, eight answers instead of twenty, and the twenty were available at 40% of the price.
Why the last four days are the most expensive place to learn anything
The instinct to save everything for the peak comes from a reasonable place: that is when demand is highest and every dollar feels like it converts. But demand being highest is exactly why the price is highest, and a seller with no tested creative is paying peak prices for the privilege of guessing. The four days do not reward the seller with the biggest budget, they reward the seller whose creative has already been through twenty versions of the question.
There is a second thing the early weeks buy that does not show up in any dashboard: the list of creators who actually deliver. By 26 November you know which four people posted on the date they promised and which four went quiet, and that knowledge is what makes the peak spend reliable. A seller who starts on 27 November is hiring people during the four days when everyone in the category is hiring the same people.
Step 1 — Write three phases with a job attached to each
Write the twenty days as three phases with a number attached to each, and write what each phase is allowed to buy. Warm-up, 11.11 to 11.19: 20% of the budget, buying tests. Build, 11.20 to 11.26: 30%, buying confirmation. Peak, 11.27 to 11.30: 50%, buying volume. If a phase spends on something outside its job, the next phase has less to work with, and the plan quietly becomes the all-in-peak plan it was supposed to replace.
On a $12,000 budget the warm-up number is $2,400, which at $100 a brief is twenty-four creators. That is not too many, because most of them will produce a video you never use and that is the point of the phase: the twenty-four exist to find the two or three that work. Keep each brief small — one product, one hook, one format — so that what you learn is attributable to the hook rather than to the creator.
Before the first brief goes out, put the three phases on one page with their budget, their job and their target count — twenty-four tests, three scale-ups, one peak. Get the 20/30/50 creator budget planner from a DAMI specialist and mark each creator against it as they deliver, because the second phase is only as good as the record of who actually posted in the first one.
Step 2 — Spend the warm-up on tests, and read them on cost per order
Judge a warm-up test on one question: did it produce orders at a cost per order you could live with at three times the spend? Not “did it get views”, not “did the creator like it”. A test that produced two orders is not a failed test, it is a read: at that cost per order, scaling would either work or not, and you now know which. A test that produced views and no orders is also a read, and it is the more valuable of the two because it tells you the product page is the problem rather than the creative.
Twenty-four briefs at $100 will not all come back as videos, and you should plan for eighteen to twenty. The four to six that never arrive are not a loss, they are the reason the briefs were small: at $100 the cost of a no-show is bearable, at $1,000 it is a decision you have to defend. This is also why the warm-up is the phase where you can afford to take a chance on a creator nobody has tested.
Step 3 — Spend the build on confirmation, and only on the winners
In the build phase, spend only on creators whose warm-up video already cleared your cost per order. No new names, no exceptions, and this is the hardest rule in the plan because a new name always looks promising and the budget always feels like it should be doing something. What the build phase is doing is confirming: taking the two or three that worked and giving them three to five times the spend to see whether the result holds at volume.
Set the scale-up number before you start, so the decision is not made in the mood of a good week. If a warm-up video produced orders at $18 each and your ceiling is $22, scale-up is a yes and the only question is how much. If it produced them at $26, the answer is no, and the money stays in the peak rather than being spent proving a number you already have.
Step 4 — Spend the peak on volume, with no new names in it
The peak is where you stop testing and start spending, and the reason is arithmetic rather than discipline. At $18 per thousand impressions, the same read costs 2.4 times what it cost in the first week, and you have four days to act on whatever you learn — which is not enough time to learn it. So the peak phase should contain zero new creators, one or two creative variants on the winners, and a daily spend cap that you do not raise because the morning looked good.
One thing does belong in the peak that is not volume: the returns read. Orders placed from 11.11 through 12.31 can come back until 10 February under the extended holiday policy, and creator commission settles against net sales after delivery, which means a December return wave reduces January creator payments. Plan the peak spend against net revenue rather than gross, or the four best days will look better in the dashboard than they do in the bank.
Where DAMI fits: three moves that keep twenty-four briefs on one page
DAMI brings an 8-million-plus creator database, competitor and competitor-store creator search, automated outreach tasks, AI-assisted outreach, AI video creation, sample management, campaign management, social media management, store analytics, and multi-store team dashboards into one place. Three phases, three jobs, one budget line. The reason this plan fails in practice is not the split — it is that twenty-four briefs, three scale-ups and a peak week are more commitments than anyone tracks in their head while also running the store.
Step one: put the three phases and their budget in one plan
Lay out warm-up, build and peak with their dates, their budget share and their target counts, then keep the plan visible for the whole twenty days. Output: a plan where a phase spending outside its job is visible the same day rather than in the December review.
Step two: run the warm-up as an automated task, not as messages
Send the small brief to twenty-four creators through one outreach task with a scheduled follow-up, so the tests actually happen inside the first week instead of dribbling into the second. Output: twenty or so videos inside nine days, which is the raw material the next two phases run on.
Step three: rank by cost per order and scale the winners in one place
Read each creator's cost per order next to their spend, mark the ones under your ceiling, and move budget to those before 27 November. Output: a shortlist of two or three proven creators going into the peak, which is the only thing that makes peak spend more than a guess.
Questions sellers ask about front-loading the budget
My product has never sold on TikTok Shop. Does the warm-up still make sense?
It is even more necessary, because you have no baseline at all and the peak would be your first test at the worst price. The difference is what you count as a win: for a new listing, a warm-up that produces orders at any cost you can afford to repeat is a strong result, and one that produces views without orders tells you to fix the listing before spending again.
Is 20% enough for warm-up if I have a smaller budget?
On a small budget the percentage matters less than the count. You need enough tests for the results to mean something, which is roughly fifteen to twenty pieces of creative. If 20% of your budget cannot buy fifteen tests, either shrink the brief so it can, or accept that this campaign is a single-shot spend and plan it as such rather than pretending it is a testing plan.
What counts as a tested winner?
A video that produced orders at a cost per order you would still accept at three times the spend, on a product whose return rate you have seen. Two orders at the right cost per order is enough if the audience was reached properly; twenty orders at the wrong cost per order is not a winner, it is a subsidy you paid yourself.
Should I use the same creators in the peak that worked in warm-up?
Yes, and one thing more: ask them to make a second video rather than boosting the first. A video that has been running for two weeks has already reached most of its organic audience, so additional spend on it buys incrementally less each day. A second video from the same creator reaches the same audience with a new hook, which is the cheaper way to extend a proven relationship.
What if nothing in the warm-up works?
Then the peak budget should go down, not up, and that is the plan working rather than failing. A campaign where you discover in week one that the creative does not convert has cost you $2,400; the same discovery in the last four days costs you $12,000 and there is nothing left to do with it. Keep the remaining budget for the next window rather than spending it on a hope.
The number to check before 27 November
The most expensive thing in this window is not a creator. It is not knowing which creator works, while still paying for all of them. DAMI brings an 8-million-plus creator database, competitor and competitor-store creator search, automated outreach tasks, AI-assisted outreach, AI video creation, sample management, campaign management, social media management, store analytics, and multi-store team dashboards into one place.
Open last month's creator spend and write down three numbers: how many creators you paid, how many produced a video you actually used, and what the cost per order was on the best one. Fifteen minutes. If the third number is above what you would pay at three times the volume, you already know the peak is not the place to fix it — and you have four weeks to fix it somewhere cheaper.
Start with DAMI while the numbers are still small enough to read.
