Watch the episode
Watch the episode: How Quince Runs 5,517 Meta Ads: 4 Strategies to Steal (9:12) · Watch on YouTube · More breakdowns on the channel
Quince had 5,517 live Meta ads on September 27, 2026. I pulled 1,000 of them from the Meta Ad Library and went through the creative itself: the headlines, the captions, the prices printed on the pictures and the opening lines of the videos. This is the written version of episode three of Steal the Strat, my series where I take a brand’s public ad account apart and pull out what a smaller brand can copy. The full episode is above if you would rather watch it.
Key takeaways
- Quince had 5,517 live Meta ads on September 27, 2026, about ten times the 229 Liquid Death and 316 Native ads I pulled the same week, combined.
- In a 1,000-ad sample of Quince’s Meta account, 784 ads were under 30 days old and 298 launched on a single day, August 31, 2026.
- Quince’s 1,000 sampled Meta ads used 678 different pictures and videos but only 143 different captions, so a new picture almost never waited on new copy.
- The 1,000 sampled Quince ads carried zero coupon codes and zero sitewide sales; Quince sells its price by printing a struck “traditional retail” price next to its own.
- For a small brand selling on its own site, the number to manage is new Facebook ads launched per week, and five a week on a fixed day is a workable start.
The short answer
How many Facebook ads should you run?
There is no correct total. The number that matters is how many new ads you launch each week, and whether you can keep doing it without your copy, your designer or your budget becoming the bottleneck. Most small accounts I look at have a handful of ads that have been running for months and nothing new behind them. When those ads tire, there is nothing ready to take their place.
Quince is the extreme opposite. It runs its Meta account like a publishing schedule. You do not need its volume. You need its habits, scaled down: a fixed launch day, a set of words that is already written, a price story that does not depend on a sale, and ads that answer the questions a store would answer. Each of those is below, with what Quince does and what to do at your size.
The numbers
How many Meta ads does Quince run?
live Meta ads on September 27, 2026
of 1,000 sampled ads under 30 days old
launched on a single day, August 31, 2026
of 1,000 clicks end at quince.com, none at a retailer
On September 27, 2026 the Meta Ad Library showed 5,517 live ads for Quince. For scale, Liquid Death had 229 live Meta ads when I pulled it on September 23, and Native had 316 on September 25. Quince was running about ten times both of them combined.
I pulled 1,000 of the 5,517, sampled across the whole account, so every count below except 5,517 is a count within that sample. Of those 1,000 Quince ads:
- 784 (78.4 percent) had been running for under 30 days, with a median of 27 days.
- Something new launched on 25 of the last 30 days.
- 522 launched in the week of August 31, and 298 of those on August 31 alone. Native’s biggest single day was 42.
- 980 of the 1,000 clicks ended at quince.com, 10 opened the Quince app, and none went to a retailer.
That last point is why the volume makes sense. Quince sells on its own site and nowhere else. There is no Target shelf or Amazon listing doing the work, so the feed is the only shelf Quince has, and Quince restocks that shelf every week.
The check
Are those really new ads, or old ads switched back on?
A start date in the ad library can belong to an old ad that was paused and restarted, so I checked the August 31 launches. Of the 298, 267 (89.6 percent) carry a headline and caption that appear nowhere in the older ads in my sample, matched character for character. If you ignore differences in spacing and line breaks, the figure is 82.9 percent. Either way, most of that day’s ads were new.
Two caveats apply to every age figure in this post. Days running means days since an ad last started delivering, not the day it was created, so a pause and a restart resets the clock. And an ad that has run a long time has survived something, but what it survived is not visible from outside. The ad library shows no spend and no results.
Steal the strat · four plays
Steal one: how often should you launch new Facebook ads?
Quince does not so much run ads as run drops, and each new wave looks nothing like the old one. The oldest carousels in my sample, including a maternity hoodie that had been running for 275 days, are plain catalog photos: a model against a pale studio wall, the product in the center, not one word on the picture. The wave that started on August 31 is tall and vertical, in warm browns and creams, with one line of thin serif type across the middle and often a price. The videos are customers and creators holding the jeans, the sweater or the quilt, often cropped below the face. 420 of the 435 videos in my sample were under 30 days old.
The reason this matters for your account is simple. A feed moves fast, and the same picture seen for the fifth time turns into wallpaper. A new picture is a new chance to stop someone’s thumb.
What to do
Sort your active ads by start date and count how many started in the last 30 days. Then pick one day a week and launch new pictures on that day, every week. Leave what is running alone and add to it.
At your size
Five new ads a week, not five hundred. Three new pictures and two customer videos is plenty. The discipline is the day, not the number. If making that many is the hard part, that is what static ad creation and UGC content creation are for.
Steal two: do you need new ad copy for every new creative?
You cannot launch hundreds of ads in a week if every ad needs new copy, so Quince does not write new copy. In my sample there are 678 different pictures and videos and only 143 different captions. Seven captions, the price comparisons, the factory line and one five-star review, ride on 472 of the 1,000 ads.
A new product gets all of them at once. On March 14, 2026 Quince launched one pair of chinos, shot on a white wall, as six carousel ads with six different captions, all six still running when I pulled the account. On July 13 a gold ring photographed on plain cream went out as seven ads with seven captions. The picture is a product photo. The caption does the selling.
The words are the slow part of making an ad. Settle them once, and the only new thing a launch needs is a picture.
