Apparel is a category where the buyer is not only judging the product. They are judging themselves in the product. The same jacket sits on twenty storefronts at once, so what separates a sale from a scroll is whether the viewer can predict how that jacket will fall on their own shoulders. Every apparel purchase is a prediction, and a wrong prediction comes back as a return that is paid for out of the margin the sale just created. The number that matters in this category is not units sold. It is units kept.

That single fact reshapes the mechanics. Size and colour multiply every style into a dozen listings, and the data plumbing is built around that multiplication. Google Merchant Center requires colour, age group and gender on every Apparel and Accessories offer in free listings, and size on clothing and shoes; for Shopping ads it requires the same attributes when you target Brazil, France, Germany, Japan, the United Kingdom or the United States. Variants also have to share an item group ID, or Google shows them as separate products instead of one listing with options. Instagram allows up to twenty products and shops to be tagged in a photo feed post, and a carousel carries twenty tags across all its media. Where checkout happens depends on the surface: Meta's Commerce Platform developer documentation describes buyers discovering products on Facebook and Instagram and completing checkout on the seller's website, and Pinterest states that direct links take users straight to your site. So most apparel funnels end on a page you control, and if that page does not repeat the fit information the video gave, the sale dies at the size dropdown.

As a category, apparel is largely unrestricted. Clothing appears on none of the major prohibition lists: not Meta's twenty-four prohibited commerce categories, not TikTok Shop's US prohibited categories, not Snap's merchant prohibitions, not Pinterest's merchant guidelines. The constraints that actually bite sit one layer below platform policy: textile fibre and origin labelling, children's product safety, counterfeit and trade dress, environmental claims, and the creative rules for swimwear, intimates and shapewear. Three sub-segments do hit platform limits, and they are the ones social sellers gravitate to: secondhand, kidswear, and anything cut close to the body. The expensive failures in this category are usually compliance failures made by people who assumed clothing was unregulated because no platform policy page said otherwise.

Who is buying, and how they decide

Womenswear carries the volume and skews young on short video, menswear and streetwear behave like collectibles, and kidswear is bought by a parent solving a size problem on a deadline. Across all three the decision runs in a fixed order, and it is not the order marketers assume. First comes recognition: does this look like me, or like the version of me I am currently trying to be. That step is emotional and fast, and it is what a scroll-stopping video buys you. Then comes interrogation, which is where most sales are lost. How tall is the person wearing it, what size are they in, is the fabric heavy or thin, does it stretch, is it see-through in daylight, does it sit at the hip or the waist. Buyers answer these questions in the comments and in reviews before they ever reach your site, which means the comment section is functionally part of the product page and should be staffed like one. Then comes the risk check: what happens if it does not fit, how long is shipping, and is this a real brand or a page that dropships from a catalogue. Price is usually last, and it is judged as cost per wear rather than as a number, which is why styling a single piece three ways is usually a better lever than discounting it. Repeat purchase turns almost entirely on fit confidence rather than brand affection. Once a customer knows they are a particular size in your brand, the second order is nearly frictionless, and the whole economics of the account change. That is why a resolved size question is worth more than a new follower.

Where this actually sells

Not every platform is worth your time in this category. These are the ones that are, and why.

