I got tired of burning money on saturated dropshipping products, so I scraped 1.3M SKUs to do the math before spending $1 on ads.
Hey guys.
I just wanted to share something I've been working on because I was honestly sick of the whole "test and pray" method. We've all been there: you see a product that looks like a winner, you spend $200 on Meta ads, and then you realize the landed cost eats all your profit or 50 other stores are already scaling it.
I got so fed up that I spent the last few months coding a script to do the dirty work for me. Instead of manually looking for products, the tool scrapes the CJ Dropshipping catalog (around 1.3 million products) and calculates the true landed cost by weight. Then it automatically checks the Meta Ad Library.
But here is the trick I realized: looking at active ads is useless. My script looks for abandoned ads. If 5 stores are running it but 8 stores tried it and killed the ad after 19 hours, the script flags it as a trap and kills the product.
I let it run today targeting the Netherlands just to see what would happen. It spit out a weird product: a full-color Pomeranian dog chef apron.
The math it calculated was actually insane. The shipping cost to NL is $7.27, and based on competitor pricing it suggested a retail price of €94.95. That leaves a net margin of around 71%. It even calculated my exact break-even CPA before I even opened the ads manager.
It went a bit overkill and even generated the Dutch copy for the landing page using the supplier photos (I attached a screenshot of what it outputted).
I’m still tweaking the code and the servers are costing me a bit to run, but the amount of time it saves is ridiculous. Has anyone else tried building custom scrapers for product research instead of paying for the typical spy tools?
I have some server credits left for the month, so if anyone is struggling to find products for Q4 and wants me to run their specific niche through my script, just shoot me a DM and I'll send you whatever it finds.