What to do
Write a kit of six sentences, each a different argument: your price against theirs, the reason for the price, a real review in the customer’s own words, the guarantee, the material, and the one line that sounds like you. Every new picture goes out on those six sentences, one ad each.
At your size
The kit turns a week of writing into an afternoon with a camera. If you want help writing the six, that is conversion copywriting.
Steal three: how do you sell on price without running discounts?
With this many ads you would expect a sale somewhere. In the 1,000 I pulled there is not one coupon code and not one sitewide sale. The closest Quince comes is a bundle price or store credit. Instead, Quince sells the price by comparison.
In the captions: “Our full price < their sale price.” and “Competitors marked it down. We never marked it up.” On the pictures: a pair of travel pants at $49.90 with $128.00 crossed out underneath and the words traditional retail, or a black sweater at $119.90 next to a struck $298.00. “Traditional retail” is the phrase printed most often across the pictures in the sample.
Then the reason, before anyone can doubt it. The most used caption in the pull, on 233 of the 1,000 ads, opens: “Ever wonder why luxury products are so expensive? At Quince, they don’t have to be. We connect you directly with the factories that make for the world’s leading luxury brands.”
A discount tells people your price was too high yesterday. A comparison tells them someone else’s price is too high every day. And because a low price makes people suspicious, the explanation arrives before the question does.
What to do
Name the brand your customer would buy if you did not exist. Put your normal price next to theirs on the picture, and only if it is true. Then write one sentence that explains why yours costs less. Both go in your kit.
At your size
You probably do not share factories with luxury houses. You have something else: you make it yourself, you skip the retailer, or you buy fabric by the roll. Say that. The reason is the ad.
Steal four: what should your ads do that a store does for retail brands?
In a store you can pick a product up, you can see other people buying it, and you know you can bring it back. Quince’s ads do all three on purpose.
- Pick it up. The videos put the product in someone’s hands, pulling at the fabric and turning it over.
- See other people buying it. The pictures say sold out and back again: “The sweater that sells out every season is back.” “Waitlisted quilts are back.” A five-star review is printed across a leather chair: “Looks like a $5k chair.”
- Bring it back. The most used headline in the sample, on 157 of the 1,000 ads, is “Free shipping on everything and 1 full year to return.” One of those ads prints a customer review on the picture: “I was skeptical to order a couch online, but the risk paid off.”
Nobody worries about returning a sweater to Target. A brand that only sells on its own website has to put that promise in the ad, and the headline is the line everyone reads.
What to do
Pull your last ten headlines. If none of them says what happens when a customer changes their mind, write one that does. Then take your best review and print it on a picture. Run both alongside what you have now.
At your size
You have one review that names what your customer was afraid of and says it turned out fine. Print that one. If your return policy is too thin to put in a headline, fix the policy before the ad.
The one that decides
Which of these should you copy for your brand?
It depends on where your clicks end. Liquid Death’s ads quietly admit the product is available at Amazon. Native sent nine out of ten clicks to somebody else’s store. Quince sends almost every click to its own site and none to a store. I broke down Liquid Death’s Meta ads and Native’s Meta ads the same way, if you want to compare the three.
If your clicks end on your own site, take all four, and start with the weekly drop, because everything else feeds it. If your clicks end on a retailer’s shelf, take the first three and leave the fourth to the store, because Amazon already shows the reviews and the return policy.
If you want to see how your own account compares, my free Facebook ad audit reads your page and a competitor’s from public data and emails you the write-up in about a minute. If you sell clothing, my page on Facebook ads for apparel brands goes further on creative for this category.
FAQ
Frequently asked questions
How many Facebook ads should a small brand run at once?
There is no right total. Track how many new ads you launch each week instead. Quince launched 298 Meta ads on one day in my sample, but for a small brand selling on its own site, five new ads a week on a fixed day, three pictures and two customer videos, is a workable rhythm.
How often should you refresh Facebook ad creative?
Weekly is a good default for a brand that sells on its own site. In a 1,000-ad sample of Quince's Meta account, 784 ads were under 30 days old and something new launched on 25 of the last 30 days.
Do you need new ad copy for every new Facebook ad image?
No. Quince's 1,000 sampled Meta ads used 678 different pictures and videos with only 143 different captions. Write a kit of about six sentences, each a different argument, and pair every new picture with them.
Does Quince run discounts in its Meta ads?
Not in the 1,000 ads I pulled: no coupon codes and no sitewide sales. Quince prints its price next to a struck traditional retail price and explains the gap with a line about the factories it works with.
Where does this data come from?
The public Meta Ad Library, read on September 27, 2026. It shows which ads are live, their copy and creative, and when they last started delivering. It does not show spend or results.
Method
How was this measured?
Everything here comes from the Meta Ad Library, read on September 27, 2026 through my own teardown tool. The library reported 5,517 live ads for Quince. I pulled 1,000 of them, 18.1 percent, sampled across the account, so every count except 5,517 is a count within those 1,000. Liquid Death’s and Native’s figures come from their own pulls on September 23 and September 25, 2026. Meta publishes no spend, impressions, click-through rate, cost per acquisition, return on ad spend or revenue for US commercial ads, so none of those is claimed here. Why Quince launches in waves, reuses its captions and sells by comparison is my reading of public data, not a measurement of results. I have no relationship with Quince.
Want this built for your brand?
If you want a weekly launch rhythm like this built for your brand, with the creative and the kit of sentences behind it, that is what I do for DTC brands.
