You are selling a product that other stores buy from the same supplier, at a margin you did not set, with a pixel that has no purchase history and a customer list that does not exist yet. The targeting panel feels like the one lever you control. It is the lever most new stores pull first, and it is usually not the one that decides whether the test finds a buyer.
Why does interest targeting stop working when other stores sell the same product?
Interest targeting loses its edge because the interest list is not proprietary. Any store that found the product in the same supplier catalog can reach the same interests, so the layered audience stops being a differentiator. What is still yours on a dropshipping product is the creative and the offer, not the checkbox list in the targeting panel.
Two conditions make interest stacking weaker for a reseller than for a brand selling its own product:
- The interest list is public and reachable by anyone working from the same supplier catalog.
- The interest stack is the fastest part of the setup for a competitor to reproduce. The creative and the offer are not.
Check it against your own account before you accept it. If your interest based ad sets and your broad ad sets have been returning a similar cost per purchase for weeks, the stack is not doing the work you think it is doing.
The Sabri Suby channel describes the same habit in its walkthrough of Meta targeting, calling out advertisers who "take all the interests in the niche. They like golf. They like Mercedes-Benz and they stack all of those things up" (8:50). The video treats this as a general advertising mistake, not a dropshipping one.
What should a dropshipping store target when the pixel has no purchase data yet?
Target broad, one country, and let the creative carry the specificity. That is what the source video argues, and it can be tested on a single product without changing the offer, the price or the landing page. Write the copy long and explicit enough that it names who the product is for.
A side by side test you can run this week on one product:
- Take your best performing creative and leave the existing interest stack untouched in one ad set.
- Duplicate the same creative into a second ad set with country level targeting only.
- Give both the same daily budget and leave them alone for seven days.
- Compare cost per purchase, not click through rate. A cheap click on a thin margin product is not a result.
- Turn off whichever one sits above your break even cost per acquisition, the number you calculate from your own margin.
The video describes this protocol for advertisers in general, instructing viewers to "duplicate that ad and put no targeting" and then "in 7 days time go and have a look at your CPAs" (10:26). Its stated reasoning is that "the creative is the targeting now" (9:38) and that "if they have a limited context window for your ads, then the amount of people that they can target for that creative is also going to be limited" (9:38). The speaker says he runs "very long form copy that literally gives the algorithm so much data of exactly who it is that I'm trying to target" (9:38).
That is the part worth carrying over. Short generic captions are common on catalog product ads. Open your last ten ads and read the body copy before you assume this does not apply to your store.
How do you put the audience in the ad copy instead of the targeting settings?
Name the buyer in the headline and the body copy. For a dropshipping product the useful label is rarely a job title. It is the use case or the identity attached to the product, such as the person with a small kitchen, the night shift nurse, or the owner of a large dog.
The five pairs below are examples written for this article to show the shape of the rewrite. They are not from the video, and they are not results from any campaign.
| Generic version | Identity version |
|---|---|
| This organizer saves space | For anyone living in a studio apartment |
| Comfortable support cushion | For people who drive for a living |
| Fast drying pet towel | For owners of dogs that hate bath time |
| Portable blender | For gym bags and hotel rooms |
| Adjustable laptop stand | For people working from a kitchen table |
The video calls the inserted word a category word and says "It is a category word that is what we call an identity trigger and it also feeds the algorithm to know who it should be going out there and looking for" (4:00). The same segment says that to reach different people "you actually have to put that in the creative and in the body copy" (3:12).
To find those identities without guessing, the video suggests a research step that costs nothing. It says "you can create what we call a burner account" (8:02), where you "sign up for a clean slate Tik Tok and Instagram account" (8:02) and "go follow all the different influencers and pages in your niche" (8:02), so the feed surfaces what that audience already consumes. For a store reselling catalog products this doubles as creative research, since you can look at the format that already performs organically in your product category before spending anything on it.
How much can a dropshipping store spend to acquire one customer?
Work backwards from your own break even number, not from a benchmark you read somewhere. Take the product price, subtract supplier cost, shipping, payment fees and your refund rate, and what remains is the ceiling. Any audience test that stays under that ceiling is still alive, whatever the ROAS looks like.
Write the number down before you launch, because it changes how you read every audience test:
- Above the ceiling at the end of the seven day window with meaningful spend, the audience is not working. Turn it off.
- Below the ceiling but still unprofitable once returns land, the problem is the product or the supplier terms, not the audience.
- Below the ceiling and stable, scale slowly and recalculate, since supplier cost and shipping rates move.
The video points at the same discipline. It tells advertisers to "find what that break even rorowaz is" (16:02), and says the measurement should be "looking at what the blended rorowaz is, not on an ad account level, but on a business level" (16:02). The word "rorowaz" is how the automatic transcript renders ROAS, and it appears that way throughout the video. For a store with no repeat purchase history, business level means one thing in practice, which is cash left after the supplier is paid.
