I scraped 10,000 products to find gaps nobody was talking about

I wanted to know what was actually selling on TikTok Shop vs Amazon — not what the gurus were saying was selling, not vibes, not "this niche is HOT right now." Actual data. So I built a scraper, pulled 10,000 products, loaded it all into a local SQLite database, and started looking. What I found was not what I expected.

I'll be honest, I went in thinking I'd find some obvious thing — like a product category that's popping on TikTok that Amazon sellers weren't touching yet. And there were a few of those. But the more interesting thing was the gap in the other direction: products with strong Amazon search volume and decent reviews that had almost no presence on TikTok Shop at all. Categories where the demand is clearly there, but nobody has made the short-form video content to bring new buyers in. That's a different kind of gap and in some ways a less risky one because you know the demand already exists.

Now here's the part people are going to want to skip past — I didn't need a data science degree to do this. I didn't write complex ML models or do anything that required more than some basic SQL and a willingness to look at spreadsheets for a few nights. The tool that did the heavy lifting was Apify, which handles all the scraping infrastructure so I don't have to maintain proxies and figure out why Amazon blocked me again. I wrote the queries. I looked at what came back. That's it.

Data is intimidating to people who've been told it should be intimidating. Like, somewhere along the way the message became "you need a specialized degree and a $200/month tool stack to understand what's selling." You don't. You need a way to collect the data, a place to store it, and enough patience to sit with it and look for patterns. SQLite is free. SQL is learnable. The scraping tool costs less than a dinner out.

The bigger shift for me was moving from "I think this product has potential" to "here's what the data says about this product." Those are very different decisions and they feel different too. One is a bet, the other is a bet you placed on purpose with information. Both can be wrong, but one of them you can actually learn from because you wrote down what you expected to happen.

I built Versus because I wanted to have this conversation with my own data on a regular basis, not just once. Price alerts when a product I'm watching drops. Trend detection across categories. Gap analysis across platforms. It's not a public product yet — it's a tool I use — but this is what building for yourself looks like. You make the thing that scratches your itch and then you decide later if you want to hand it to other people.

Apify. $29 a month and it has scrapers for Amazon, TikTok Shop, Alibaba, and a bunch of other platforms that would take you weeks to build yourself and maintain. You run an actor (that's what they call their scrapers), pipe the results to a dataset, pull it down as JSON or CSV, load it wherever you want. I load it into SQLite and query from there.

The free tier lets you poke around and understand how it works before you commit to anything. But if you're serious about product research, $29 is genuinely worth it. You're not paying for data, you're paying for the infrastructure that goes and gets the data for you so you can focus on what the data actually means.

If you want to go deeper on this stuff — the building, the data, the what-actually-works-in-practice conversations — come hang out at Build con Chispa. It's my community on Skool and it's where I share things I don't put anywhere else. skool.com/build-con-chispa-9733.

Yesenia M. Perez Lead Backend Engineer. Mom of 3. Building at 11pm.

yessieperez.com | IG @yeseniavperez | TikTok @iamyessieperez | LinkedIn /yessiemalone