Most of a dropshipping store’s sales come from Instagram and Facebook ads. The reason is simple: nobody searches for these products. People don’t type “tablet stand for the kitchen” into Google — they see the product in their feed, understand its benefit and buy it right there. In other words, the ad doesn’t meet demand here; it creates it.
Three things determine the results of Meta ads in dropshipping: the creative (at least 70%), the offer and accurate measurement. Audience selection, campaign naming and budget strategy are the rest. If the creative is weak, no targeting will save it.
If you’ve already chosen your product and built your store, you’re now at the traffic stage. For the big picture of the model, read what dropshipping is.
Three things to set up before running ads
These must be ready before you launch a campaign. If they aren’t, your ad money goes somewhere you can’t measure.
- Pixel and Conversions API. Working with the browser pixel alone is no longer enough — some orders don’t reach the report and the account learns the wrong things.
- Conversion events. At minimum: page view, view content, add to cart, initiate checkout, purchase. Each must be checked in test mode.
- Domain verification and event priority. If you skip this step, most sales from iOS users won’t be visible.
Getting this setup right matters more than the ad budget itself — because with wrong measurement you can switch off the winning product and scale the losing one. If you need help on the technical side, see our analytics and tracking service.
Campaign structure: the simpler, the better
The most common beginner mistake is overcomplicating the account: 8 campaigns, 20 ad sets, 50 AZN budget in each. As a result, no ad set finishes the learning phase. A simple structure looks like this:
| Stage | Campaign | Budget logic |
|---|---|---|
| Test | 1 campaign · 1–2 ad sets · 3–4 creatives | 20–30 AZN a day, 3–5 days |
| Validation | Keep the winning creative in a separate ad set | 30–50 AZN a day |
| Scaling | Increase the budget in steps of 20–30% | Continue as long as results hold |
| Retargeting | A separate campaign: site visitors, add-to-carts | 10–15 AZN a day is enough |
The objective (optimisation goal) should almost always be purchase. Optimising for clicks or views shows cheap numbers but doesn’t bring sales — the account brings you exactly what you ask it for.
The creative — 70% of the ad
In dropshipping, the winning ad is almost always a video, and its structure works in this order:
The first 2 seconds — a thumb-stopping shot
Show the problem moment or an unexpected image in the very first second. A video that opens with a logo is dead in practice.
Seconds 3–7 — the problem
The viewer should recognise themselves: “yes, that happens to me too”. Showing the problem is stronger than talking about it.
Seconds 8–15 — the solution
Show the product working: it opens, folds, cleans. Real use, not a catalogue photo.
Seconds 16–22 — the proof
A review, a before/after shot, a counter. In the local market, a real user video is the strongest proof.
The last 3 seconds — a clear call to action
“Tap the link”, “Add to cart”. If there’s a discount, say it here.
Prepare at least 3 different creatives for one product: one based on the problem, one on the result, one on a review. Using the supplier’s catalogue video as it is is the fastest way to lose money — dozens of stores run the same video. If you can’t set up creative production yourself, see our ad creatives service.
Audience: broad or narrow?
A few years ago, testing with 20 different interest groups was standard. Now Meta’s algorithm learns who to show the product to from the creative. The approach that works in practice:
- Start with a broad audience. Just country, language and a wide age range. Give the algorithm room to work.
- Use interest groups only if the product is very specific. For example hunting, fishing, baby products.
- Don’t narrow the audience too much. In Azerbaijan an audience of 30–50 thousand people runs out in a few days and results get worse.
- Build lookalike audiences only after 50+ sales. Before that there isn’t enough data.
Retargeting: the cheapest sales are here
95% of visitors don’t buy on their first visit. A retargeting campaign brings exactly them back, and the cost per sale is 2–4 times cheaper than cold traffic. The minimum setup:
- People who added to cart but didn’t buy (7 days). The most valuable group — a discount or free delivery offer works here.
- People who viewed a product but didn’t add it to cart (14 days). Show them a review and a proof video.
- People who watched 50% of the video. A cheap and broad group.
Which numbers should you look at?
| Metric | What it tells you | When to worry |
|---|---|---|
| CPM (cost per 1,000 impressions) | The level of competition in the auction | If it’s risen sharply — the audience is running out |
| CTR (click-through rate) | The strength of the creative | Below 1% — the creative is weak |
| Add to cart / clicks | The strength of the product page | Clicks but no add-to-carts — a page problem |
| CPA (cost per sale) | The main decision number | Above break-even CPA — you’re losing money |
| ROAS | The revenue the ads return | Below 2 can be dangerous, depending on margin |
I showed how to calculate break-even CPA with numbers in the product selection post. In short: subtract the product, shipping, fees and returns reserve from the selling price — what’s left is the maximum ad cost you can pay for one sale.
When to switch off, when to scale?
- Switch off: if in 3 days you’ve spent twice the product’s price with no sales.
- Fix the page: if there are add-to-carts but no sales — the problem isn’t the ad but the checkout flow.
- Change the creative: if CTR is below 1%, don’t touch the targeting — change the video.
- Scale: if CPA is at half of break-even, increase the budget by 20–30% a day. Doubling it in one go breaks the learning.
- When the audience runs out: if frequency goes above 2.5, you need a new creative.
The ad account doesn’t sell a bad product — it just shows faster that the product is bad.
The 6 most common mistakes
- Launching ads before measurement is set up. If the numbers are wrong, every decision is wrong.
- Deciding after one day. The account needs at least 3 days to learn.
- Changing the budget often. Every change resets the learning.
- Testing with one creative. You can’t tell whether the failure is the product or the video.
- Optimising for clicks. A cheap click doesn’t mean a sale.
- Not setting up retargeting. That means leaving your cheapest sales at the door.
How much budget do Meta ads for dropshipping need?
To test one product properly, you need 20–30 AZN a day, or 100–150 AZN over 3–5 days. If you’ll test 3–5 products, a realistic starting ad budget is between 500 and 900 AZN.
Instagram or Facebook — which works better?
Both are managed from the same account, and at the start it’s better to keep placements automatic. The algorithm finds on its own which platform gives cheaper sales. If your audience is young, in practice most of the budget flows to Instagram and Reels.
Which objective should you choose for ads?
The Sales (conversion) objective with the “purchase” event. Traffic or engagement objectives bring cheap clicks but no sales.
How many creatives should you prepare?
At least 3 different videos for one product: based on the problem, the result and a review. Even after you find a winning creative, you need a new one every 2–3 weeks, because the audience gets tired of seeing the same video.
The ads are running but there are no sales — why?
Clicks but no add-to-carts — the product page is weak. Add-to-carts but no sales — there’s a barrier in the checkout flow. No clicks at all — the creative doesn’t stop attention. In all three cases only properly set-up measurement gives you the answer.
What should I do if my ad account gets blocked?
The most common causes: health claims, brand copies, banned “before/after” style shots, or broken website pages. Fill in the return and privacy pages, tone down bold claims and don’t run the same product from more than one account.
Conclusion
Meta ads in dropshipping aren’t a magic button — they’re a fast testing environment. If your creative is good and you see the numbers correctly, you learn within a week whether a product works. If measurement is wrong, you burn money for months and draw the wrong conclusions.
If you want your campaigns launched with a professional setup, see our Meta Ads service page or go straight to the get a quote form.
