A lot of sellers blame the product team when things don't sell, and the product team blames the operators. This happens everywhere. But honestly, operators should learn how to pick products. It's not just the product team's job. If you don't know what makes a good product for your own skills, you're basically gambling.

Here's my framework for operators who want to choose products that actually have a chance.

First, understand your own capabilities

Operator skills come in two parts:

  • General skills – writing listings, PPC, optimization – things you can apply anywhere.

  • Category‑specific experience – deep knowledge of a product niche, knowing how customers search, what images work, how ads behave in that category.

A great operator in one category might struggle in another. So don't assume you can jump into any category and win. Stick with what you know, or be very careful.

How to choose a category (from an operator's perspective)

1. Start with what you already know and have succeeded in.

Copying a winning formula is the fastest and most reliable path. Yes, markets change, but you're still reducing unknowns.

2. Improve an existing product you've sold before.

Take customer feedback and add a new feature, better material, or slight design change. This is often the best option – you avoid price wars and customers already understand the product.

3. Go for similar or complementary products.

If you know pet beds, try pet carriers. It's not as safe as sticking to the exact same product, but better than a completely new category.

4. For brand new categories – run the numbers.

Estimate CPC, average price, conversion rate, return rate. In the worst case (all sales from ads), would you still break even? If yes, it's worth testing. Also, think about related products (scenario‑based, series) so you don't run out of ideas after one or two SKUs.

What to look for in a specific product

  • Variations, upgrades, or complementary products of something you already made profitable. You know the launch rhythm, the listing, the customer.

  • Run the numbers – higher price, lower CPC, higher conversion, lower returns. Compare search volume to number of sellers (rough competition measure) against a category where you've already succeeded.

  • Clear differentiation – design, function, or both. But don't just invent random features; base them on best‑sellers and customer feedback.

  • Large enough market size – check competitor sales volume. Too small = poor ROI.

  • A clear competitor to benchmark – best‑sellers have already proven the market. Copying their approach reduces testing cost.

  • Upward trend – ride the wave, don't fight it.

  • Risk assessment – patents, bad reviews, high returns, low conversion. Avoid known risks (especially patents). If risks are unclear, test small.

Final thought

Don't chase hype products. Most people who jump on trends end up stuck with clearance inventory. Controlling risk is more important than chasing profit. The best product is the one that fits your skills and experience.