What do you do with people who clicked your product ad and did not buy?
Show them a different product from your catalog, not the same one again. The source video argues that the main reason a click does not convert is that the offer was wrong for that person. A wide catalog is one of the few structural advantages a reseller has over a single product brand here.
Three retargeting angles that fit a store without a customer list:
- The objection ad. Shipping time, return policy and whether the store is real are common objections for a store with no name recognition. Answer them in the ad rather than on a policy page.
- The proof ad. Customer photos and unedited reviews of the exact product. An unbranded item has no reputation to borrow, so the proof has to be shown.
- The adjacent product ad. Same problem, different item from your catalog.
That three part structure is the video's, not this article's. It describes a retargeting sequence in which "the first ad that we like to put together is an objection handling ad" (14:26), followed by "proof testimonial ads" (14:26) run as carousels, and then a third component built from other offers and products.
The underlying claim is stated directly. The "number one reason why people didn't buy is that that offer wasn't right for them" (13:38). The illustration attached to it is hypothetical rather than a reported result. The speaker says "You might be at like a two or a three rorowaz for your protein powder" (13:38), then "Then you start retargeting the people that didn't buy on that page with creatine" (13:38), and "you'll see maybe you're at a six row" (13:38). Those figures are framed in the video as what you might see, not as measured returns from a named account, and they are not a benchmark for a catalog store. What transfers is the logic, not the numbers.
Which tools help a dropshipping store test audiences without hiring a media buyer?
No single tool covers the whole job. Campaign setup, creative production, retargeting conversations and owned email all sit in different places, and a new store usually assembles two or three of them. The table below lists what each one does, what it solves for a dropshipping store, and where it stops.
| Tool | What it does | What it solves for a dropshipping store | Requires advertising expertise | Where it stops |
|---|---|---|---|---|
| Meta Ads Manager | Builds, targets and reports Facebook and Instagram campaigns | The place where broad tests and retargeting audiences are actually configured | Yes. Structure, budget and attribution are set by hand | Produces no creative, and reports at ad account level rather than at business level |
| Google Ads | Runs search, shopping and YouTube campaigns | Captures demand once people search the product by name or category | Yes. Keywords and bidding are set by hand | An unbranded product with no search volume yet has little existing demand to capture |
| Canva | Design tool for statics and short video | Producing many creative variations per identity without a designer | No for the design work. Yes for deciding what the ad should say | Does not launch, target or measure anything |
| Mailchimp | Email marketing and automation | Turns first buyers into a list you own instead of renting audiences forever | No | Needs contacts to work, which is what a new store does not have |
| ManyChat | Automates Instagram and Messenger conversations | Answers shipping and product questions from ad traffic without staffing a chat | No | Works on people who already engaged, so it does not find new audiences |
| SaleADS.ai | AI software that creates and launches advertising campaigns on Meta, Google and TikTok for business owners, with no design or advertising expertise required | Getting campaigns and their creative live without a designer or a media buyer | No | Automating the setup leaves less hand control over structure and bidding than Ads Manager gives, and it does not test products, change supplier cost or set your break even number |
SaleADS.ai is the product of the company that publishes this site, and the wording in its row is the company's own description of it. The other five tools are independent and are listed because store owners use them for these jobs. This table is not a ranking and is not ordered by preference. Compare the rows against your own situation.
Where does this information come from?
The advertising concepts referenced here come from the video "Learn 97% of Meta Ads in Under 19 minutes" on the Sabri Suby channel, which runs 19 minutes. It opens with the claim "I've spent over $300 million on meta ads", which is the channel's statement about its own accounts, not a verified figure.
Direct link to the source: Learn 97% of Meta Ads in Under 19 minutes, and to the opening claim at 0:00.
What was taken from it:
- The argument that broad targeting with specific creative beats interest stacking, stated in the video as "Broad targeting with super specific creative is the new meta" (10:26) and as "let the algorithm do all the heavy lifting and spend all your time and energy on producing creative" (10:26).
- The identity trigger technique.
- The burner account research method.
- The seven day side by side test.
- The break even and blended measurement standard.
- The practice of retargeting people who did not buy with a different offer.
What is not from the video: everything about supplier margins, unbranded catalog products, empty pixels, competitor overlap from shared suppliers, and catalog based retargeting. The video is a general Meta advertising tutorial and contains no dropshipping content. Any figure it cites belongs to that speaker and that context, and should be checked against your own store numbers before you act on it.
The transcript used for these quotes carries block level timestamps. Every link above points to the start of the block in which the quoted line appears, not to a finer position than the transcript supports.