Thin margins die in the ad auction.

Yes, but only at specific price points and order values. A single $24 tee with a $12 base cost almost never survives Meta or TikTok economics, because break-even ROAS lands around 4x. The same shop selling a $68 personalized three-piece set with a 38% contribution margin has a break-even near 2.6x, which paid traffic can hit consistently. Our first job is telling you which side of that line your catalogue sits on, and we will say so plainly if the answer is the wrong side.

We ask for a minimum of $4,000 in media over the first thirty days, and $6,000 to $10,000 is where testing gets genuinely informative. That budget is split across a shortlist of products and a batch of creative angles, with a fixed spend ceiling per test. Below roughly $3,000 the data is too noisy to separate a bad product from an unlucky week, and we would just be spending your money to generate opinions. If your budget is under that, take the teardown first and come back when the media is ready.

Our in-house team produces the advertising creative: statics, mockup motion, UGC-style video, hook variants and landing visuals. Product artwork can come from your existing library, your own designers, or our team when you need new directions to test. Every asset we make is yours, delivered in source format, and we keep a documented angle library so you can see which hooks won and why. We do not use your competitors' artwork and we do not touch designs we cannot verify the rights to.

On the defensive side, we screen your catalogue for trademark and copyright exposure before we put media behind anything, because a takedown mid-scale is far more expensive than a design change. On the offensive side, we assume you will be copied and we build for it: personalization, bundle exclusives, owned audience lists and a rolling pipeline of next products. We can support your platform takedown filings with documentation, but we are not attorneys and we do not give legal advice. The real protection is an offer that is structurally hard to clone, not a report to Etsy.

Both, with different jobs. Marketplaces like Etsy and Amazon give you cheap intent and fast validation signal, so they are excellent for discovering which designs and niches have real pull. Your own storefront gives you margin, customer data, retention and an asset that compounds, so that is where paid traffic and personalization live. Running them without one shared pricing logic is how sellers end up cannibalizing their own listings, so we set the channel roles first and then let marketplace data feed the storefront testing queue.

Realistically, three to seven weeks from the start of testing. The first two weeks are teardown and offer rebuild, then testing runs in waves, and a genuine winner usually surfaces in the second or third wave. Some accounts find one in week three; some need two full cycles because the niche is crowded or the fulfilment cost caps the price. We report progress against test thresholds every week, so you never have to guess whether the program is working while you wait.

It happens, and we tell you rather than quietly extending the engagement. If a full testing cycle produces no product clearing threshold, we review whether the constraint is the product category, the fulfilment cost base, the price ceiling of the niche or the creative angles, and we change one variable deliberately instead of all four at once. Sometimes the honest conclusion is that the niche cannot support paid acquisition at your cost structure, and we will recommend a different category or a different channel. Results vary between sellers and nothing here is a guarantee of revenue or profit.

You send the application form on this page and we review it within one business day. If there is a fit, we book a forty-minute call to walk through your costs, your current spend and your top products, then send a scoped proposal with the engagement tier we recommend. On acceptance, we request read access to your ad accounts, store analytics and fulfilment invoices, and the teardown starts the same week. If there is no fit, we say so on the call and point you at what to fix first.