TikTok This is where demand for a garment gets created rather than met. Buyers are not searching for a ribbed knit midi, they see one move on a body and want it within the minute. Motion is the information a photo cannot supply, and clothing is mostly judged on motion. The commerce layer is geographically uneven, and that shapes strategy more than creative does: TikTok Shop reached ten European markets on 15 June 2026 when Austria, Belgium, the Netherlands and Poland joined France, Germany, Ireland, Italy, Spain and the UK, and it is not available in Canada. Where Shop exists you can sell in feed. Where it does not, TikTok is a demand engine feeding an offsite cart, and you should build the funnel that way now rather than waiting for a launch with no published date.
Instagram Instagram is where a buyer checks whether you are real before they spend. They read the grid for consistency, the Reels for movement, the tagged photos for other people's bodies, and the DMs for a size answer. Organic product tagging is live on feed posts, Reels and Stories, so the garment can be identified without a link in every caption. Instagram added affiliate product tagging in Reels on 24 March 2026, but its own announcement put the rollout in the US, Brazil, India, Indonesia and Thailand, so treat it as unavailable in Canada, the UK and the EU until Meta says otherwise. Shops availability is tiered by market, and per Meta's help documentation a shop's location is determined by the location of the commerce account rather than chosen freely, so you cannot simply pick a richer market. Treat Instagram as the verification and objection-handling surface, not the checkout.
Pinterest Pinterest catches the buyer earlier and holds them longer. People plan wardrobes here by season, occasion and body type, and a pin can keep surfacing long after it is published, which suits repeatable core styles better than a trend piece with a three week life. There is no native checkout: Pinterest states that direct links in ads take users straight to your site rather than to a close-up, and that this is the default and cannot be disabled. Catalogs are genuinely region-limited, and the prerequisites include a claimed website with contact details, a shipping policy and a refund policy, so a published returns policy is an entry requirement rather than a nice-to-have. Resale sellers should read the split carefully: pre-owned products are not allowed as organic product Pins, but Pinterest states shopping ads may include pre-owned products, with named category exclusions.
YouTube YouTube answers the question short video cannot: is this worth it after a month of wear. Long-form try-on, wash-and-wear updates and honest comparisons catch buyers with money out and doubt remaining, and they keep earning for years, which matters for evergreen styles. Shopping is tightly gated. The affiliate programme is available in fourteen countries and creators can tag their own store's products from twelve, with Canada, the UK and the EU absent from both, on top of YouTube Partner Program membership and its subscriber threshold. Store connection reaches well beyond Shopify, including Wix, FourthWall, Spreadshop, SHOPLINE and merch platforms such as Bonfire, Cotton Bureau and Threadless, which makes it a natural home for design-led apparel. Two operational facts to plan around: connected products go through a compliance review, and when an item is disapproved the creator cannot appeal it, the store admin or merch retailer has to submit it through Google Merchant Center. Commissions are paid through AdSense sixty to one hundred and twenty days after purchase, specifically to account for returns, so plan cash flow around that lag.
Whatnot For secondhand, vintage, streetwear and sneakers, the buyer wants to be talked through a single piece in real time and then win it. Live auction resale rewards sourcing skill rather than capital, and it solves the one-of-one problem that catalogue commerce handles badly: a single garment in a single size cannot carry a paid campaign, but it can carry two minutes of a live show. Confirm current market availability inside the app before building a plan around it, and treat any country list you find in a blog post as out of date.

The content that works

The fit-data try-on The creator states their height, their usual size, the size they are wearing and what they would change. It converts the viewer's guess into a comparison against a known body. It is also return reduction disguised as content, because the people who would have ordered the wrong size self-correct before checkout instead of after.
Movement and fabric proof Walking, sitting, reaching, and a deliberate backlight pass to show opacity. Buyers have specific unspoken fears in this category: drape, cling, stretch recovery and see-through fabric. A flat-lay answers none of them and a studio still answers them dishonestly. Showing the garment sit down is worth more than another angle of it standing.
One piece, three outfits Styling the same garment across different contexts raises perceived cost per wear without ever discussing price, which is how this category's buyers actually calculate value. It also multiplies one production day into several posts and gives you a clean test of which context converts, which is usually a sharper signal than which garment converts.
Live selling with open size questions Live is the format where a hesitating buyer can get their exact fit question answered before the intent fades. Check which live surfaces still exist before planning around one: Instagram Live Shopping ended in March 2023 and Facebook Live Shopping closed in October 2022, so live commerce in practice means TikTok LIVE where Shop operates, YouTube live streams in eligible markets, or a live marketplace. Treat the recurring questions as a content brief, not as support tickets.
AR try-on for eyewear, footwear and accessories Where the fit variable is narrow, augmented reality carries information that video does not. Snapchat's catalog-powered shopping Lenses are built for this, and Snap states a single Lens can hold up to twenty AR try-ons, so a shopper can move through styles in one session. Check your catalogue against Snap's merchant policies first, because they prohibit products that require age gating outright, which narrows what you can put on that surface.
Customer bodies, credited and measured Reposted customer content outperforms studio imagery when the customer's body resembles the viewer's, and it matters most for anyone outside sample size. Ask for height and size worn when you request permission, then publish both. The FTC's material connection rules apply here: if you gave the garment away or paid for the post, the disclosure has to be unavoidable, and for video it belongs inside the video rather than only in the description.

