Launch Deal60% off your first year — use code at checkout60% off — code Limited time only

← All posts

The Looks I Save by Hand: An Outfit Studio Playbook

Wardrowbe Team8 min read
A woman in a deep red dress getting ready for a date night, with carefully selected accessories laid out on a marble surface

The AI in Wardrowbe is good at picking what to wear when you don't want to think. Weather, occassion, what you wore last week, what you've rated up. It runs every morning, it gets dressed for you, you accept or you swap. That covers about 80% of my mornings, and probably yours.

The other 20% is where Outfit Studio earns its keep. There are certain outfits I want sitting in my lookbook, named, ready, one tap away. Not generated fresh every morning. Locked in. The date-night look I figured out in the mirror last Saturday. The Monday rotation I run on autopilot. The wedding outfit I want to stop re-deciding two weeks before every wedding.

This is the playbook I actually use. Five looks worth building by hand, why they belong in Studio instead of being left to the AI, and the small habits that make a saved lookbook actually pay off.

1. The Date Night Look

This is the one everybody asks me about, so it goes first.

Date-night outfits are where AI suggestions break down. The model knows your wardrobe, it knows the weather, it knows "date" is a formal-ish occasion. What it doesn't know is the specific cardigan-and-silk-top thing you and your partner both like, or the boots that are technically too warm for May but you wear them anyway because they look right. That's taste, not data.

So I build it once, in Studio, in about ninety seconds. Open the canvas. Pick the top, the bottom, the outerwear, the footwear, one or two accessories. The picker has search now that actually works (name, subtype, tags, color all hit at once), so I can find the silk top by typing "silk" and not scrolling for thirty seconds. Hit the Date occasion chip. Name it something I'll recognise like "deep red, going-out". Save to lookbook.

Now it lives in My Looks. Friday rolls around, I open Outfits, filter to My Looks, tap it, done. The AI didn't have to guess and I didn't have to think.

A date-night look saved in Outfit Studio with role slots filled, occasion set to Date, named and ready to wear

One tip though: tap any item on the saved outfit to open the lightbox, pinch to zoom. I do this before I leave the house to remember exactly how the fabric drapes. Sounds silly until you've second-guessed yourself in the bedroom for ten minutes.

2. The Monday Uniform

I have a Monday outfit. It's almost the same outfit every Monday. Charcoal trousers, a fitted knit, a clean pair of leather sneakers, a watch. Office occasion. It's not exciting and it doesn't need to be. Monday is hard enough.

Before Studio, the AI would keep trying to be clever on Mondays. It would suggest the knit one week, the oxford the next week, a chore jacket the week after that because I'd rated it up three months ago. All of those work, none of them are the Monday outfit. I'd reject the suggestion, swap to what I actually wanted, and the cycle would repeat.

The fix was to build the Monday uniform in Studio once and then schedule it. Save the outfit, set occasion to Office, name it "monday default". Then in Notifications, set up a Monday morning schedule with Office as the occasion. The morning push now nudges me toward office-appropriate items, and the saved look is sitting in My Looks if I want to skip the picker entirely.

The thing I underestimated about saving uniforms: it's not really about the days you wear them. It's about every other day, when you can ignore them and focus on actual decisions.

3. The Wedding-Guest Capsule

This one is the most "Studio is necessary" use case I've found.

Wedding invites come weeks ahead. I used to think about what I'd wear about three days before, panic-shop or panic-iron, and then forget the outfit entirely by the next wedding. Three weddings later I'd have three slightly different formal looks and no memory of which one worked.

Now when a wedding invite lands, I build the outfit immediately. In Studio. Formal occasion, full canvas: dress shirt, suit pieces, tie, dress shoes, the right belt, pocket square if it calls for one. Save to lookbook, name it like "spring wedding, navy suit". Don't mark it worn yet. It just sits there.

Two things happen. One, the decision is made; future-me doesn't have to redo the work. Two, when the wedding actually happens, I open the saved look, tap Mark as Worn, and that look becomes a record. Worn outfits are locked from edits on purpose (the wear history needs to match reality, otherwise the learning engine gets confused), but I can clone-and-edit if I want a variant for the next wedding.

