Two years ago I lost $1,400 chasing a "winning product."
It was a posture-corrector gadget — the kind with the dramatic before/after ads. It sat at the top of every "winning products this month" listicle I could find. The comments were full of sellers bragging about their ROAS. So I did what the gurus said to do: I ordered 300 units, spun up a one-product store, and launched ads.
The first week felt incredible. Clicks were cheap, add-to-carts were rolling in, and I genuinely thought I'd cracked it. Then, around day ten, everything fell off a cliff. CPMs doubled. Then tripled. My "unique angle" turned out to be the exact same angle as forty other stores, because we'd all copied the same listicle, which had copied the same original store.
I ended up selling the remaining inventory at cost to a liquidator and shutting the store down. The worst part wasn't the money — it was the realization that I'd done everything "right" according to the playbook, and the playbook itself was the problem.
Why winning-product lists are always late
Think about how a "winning products" list actually gets made. Someone notices a product selling well. They write it up. An editor publishes it. Aggregator sites copy it. By the time a product earns the label "winning," here's what's already happened:
- The original seller has been running it profitably for weeks or months.
- Dozens of copycats have entered and started bidding up ad costs.
- Suppliers have noticed the demand spike and raised wholesale prices.
- Customers have seen the same ad creative seventeen times.
A winning-product list is a lagging indicator. It's a photograph of a trend that already happened. Trading on it is like buying a stock after the earnings call — the information is public, priced in, and the edge is gone.
The uncomfortable question I sat with after my $1,400 lesson was: what would a leading indicator look like?
The signal I almost ignored
A few months after the posture-gadget disaster, I noticed something odd. Within about three weeks, I kept stumbling across brand-new stores — unfamiliar names, fresh domains, janky-but-earnest branding — all selling variations of the same niche product: a magnetic car-mount-style phone holder, but redesigned for desks and gym equipment.
No listicle had mentioned it. No guru was talking about it. Each store on its own looked like nothing — a handful of visitors, a basic Shopify theme, maybe a dozen products. But there were five of them, all launched within a month, all converging on the same idea.
I didn't act on it. I told myself it was a coincidence, that I was pattern-matching noise. Six months later that product category was everywhere — including, ironically, on the winning-product lists. I hadn't lost money this time, but I'd left the lesson on the table: the earliest real signal of a trend isn't a product going viral. It's multiple new stores betting on the same thing at the same time.
That changed how I do product research entirely. I stopped asking "what's selling now?" and started asking "what are new stores being built to sell?"
The method: build a new-store radar
The core idea is simple: newly registered domains + sudden traffic growth = someone spending real money on a bet. Domain registration costs money. Building a store costs time. Running ads costs serious money. When a brand-new store starts gaining traffic fast, that's conviction you can observe from the outside.
Here's the DIY version, step by step. It's unglamorous, but it works.
Step 1: Harvest newly registered domains
You need a feed of fresh domains. Practical sources:
- Certificate Transparency logs (via crt.sh) — every new HTTPS cert is public. Query for recently issued certs and extract the domains.
- RDAP lookups — for any domain, RDAP tells you the exact creation date. This is your age filter: keep domains registered within the last 90–180 days.
- New-domain feeds and drop lists — several free feeds publish daily registrations.
Filter aggressively by TLD. Sellers overwhelmingly use .com, .store, .shop, and .co. A .xyz domain can be a store, but your time is better spent where the signal density is highest.
Step 2: Detect stores, not blogs
Most new domains are junk — parked pages, spam, affiliate blogs. You need store fingerprints:
- URL paths like
/cart,/checkout,/products/,/collections/ - Platform signatures: Shopify (
cdn.shopify.com,/cart.js), WooCommerce (/wp-content/plugins/woocommerce), BigCommerce, Wix Stores - Commercial intent markers: "Add to Cart" buttons, price schema, checkout flows
A simple crawler checking the homepage for these markers will eliminate 90%+ of non-stores. This filtering step is the boring part, and it's also the part most people skip — which is exactly why the signal stays valuable.
Step 3: Measure velocity, not size
This is the key mental shift. You don't care how big a store is. You care how fast it's growing.
- Track ranking movement over 2–4 week windows (lists like Tranco publish rank data you can diff over time).
- Watch for domains that go from "doesn't exist in the data" to a meaningful rank within weeks. That's the explosion pattern.
- Ignore absolute traffic numbers — a store doing 50k visits/month that's flat is less interesting than one doing 5k that's 10x'd in a month.
New + accelerating beats big + flat, every time.
Step 4: Cluster by niche
One fast-growing new store is a curiosity. Five new stores in the same niche within a month is a trend forming.
Group the stores you're tracking by what they sell — category, product type, even visual similarity of hero products. When a cluster forms, dig in: are they all dropshipping the same supplier's product? Is it a genuinely new product, or a new angle on an old one? The cluster tells you where to look; the individual stores tell you what's working (compare their angles, pricing, and creatives).
Step 5: Confirm conviction with ad libraries
Before you commit to a niche, verify that these stores are spending money — not just existing:
- Search the store domains in the Meta Ad Library and TikTok Creative Center.
- Stores running active, recent ads are putting cash behind their bet. That's the strongest confirmation signal available for free.
- Bonus: the ad creatives themselves are a free education in what's converting — hooks, angles, offers.
Red flags to filter out
- Parked or redirect domains with no real storefront.
- Copy-paste clone rings — dozens of identical stores from the same operator (same theme, same copy, same products). That's one bet, not a trend.
- Brand launches with PR — a celebrity brand will spike on launch day; that's marketing, not organic demand.
- Seasonal spikes — Halloween stores in October are a calendar event, not a discovery.
The honest caveats
I'll be straight with you: this method has costs.
First, it's tedious. For every interesting new store you'll sift through hundreds of dead domains. The filtering is 80% of the work and 0% of the fun.
Second, it needs discipline. Checking once gives you noise; checking weekly gives you velocity data, and velocity is the whole game. A radar you don't maintain is just a spreadsheet.
Third, early signals are probabilistic, not certain. A cluster of new stores means several people independently bet money on the same idea — that's meaningful, but it's not a guarantee. You'll still be wrong sometimes. You'll just be wrong earlier and cheaper than the listicle crowd.
If you don't want to build this yourself
Here's the thing I wish someone had told me two years ago: steps 1 through 3 above — harvesting domains, filtering stores, computing velocity — are pure infrastructure. No product insight happens there; it's plumbing. The insight starts at step 4, clustering and interpretation.
If you'd rather skip the plumbing, that's exactly what RainbowSea does. It tracks newly registered domains, filters for real stores, and ranks them by traffic growth velocity — so you can discover trending new stores while they're still accelerating, instead of reading about them in a listicle two months later. The e-commerce category alone has 1,000+ new stores ranked this way, which is a solid starting feed for the clustering step.
Start here if you're curious: https://rainbowsea.org
Whether you build the radar yourself or use a tool, the principle is the same: watch where new stores are being built, not what old lists are celebrating. The list tells you where the parade has been. The new stores tell you where it's going.
Your turn
What's the earliest signal you've ever caught — the time you knew before everyone else? Or the flip side: the trend you saw forming, talked yourself out of, and watched explode without you?
Drop it in the comments. I'm genuinely curious which niches you're watching right now — and I have a feeling the best answers here will be better than any listicle.
Top comments (0)