The path from scroll to sale

  1. The garment does something in the first second. Not a logo, not a face, not a caption card: a twirl, a sit, a layer coming off. Motion is the information the buyer came for, and stills lose this category before the sound loads.
  2. The second watch carries the fit data. Height, size worn, fabric weight and one honest caveat, on screen and in the caption, so a viewer can locate themselves without asking. Every question you answer here is a DM you do not have to staff and a return you do not have to fund.
  3. Objections get handled in public. Answer size questions in the comments where the next hundred viewers can read them, and keep a saved reply that routes to the size guide. If you are running DMs, respect the messaging windows: Meta's policy gives businesses up to twenty-four hours to respond, and promotional content beyond that window needs an approved message tag or a sponsored message. Note also that someone logged into Facebook cannot message your Instagram professional account, so do not build a single inbox plan across both apps.
  4. The landing page repeats the video rather than replacing it. Same garment, same model, same stated height and size worn, a size chart with real measurements, and the exchange terms visible before the size dropdown. Because checkout is offsite on Meta and Pinterest, this page is where most apparel funnels break, and the break is nearly always an unanswered fit question.
  5. Checkout carries an exchange promise rather than a returns policy. The buyer is not asking for their money back, they are asking what happens if they guessed wrong. Capture the size they bought against the customer record at this step, because that field is what makes the second sale cheap.
  6. Post-purchase, the first resolution offered is a size swap, not a refund, and the second purchase is marketed by size and category rather than by garment. A customer whose size is known in your brand is a different economic proposition to a new follower, and most accounts in this category never separate the two.

Rules that constrain this category

Policies and regulations change. Treat this as the starting point for your own check, not as legal advice.

