K-beauty Meta ads, Oct 20 to Oct 26: deadline wording at 8.3%, about double the previous 7 days
From Oct 20 to Oct 26, 2026, KOAD saw 3,635 new K-beauty ads on Meta and 30 on TikTok. Among new Meta ads, 8.3% carried deadline wording, about double the 4.3% of the previous 7 days, while coupon and discount wording fell from 10.9% to 7.4%. Ad spend and sales are not visible.
New ads are ads whose Meta start date falls between Oct 20 and Oct 26, compared with the prior 7 days (Oct 13 to Oct 19). Recently confirmed ads are ads KOAD confirmed again within the last 14 days. Source: KOAD archive (public delivery information from the Meta Ad Library).
Offer wording: what went up this week
Of 3,635 new Meta ads, 25.1% carried at least one offer signal. Shares below use all new ads as the denominator. An ad with several signals is counted for each.
| Offer signal | This week | Prior 7 days | Change | Brands |
|---|---|---|---|---|
| Deadline wording | 8.3% | 4.3% | +4.0 pts | 23 |
| Coupon or discount | 7.4% | 10.9% | -3.5 pts | 35 |
| 1+1 or bundle | 7.2% | 6.9% | +0.3 pts | 35 |
| New launch | 3.4% | 5.6% | -2.2 pts | 30 |
| Free gift | 1.7% | 3.5% | -1.8 pts | 14 |
| Free shipping | 1.3% | 1.4% | -0.1 pts | 4 |
This weekPrior 7 days
Most used phrases
| Phrase (Korean) | Ads | Brands | Share | Change |
|---|---|---|---|---|
| 선착순 (first come, first served) | 264 | 15 | 7.3% | +3.7 pts |
| 1+1 (1+1) | 259 | 34 | 7.1% | +0.3 pts |
| 최대 할인 (up to X% off) | 187 | 32 | 5.1% | +1.1 pts |
| 한정 off (limited off) | 181 | 4 | 5.0% | +4.7 pts |
| 추가 증정 (extra gift) | 112 | 12 | 3.1% | +1.5 pts |
Phrases were found in 66.6% of new ads. "한정 off" came from 4 brands, so it may reflect a few brand promotions.
What ads led with
85.9% of new ads had a classified appeal. An ad can have several appeals, so the total is above 100%.
| Appeal | Ads | This week | Prior 7 days | Change |
|---|---|---|---|---|
| Price and offers | 2,167 | 59.6% | 61.8% | -2.2 pts |
| Ingredients and evidence | 1,831 | 50.4% | 52.4% | -2.0 pts |
| Product introduction | 593 | 16.3% | 12.7% | +3.6 pts |
| Reviews and experience | 76 | 2.1% | 3.0% | -0.9 pts |
This weekPrior 7 days
Ingredients named in new ads
19.1% of new ads named an ingredient KOAD recognizes. Shares use all new ads as the denominator.
| Ingredient | Ads | This week | Prior 7 days | Change |
|---|---|---|---|---|
| Collagen | 223 | 6.1% | 6.7% | -0.6 pts |
| Retinol | 165 | 4.5% | 6.7% | -2.2 pts |
| PDRN (salmon DNA) | 157 | 4.3% | 7.6% | -3.3 pts |
| Panthenol | 127 | 3.5% | 2.8% | +0.7 pts |
| Peptides | 86 | 2.4% | 3.7% | -1.3 pts |
| Ceramides | 74 | 2.0% | 2.8% | -0.8 pts |
| Retinal | 45 | 1.2% | 0.3% | +0.9 pts |
This weekPrior 7 days
Product types
83.5% of new ads had a classified product type.
| Product type | Ads | This week | Prior 7 days | Change |
|---|---|---|---|---|
| Skincare › Cream | 503 | 13.8% | 16.8% | -3.0 pts |
| Base makeup › Foundation | 340 | 9.4% | 10.1% | -0.7 pts |
| Skincare › Serum | 318 | 8.7% | 9.9% | -1.2 pts |
| Masks › Sheet masks | 198 | 5.4% | 4.6% | +0.8 pts |
| Skincare › Ampoule | 187 | 5.1% | 6.8% | -1.7 pts |
| Base makeup › Cushion | 178 | 4.9% | 2.0% | +2.9 pts |
This weekPrior 7 days
TikTok
KOAD saw 30 new TikTok ads (prior 7 days 37, 18 accounts). The sample is small, so this issue shows counts only.
Long-running ads: observed facts only
Of 15,197 recently confirmed ads, 3,514 (23.1%) had run past 30 days and 1,788 (11.8%) past 60 days. It differs a lot by category.
| Category | Recently confirmed | Past 30 days | Share | Past 60 days | Bar |
|---|---|---|---|---|---|
| Beauty devices | 971 | 432 | 44.5% | 151 | |
| Masks and packs | 1,525 | 418 | 27.4% | 244 | |
| Lip makeup | 499 | 125 | 25.1% | 60 | |
| Hair and scalp | 598 | 143 | 23.9% | 102 | |
| Cleansing | 1,119 | 262 | 23.4% | 130 | |
| Skincare | 6,838 | 1,501 | 22.0% | 736 | |
| Sun care | 769 | 113 | 14.7% | 59 | |
| Base makeup | 1,977 | 230 | 11.6% | 115 |
Run days count from the Meta start date to the last day KOAD confirmed the ad. Running long is an observation that the ad stayed live, not a measure of performance.
Founder's reading
ReadingThis week, new ads led with time pressure more than with the size of the discount. Deadline wording and "first come, first served" went up, while coupon, discount and free gift wording went down.
The 303 new ads with deadline wording came from 23 brands, and the 181 ads with "limited off" came from 4 brands. This may be a few brands whose promotions fell in the same week, so it is too early to read it as a market-wide shift. We will check whether it continues next week.
How long ads stay live differs a lot by category. In beauty devices, 44.5% of recently confirmed ads had run past 30 days. In base makeup it was 11.6%. Base makeup looks like a category where brands change creatives often.
What is different from last week. This is the first issue. The biggest change this week is the type of offer wording: deadline wording went up while coupon and discount wording went down.
What these numbers cannot tell
- Ad spend, impressions, clicks and sales are unknown. Meta does not publish spend or impressions for regular commercial ads.
- KOAD counts only the ads it collected. Fewer new ads than the prior week does not mean the market shrank.
- Offer signals and appeals come from ad copy and text inside images. Some ads are not classified, so each item shows the classified share.
- One week can move a lot because of a few brand promotions. Trends over several weeks are safer to read.
Common questions
Where do these numbers come from?
From the KOAD archive (public delivery information from the Meta Ad Library). Numbers are copied from daily snapshots and shown with their reference date and sample.
Is a long-running ad a good ad?
That cannot be said. Run days only show how long an ad stayed live. Spend and sales are not public, so performance is unknown.
If there are fewer new ads than last week, did the market shrink?
No. KOAD counts only the ads it collected, so the number depends on collection. When the two weeks differ by more than two times, KOAD skips the comparison.
Source: KOAD archive (public delivery information from the Meta Ad Library). Numbers are copied from the 2026-10-26 snapshot. Definitions and limits are on the method page.
Numbers were copied from snapshots, AI was used for the first draft of the text, and the founder checks the numbers and sentences before publishing. Rights holders can request corrections or removal at [email protected].