Fashion operators search "best time to run Facebook ads for clothing" and get a blog that says 8pm. Sometimes evenings work. Sometimes you just advertised a sand hoodie whose M and L died at lunch.
Priors you may test (and must measure)
| Prior | Why people believe it | How it fails |
|---|---|---|
| Evening 17:00-22:00 | Mobile sofa browse | Your buyers are at work in that window, or the ad account is in the wrong TZ |
| Sunday night | Lookbook energy before the week | One pretty Sunday is not a schedule |
| Payday Friday | AOV lift | UK payday is not the last Friday everywhere |
| Launch morning | Drop culture | You need stock and support on, not a clock trick |
| Overnight OFF | Cheap, tired clicks | If those hours do not spend, you changed nothing |
How to set the Meta clock: Meta dayparting. Google: ad schedules.
Stock still beats the clock
If paid conversions are M and L, and those are gone, an 8pm boost buys refunds and "is this in stock?" tickets. Do the merch week first: core-size, remove a size, UGC that still wears a 10.
Dayparting a broken size run is a more precise way to waste the evening.
Timezones, markets, wholesale
- Two ad accounts, two clocks. Do not paste a London evening onto US East.
- Shopify Markets plus one Meta account is how "evening" becomes undefined.
- Wholesale / B2B fashion should not inherit DTC sofa hours. Different buyer, different lunch.
Returns after midnight
Late-night fashion impulse returns. If you keep 22:00-24:00 on, watch refund rate by hour for a month, not only ROAS. A cheap CPA that comes back in a poly bag is expensive.
A four-week fashion test
- Week 0: export hours + variant cover on top 10 paid styles.
- Week 1: overnight OFF on one lifetime-budget prospecting ad set. Catalog stays wide.
- Week 2: hold. No payday plus yet.
- Week 3: if evenings truly outperform, a small Search bid adj or a note in the brief. Still no +40%.
- Week 4: keep only the cut you would trust when the founder is on a plane.
Weather that week is a demand story, not an hourly kill: weather-based dayparting.
Worked apparel example
UK womenswear. London ad account. 60-day Meta hours: 00:00-06:00 is 12% of spend, 4% of purchases, mostly prospecting. 18:00-21:00 is 24% of spend and 27% of purchases. Sunday 18:00-22:00 looks fat. Sand hoodie M/L went OOS Friday 16:00. The best static still shows a size 10.
Week 1: overnight OFF on the lifetime-budget prospecting ad set only. Do not boost Sunday yet. Pause or swap the size-10 static. Hide sand M/L in the catalogue. Week 2: hold. If Sunday is still fat and the size run is healthy, a small day-of-week note in the brief. Not a +40% bid.
That is fashion dayparting. The clock is the last cut, not the first.
Build a fashion daypart without folklore
- Export Meta and Shopify hours. 60 days of ad hourly plus order timestamps. Same timezone or you are lying to yourself.
- Mark the converting size run. If M/L are gone, do not buy more evening traffic for that hoodie. Fix the set first.
- Cut one overnight OFF on prospecting. Lifetime-budget ad sets only, for a native Meta clock.
- Test one day-of-week story. Sunday evening or payday Friday, not both, not plus 40% yet.
- Hold through one full retail week. Fashion weeks are lumpy. Do not retune Tuesday because Monday was quiet.
Questions people actually ask
What is the best time to run Facebook ads for a fashion brand?
There is no universal 8pm. Many apparel accounts do see evening and Sunday-night browse. Your Meta hourly breakdown in the ad account timezone is the only schedule that matters. Start by cutting a proven dead overnight, not by boosting folklore.
Should fashion brands daypart catalog ads?
Lightly, if at all. Catalog and Advantage+ need room. If spend is leaking, check sold-out sizes and core-size health before you shrink the clock.
Do payday and weekends change the schedule?
Often. UK payday windows and Sunday evening can lift AOV. Treat them as day-of-week tests, not a 168-cell redesign every month.
Does Ralph have fashion-specific dayparting?
No. The blocks are generic retail hours. Fashion advice on this page is merchandising and evidence, not a hidden apparel model.
US store, UK factory, ads in both?
Then you have two clocks. Do not copy a London evening onto an Eastern ad account. Split campaigns or accept that 'evening' is undefined.
Late-night fashion impulse and returns?
Watch refund rate on hours you keep on after 22:00. A pretty ROAS that comes back in the post is not a win.
If the hours are the problem: Ralph reads sixty days of your hourly results, proposes the switch-off with floors, and applies it on Meta after you approve. On Google he recommends the schedule. See Ralph · How Ralph dayparts · Docs