  • Clothing itself is not a restricted category, and it is worth being plain about that. Apparel appears on none of the major prohibition lists: not Meta's twenty-four prohibited commerce categories, not TikTok Shop's US prohibited categories, not Snap's merchant prohibitions, not Pinterest's merchant guidelines. The hard line is authenticity, not clothing. Snap's merchant policies name "Counterfeit, fake or bootleg products, or replicas or imitations of designer products". Pinterest prohibits counterfeit goods. Counterfeit goods heads Google's prohibited content list for both Ads and Merchant Center. TikTok's Branded Content Policy bars promotion of counterfeit products. Dupe marketing lives right beside that line, and what crosses it is usually the caption rather than the garment: naming the brand you are imitating, in text or on screen, converts a design similarity into a trademark complaint a rights holder can file. Write the copy so the garment stands on its own without the comparison.
  • Secondhand and vintage is where organic selling and paid advertising diverge hardest, and resale sellers get caught by this constantly. TikTok Shop's US policy lists Secondhand Products among its prohibited categories. Pre-Owned Products covering luxury bags, luggage, accessories and footwear sit behind category-level qualification, meaning documents submitted through the Qualification Center, and Pre-Owned Luxury Apparel, Jewelry and Watches is invite-only, so a seller cannot self-apply into it. Pinterest splits the other way: pre-owned products are not allowed as organic product Pins, but Pinterest states shopping ads may include pre-owned products, excluding named categories such as cosmetics, personal care, baby products and safety gear. The same inventory can therefore be unlistable and advertisable on one platform at the same time. Plan resale around live marketplaces and your own checkout, and read the current prohibited-items detail inside each platform's seller policy for your own market rather than the top-level category name, because the detail is where the exclusions live.
  • Kidswear carries two layers, and the legal one is heavier than the platform one. TikTok Shop US puts Kids' Fashion behind category-level qualification, and handles designated Baby and Maternity items, sleepwear among them, at product level instead. Underneath that sits US product safety law. Children's sleepwear must meet flammability standards at 16 CFR part 1615 for sizes 0 through 6X and 16 CFR part 1616 for sizes 7 through 14, and the Children's Product Certificate has to cite the applicable part. CPSC treats hood and neck drawstrings on children's upper outerwear in sizes 2T to 12 as a substantial product hazard under 16 CFR part 1120, and it also reaches waist and bottom drawstrings in sizes 2T to 16, which must not extend more than three inches outside the channel when fully extended, must carry no toggles or knots, and must be bar tacked if continuous. The second half is the part people miss, because they only ever heard about hoods. Selling kidswear socially without the certification behind it is a recall exposure, not a policy exposure, and no platform appeal fixes it.
  • Labelling and origin rules reach into your marketing copy, not just your neck tags. Under the FTC's textile guidance a label must disclose fibre content, country of origin and either the company name or a Registered Identification Number, and the guidance is explicit about advertising: when a textile or wool product is advertised in a catalog or other mail order promotional material, including on an internet site, the description must include a clear and conspicuous statement that the item was either "made in U.S.A.", "imported" or "made in U.S.A. and [or] imported". That covers a product page and, read plainly, any post carrying the offer. In Canada the Textile Labelling and Advertising Regulations require fibre content information to appear in both English and French, and dealer identity as a name and postal address or, in its place, a CA identification number. Unqualified Made in USA claims sit under the FTC's Made in USA Labeling Rule at 16 CFR part 323 and its all or virtually all standard, with a textile-specific wrinkle worth knowing: for textiles the FTC ties the claim to the final product being manufactured in the US from fabric manufactured in the US, rather than tracing every earlier input. If you private-label blanks and cut the supplier's tags out, these duties become yours.
  • Fibre naming has a long enforcement paper trail, and apparel is where it happens. The FTC warned seventy-eight retailers in February 2010 to stop labelling and advertising rayon textile products as bamboo. It brought penalty actions against four national retailers in January 2013, against Nordstrom, Bed Bath and Beyond, Backcountry.com and J.C. Penney in December 2015 on the same claim, and in April 2022 it used its penalty offense authority against Kohl's and Walmart for bamboo marketing. Four rounds over twelve years, one claim. If your caption says bamboo and your spec sheet says rayon, you are inside a pattern the regulator has already litigated. Green claims are tightening in parallel, and this part is region-specific and moving: the EU's Empowering Consumers Directive amends EU consumer law to restrict vague environmental claims and climate-neutrality claims resting on offsetting, and it lands through national transposition on a timetable running through 2026, so confirm the position in each market you sell into rather than working from one date. France has legislated separately against ultra fast fashion, including advertising restrictions that reach influencer promotion, and the text and its commencement have shifted repeatedly through the process, so verify the current French position before briefing creators for a French audience. Write the caption against the composition, not the brand story.
  • Creative rules for swim, intimates and shapewear differ enough between platforms that one asset set will not clear all of them. Meta's advertising standards put ads depicting people in revealing clothing or sexually suggestive poses, or focused on individual body parts such as groin, buttock or breasts, behind eighteen plus targeting. TikTok's adult content ad policy goes further and lists "Swimsuits, lingerie, or clothing that exposes or implies exposure of intimate areas" among prohibited sexually suggestive visuals, so swim creative that runs on Meta behind an age gate can fail outright on TikTok. Shapewear and anything framed around slimming is stricter again: Meta's health and wellness rules prohibit statements of inferiority about physical appearance and close-up pinching of body areas, and Pinterest does not allow weight loss language or imagery at all, with a narrow carve-out for approved healthcare advertisers in the US and Canada promoting prescription GLP-1 medication. Snap's merchant policies prohibit products that require age gating outright, which makes its merchant surface narrower than its ads surface for this inventory. One caution across all of it: these are written policies, and automated enforcement is frequently stricter than the text, so plan for rejection and appeal rather than assuming a written allowance is what you will experience.
  • Gifting is endemic in fashion, which makes disclosure a category risk rather than a legal footnote. Under 16 CFR 255.5 a connection that might materially affect the weight or credibility of an endorsement, and that the audience would not reasonably expect, must be disclosed clearly and conspicuously. Free or discounted product counts, and so does the mere possibility of payment. The FTC's clear and conspicuous standard says that in an interactive electronic medium the disclosure should be unavoidable, which is what defeats a label a viewer can scroll past, and for video it belongs in the video rather than only in the description. The FTC also states that a platform's built-in disclosure tool might not be an adequate disclosure on its own. Accepted wording is not uniform across markets, and this is where multi-market fashion accounts trip: the FTC accepts "sponsored", while UK guidance lists "#spon" and "#sponsored" among terms to avoid and the ASA prefers "Ad" placed prominently upfront, and Canada's Ad Standards guidance names "#Ad" used on its own as the gold standard. One caption template across the US, the UK and Canada is wrong in at least one of them.

