How to find winning products

A winning product is not a product that looks clever in a video. It is a product that other stores are already selling at a profit, with demand that is still growing and room left for one more seller. You cannot see anyone's bank account, but you can see a lot of public evidence: what stores rank as best sellers, which ads keep running, how catalogues change and how much traffic a store gets. This guide shows how to combine those signals into a short list you can actually test.

Key takeaways

• Look for agreement between several signals. A product that ranks high on a store, has ads running for weeks and appears on several stores is far stronger than one viral clip.
• Start from stores, not from products. Successful stores lead you to products that are proven in a real checkout.
• Ad longevity is the cheapest proof of profit you can get. Advertisers rarely keep paying for an ad that loses money.
• Every sales figure from an outside tool is an estimate. Use it to rank and compare, not as a promise of what you will sell.
• Finish with a validation checklist before you spend on ads.

What makes a product a winner

The term winning product gets used loosely. For research purposes, a product qualifies when most of these are true:

• It solves a clear problem or creates an obvious "I want that" moment in a few seconds of video.
• It has a healthy margin after product cost, shipping, payment fees and ad spend. As a rough rule, a selling price of three times landed cost leaves room for advertising.
• Several independent stores sell it, and at least one of them has run ads for it for weeks.
• Demand is rising or stable, not falling. A product that peaked two months ago is a trap.
• It is not dominated by one brand with a patent, a strong private label or deep pockets.

No single signal proves all of this, which is why the method below stacks them.

Step 1. Start from stores that are already selling

The fastest route to proven products is a list of stores that are clearly doing well in your area. In RivalSonar, open the Store explorer and filter by category, country and estimated monthly orders. Add a creation date filter if you want younger stores, since a store launched a few months ago that already shows strong estimated orders is usually riding a specific product rather than an old brand.

Without a tool, you can build the same list by hand:

• Search your product category on Google and note the independent stores that appear beside the marketplaces.
• Search the Meta Ad Library for category keywords and note which store domains keep showing up.
• Look at the "similar stores" or "customers also visited" suggestions that browser tools and traffic sites show.

Aim for 20 to 50 stores. That is enough to see patterns without drowning.

Step 2. Pull each store's best sellers

Once you have stores, find what each one sells most. Many Shopify stores expose a collection sorted by best-selling, and the public product feed shows the full catalogue. The full method is in how to find a store's best sellers. In RivalSonar, the Store profile lists products in best-seller rank order with an estimated share of orders, and the Product explorer lets you search best sellers across every store at once.

Write down the top three to five products per store. You are looking for repeats, the same product or the same product type showing up near the top of several unrelated stores.

Step 3. Check the ads behind the product

A product that sells is almost always being advertised. Search the product name, a distinctive feature or the store domain in an ad library. RivalSonar's Ad library covers Meta, TikTok, Pinterest and Snapchat ads, and shows how many days each ad has been active, an estimated reach and spend, and an AI score that labels an ad as testing, scaling or winning.

What to look for:

• Active days. Ads that have run for 30 days or more are usually profitable. Ads under a week old are tests. Read how to read ad longevity and reach for the details.
• Number of variations. An advertiser that has launched many creatives for one product is scaling it.
• Several advertisers. If three or four different stores advertise the same item, demand is real, but competition is too.

Step 4. Read the trend direction

A product can have all the right signals and still be on the way down. Check direction before you commit:

• Compare a store's estimated orders this month with last month. A falling estimate on the store that made the product famous is a warning.
• Look at when the newest ads started. If no new creatives have appeared for weeks, advertisers may be winding down.
• Check whether new stores are still launching around the product. A flood of copycat stores often marks the late stage of a trend and rising market saturation.

The Product explorer's trending and new filters help here, because they surface products whose estimated orders are rising rather than simply large.

Step 5. Size the opportunity honestly

Estimated orders tell you whether a product moves at all. They do not tell you that you will sell the same amount. Treat them as a ranking signal. A product with an estimated 1,500 orders a month across five stores is a more promising test than one with an estimated 80 orders on a single store, even if both estimates are rough.

Then do the unit economics yourself:

If similar products in your niche typically need 20 to 30 in ad spend per purchase, this product is marginal. A higher price, an upsell or a bundle that lifts average order value can change the picture.

Step 6. Validate before you test

Before you build a store or a product page, run the product through a checklist. Is there a supplier who can ship reliably to your target country? Are the reviews on competing stores real and positive? Can you make a better offer or a better ad than the stores already selling it? Our product validation checklist goes through each question.

Common mistakes

• Chasing a single viral video. Views are not orders. Without store and ad evidence, a viral clip is just entertainment.
• Copying the top store exactly. If you sell the same item at the same price with the same ad, you are competing on ad budget alone.
• Treating estimates as facts. Sales numbers from any outside tool are modelled. They are useful for comparing stores and products, not for forecasting your own revenue.
• Ignoring countries. A product that is saturated in the US can still be early in Germany or Australia. See how to research a niche across countries.

A simple weekly routine

1. Monday, refresh your store list with new stores that crossed your estimated order threshold.
2. Tuesday, pull best sellers from those stores and add repeats to a shortlist.
3. Wednesday, check ads and trend direction for each shortlisted product.
4. Thursday, run validation on the top two or three.
5. Friday, pick one to test, and save your searches so the daily digest brings new matches to your inbox.

Saved searches and Collections in RivalSonar make this routine faster, but the method works with spreadsheets and public pages too.