After a year of doing this I have a small library of formal looks I can flip through. "Garden wedding, beige," "winter wedding, charcoal three-piece," "rehearsal dinner, smart casual." That used to be six hours of panic per event. Now it's six minutes once, and zero panic later.

4. The Gym Set

A different shape of use case. The gym outfit is boring. There are three combinations I rotate. I don't need the AI to think hard about them; I need to log them so my wear data is accurate.

So all three live in Studio under the Sporty occasion. When I get back from the gym, I open the matching saved look, hit Mark as Worn, done. The wears-since-wash counter ticks up on the right items, the wash interval logic knows when those tees are due for a cycle, and the learning model sees real wear data instead of unrecorded gaps.

If I skipped this and just wore the gym kit without logging anything, my wardrobe analytics would slowly drift toward "you never wear half your closet." Saving the boring outfits is what keeps the analytics honest.

5. The One With a Custom Occasion

The newest piece. Occasions used to be a fixed list (casual, office, formal, date, sporty, outdoor) and the AI worked within those buckets. That's enough for most of life but not for everything.

A few weeks back a friend invited me to a rooftop dinner. Not quite formal, definitely not casual, the kind of thing where "smart casual" is the prompt but you don't actually want to look like you're trying. None of the preset occasions captured it.

So I used the Custom occasion. You set it up once in Style Preferences: a 500-character textbox where you describe in plain English what custom means for you. Mine reads something like "elevated casual, slightly tailored, dark tones, no logos, leather over canvas, comfortable enough to sit through dinner." Those instructions get appended to the AI prompt as hard constraints whenever I select Custom on a suggestion or save a Studio outfit with that occasion.

For this rooftop dinner I opened Studio, built the look I had in mind, set occasion to Custom, named it "rooftop dinner, smart casual", saved it. Now Custom isn't a one-off; the saved outfit is in My Looks under the Custom chip and the AI knows what Custom means for me in future suggestions. If your life has a recurring event that doesn't fit a preset (school pickup, gallery opening, board meeting that's also outdoors, whatever), this is the trapdoor for it.

The Saving Habit

Two save buttons on the canvas, not one. Save to Lookbook stores the outfit for later. Save and Mark Worn stores it and logs that you wore it today. They look almost identical and the difference matters: lookbook entries are editable, worn entries are locked.

My rule of thumb. If I'm building an outfit in advance (wedding, date night, packing for a trip), it goes to Lookbook. If I'm logging what I actually wore today (gym, Monday uniform, anything I built on the fly that morning), it gets Marked Worn. If I'm building an outfit and wearing it the same day, the second button does both in one tap.

The auto-draft also helps a lot. Studio saves your in-progress canvas every few seconds, so if you get interrupted mid-build, come back tomorrow, your work is still there. I keep a half-finished "summer wedding 2026" draft sitting around for weeks at a time and pick at it whenever I find a piece that might fit.

Why Build Looks at All

The honest case for hand-built outfits is this: the AI is good at what to wear today, but it's not good at remembering the specific combinations you've already figured out work for you. Every morning's suggestion starts more or less from scratch. Studio is the place where you write down the answers, so you stop re-solving the same problem.

Five looks is enough to start. Build the date-night one, the Monday uniform, the formal capsule for the next event on your calendar, the gym rotation, and the one custom occasion that keeps coming up in your life. After a month of using them, you'll notice how many low-friction decisions you've offloaded.

Open Studio on web or on the iPhone app and try one. The first one takes about two minutes. Every subsequent wear takes about two seconds.

Ready to get dressed in 60 seconds?

Start your 7-day free trial. No credit card required.

Start Free TrialSee Pricing

Keep Reading

Pack Light for Any Trip Using Your Wardrobe Data

How wardrobe analytics and AI outfit planning help you pack fewer clothes that work harder. Travel capsule wardrobes, built from data.

Introducing Outfit Studio — Build Outfits by Hand, Not Just by AI

Outfit Studio is the new hand-built outfit editor in Wardrowbe. Canvas, AI assist, edit mode, lineage, and a unified Outfits tab across web and mobile.

New: The App Now Walks You Through Everything

Wardrowbe's new guided setup shows you what the app does before you subscribe, then walks you step by step from first photo to daily outfit notifications.