Selling this without owning it

If you want to sell in this category without making the product, these are the routes that exist. The sourcing directory covers each in more detail.

Print on demand The route for design-led apparel where the graphic is the product rather than the cut. Printful and Printify both sit under the FYUL parent brand launched in November 2025, and both brands continue to operate. Amazon Merch on Demand is application-gated with tiered upload capacity, starting at ten live designs and advancing as sales accumulate, so the approval queue and the tier ceiling are the bottleneck rather than your design output. Redbubble and TeePublic, both under Articore Group, hand you marketplace traffic in exchange for the customer relationship. Check the graveyard before you build: Amaze has published that Spring stores stop accepting new orders on 15 September 2026 and that the remaining Spring Creator Dashboard is fully disabled on 15 October 2026, so any workflow routing you to Spring is dead. The trade-off specific to this category: you inherit the blank's fit and the printer's dispatch time, so gather your own fit data on the blanks you use, and never repeat a supplier's shipping estimate to a customer as a promise. If you re-tag garments as your own brand, the fibre, origin and care labelling obligations become yours.
Apparel dropship suppliers Fashion-specific suppliers rather than general catalogues. Trendsi is a US-based supplier led by women's fashion, offering dropship, open-pack wholesale and made-to-order production, with Shopify and TikTok Shop integrations. FashionGo is an online wholesale apparel marketplace aimed at retail buyers. Beyond fashion-specific options, CJdropshipping is free to use with global warehouse choices, Spocket and Syncee are curated marketplaces emphasising vetted US and EU suppliers, and AliExpress is reached through DSers now that Oberlo closed in 2022. The catch is sharp in apparel: you can set retail price and own the customer, but you cannot control fit, and fit data is the actual product you are selling. Order samples in multiple sizes and measure them yourself. Treat every published delivery window from these suppliers as marketing copy rather than measurement, and do not pass it through to a customer.
Wholesale and open-pack buying The route when you want pricing power and can hold a size curve. Faire is Shopify's recommended wholesale marketplace for its merchants, with Net 60 terms available by approval, accounts reviewed for terms once a first order is placed, payment falling due sixty days after the order date or sixty days after shipping when an order takes more than fourteen days to ship, and free returns applying only to a first order with each brand. Ankorstore is the principal European equivalent, with deferred payment terms. Watch the graveyard here too, because stale source lists are everywhere: Creoate entered administration in September 2026, Tundra announced its wholesale marketplace shutdown in 2023, and Shopify's Handshake was retired around October 2023. On the plus side, Shopify's native B2B features are not Plus-only, so a wholesale motion is available on lower plans.
Affiliate and creator commerce The route with no inventory and no pricing power, where the trade-off is that commissions can be reversed. Start by deleting the dead entries from whatever list you are working from: Awin absorbed ShareASale, which closed on 6 October 2025; Shopify's Help Centre states Collabs is not accepting new creator signups; and TikTok One replaced TikTok Creator Marketplace, which shut down on 1 April 2025. What works now: LTK is application-based and now gates access to retailer ambassador tiers, since Target Ambassadors is built on LTK's infrastructure. ShopMy is reported to be open signup with no follower minimum. Mavely became Later on 1 September 2026, with existing mave.ly links continuing to earn. Amazon Associates requires three qualifying sales referred within 180 days of applying, and the separate Amazon Influencer Program unlocks onsite commissions only after at least three videos meeting Amazon's content guidelines. Etsy's Creator Collective Policy sets a minimum of 500 followers on at least one authorised channel. eBay Partner Network suits vintage and sneakers, but note commissions are a share of eBay's earnings on the transaction rather than of the item price. Two constraints to plan around: LinkedIn prohibits affiliate advertising outright, and TikTok Shop's US affiliate programme requires being based in the United States with at least a thousand followers, so it is not a route for Canadian creators.
Consignment and live resale The route for people whose skill is sourcing rather than manufacturing: thrift, estate, deadstock and wardrobe clear-outs. Whatnot, Poshmark, Depop, Vinted and Grailed each attract a different buyer, and the practical read is that Depop and Vinted sit closer to everyday fashion while Grailed sits closer to menswear and streetwear. Live auction formats suit one-of-one inventory that paid media cannot carry economically. The compliance note above applies in reverse here: because several native shop surfaces will not accept pre-owned listings, the resale motion depends on marketplaces and on your own checkout, and your social accounts do the discovery work rather than the transaction. Cost your sourcing labour in per garment, because it does not amortise the way a restock does.

What to measure

  • Revenue net of returns, by style and by size, as the headline number. Gross sales in apparel describe a sale that has not finished happening yet. Track it per size and the size that fails becomes a spec problem you can fix rather than a marketing problem you cannot.
  • Return reason split at the size level, specifically too small against too big. When one size skews consistently in one direction, the fix is the size chart or the grading, not the creative. Sellers who ship a copied supplier chart never see this, because they never collect the data.
  • Exchange to refund ratio. An exchange keeps revenue and a size fact. A refund loses both. Making exchange the default first offer is one of the cheaper margin interventions available in this category, and the ratio tells you whether the offer is actually landing.
  • Saves and shares per view as the leading indicator, ahead of likes and ahead of clicks. Apparel is bought later than it is discovered. A save is a buyer parking a garment until payday or until an occasion arrives, and rising saves predict revenue that has not appeared yet.
  • Fit questions per post, counted rather than answered ad hoc. Volume is a demand signal, and the specific questions are a content brief. When the same question repeats, the answer belongs on screen in the next video and in the listing, not in a hundred individual replies.
  • Size-curve sell-through inside the first week of a drop, plus account health where it gates distribution. Broken sizes kill a style's conversion before the style stops trending. On TikTok Shop, seller-side health is a growth metric rather than hygiene: affiliate collaboration requires a Shop Performance Score of at least 3.5, and creators need a Promotion Performance Score of 3.5 or higher and a Creator Health Rating of at least 150. Read your own Seller Center for the current return and refund thresholds in your category rather than trusting a figure from a blog post.

The mistakes that cost the most

  • Shooting for a catalogue in a category judged on movement. Flat-lays, studio stills and mannequin shots are cheap to produce and they lose to a fifteen second clip of someone sitting down in the garment. The related error is filming everything on one body, then absorbing the returns from every buyer who is not shaped like that body.
  • Publishing the supplier's size chart without measuring a sample. Grading drifts between production runs and between factories, and the chart you inherited is often aspirational. That chart is your return rate written down in advance, and correcting it after a launch means correcting it in front of customers who already guessed wrong.
  • Scaling spend on a style before the size curve is in stock. Paying to send traffic to a page where the middle sizes are sold out produces the worst of both outcomes: you fund the demand and a competitor fills it. Pull budget when sizes break, not when the trend cools.
  • Treating returns as a customer service function rather than a revenue function. Defaulting to refund, not recording the size the customer bought, and not offering the swap first means the same customer guesses again from scratch next time. The size worn field in your customer record is worth more than most of your attribution stack.
  • Building on a surface that has been retired or was never available in your market. Instagram Live Shopping ended in March 2023, Facebook Live Shopping closed in October 2022, Meta's current developer documentation describes offsite checkout as the model for Shops ads, and TikTok Shop is not available in Canada. Canadian sellers sometimes reach for US operator partnerships or bought shop accounts to get around the last one. That route risks the account itself alongside tax and consumer protection exposure, and it is not a foundation to build a business on.
  • Writing the sustainability and fibre copy in marketing rather than against the spec sheet. Bamboo describing rayon is a claim the FTC has pursued against apparel and textile retailers in four separate rounds. Vague eco or sustainable copy aimed at EU buyers is tightening through national transposition of the EU's Empowering Consumers Directive, so check the market rather than assuming a date. A Made in USA line on a garment turns on a textile-specific test tied to fabric manufactured in the US, which is not the test most sellers think they are meeting. The same failure mode covers naming the designer brand your piece resembles, which turns a dupe into a trademark complaint and eventually into a closed account. In every case the copy was written by someone who never read the spec sheet.

In apparel the sale is made the moment a buyer stops guessing about fit, and the profit is kept only if they never have to guess twice.