AliExpress Signals That Actually Matter

9 min · video · Intermediate

AliExpress star ratings are gamed and review counts are misleading. The signals that actually predict whether a product will work for you are hidden in plain sight — order recency, store age, and a small animation most operators never notice.

What to ignore — and why

The metrics most beginners obsess over on AliExpress:

Star rating. Gamed. Suppliers run review-incentive programs. A 4.8-star average is meaningless.
Total reviews. Lagging — a product can have 12,000 reviews from two years ago and zero current demand.
"Best seller" badges. AliExpress assigns these on lifetime metrics, not current trends. Useless for trend-detection.
Generic product photos. Studio shots are reused across 200 listings. They tell you nothing about the actual supplier.

Drop these from your evaluation. They feel like data but are noise.

What to read — the seven AliExpress signals that matter

1. "N orders in last 30 days" badge

Located near the price on the product page. This is the single most useful number on AliExpress because it's harder to fake than total orders and it gives you recency. Look for products with 300+ orders in the last 30 days — that's a baseline for active demand.

2. The "recently ordered" pulse animation

A small "X people just ordered" notification that fires periodically near the product card. If you sit on the page for 60 seconds and see the notification fire 3-6 times, the product has live, real-time velocity. If it never fires, the product is dormant.

This is one of the most underused signals on AliExpress. AliExpress fires it based on actual order volume — it's not a trick, it's telemetry.

3. Store age

Hover/click on the seller name. AliExpress displays "Store opened: YYYY". Look for 3+ years. AliExpress purges fraudulent stores in waves; a store that has survived multiple purge cycles has a real operation.

A 6-month-old store, no matter how cheap or fast, is a flight risk. They could be:

A new dropshipper themselves (no inventory)
A scam operation that disappears in 90 days
An existing store that was banned and is now on a new account

4. Positive feedback rate

Found on the seller profile, separate from product rating. Aim for 95%+ positive. Below 93% is a structural reliability problem (not just a few bad reviews) — disputes, slow shipping, or product-quality issues at scale.

5. Real customer photos in recent reviews

Sort reviews by "recent" and look for photos. Customer photos in the last 14 days of reviews tell you:

The product is shipping (people receive it)
The product matches the listing (otherwise photos would be angry)
The customer base is real (bots don't take photos)

6. Variant complexity

A product with 30 variants (colours / sizes / models) is harder to dropship than one with 3-5 variants. Inventory complexity inflates the chance of out-of-stock surprises and supplier confusion.

For beginner stores: prefer products with 1-5 variants. Save 30-variant catalogues for when you're scaling and have an agent.

7. Multiple suppliers selling the same product

Search the product title or upload the image to AliExpress search. If 5+ different sellers carry the same product, you have:

Supply redundancy (one delisting won't kill you)
Negotiation leverage (you can switch)
Product validation (suppliers don't inventory losers)

If only one seller has it, you're either looking at a niche unique product (high margin opportunity) or a fragile supply chain (high risk). Beginners should pick the redundant supply chain.

How to do all seven in 60 seconds

A practiced operator opens an AliExpress listing and reads:

Price → mental note (3 sec)
"N orders in 30 days" → mental note (5 sec)
Wait 30 seconds, watch for the recently-ordered pulse (30 sec)
Click seller name, read store age + positive feedback (10 sec)
Sort reviews by recent, scan for photos in last 14 days (8 sec)
Count variants — under 10 is good (3 sec)
Search product image to count alternative suppliers (50 sec on AliExpress's image search)

Total: roughly 60-90 seconds for a practiced operator. Twenty products an hour at high signal.

Where Majorka does this for you

Majorka pre-computes signals 1, 3, 4 and 7 across the catalog. The Winning Score uses all four as inputs. When you open Products and a row shows score 88, you can trust that the underlying AliExpress signals are:

30-day order count above a category-relative threshold
Supplier age over 2 years
Multiple sellers detected
Feedback rate inside acceptable range

You still want to spot-check signal 2 (recently-ordered pulse) and signal 5 (recent customer photos) on your shortlist by clicking through to the AliExpress source page — Majorka surfaces the canonical AliExpress URL on every product.

Why this matters

The single biggest difference between an operator who picks winners and one who picks duds is what data they read. Star ratings and review counts are public — and gamed. Order velocity, store age, supplier multiplicity and feedback rates are quieter, harder to fake, and dramatically more predictive. Read the right signals, ignore the noise, and your shortlist quality 5x's overnight.

Two pet bowls, both 4.8 stars, only one has a pulse

Two slow-feed dog bowls on AliExpress in May 2026:

Bowl A — A$3.20 COGS, 4.8 stars, 2,400 reviews, 1,840 orders in last 30 days, store opened 2019, 96.4% positive feedback, 8 sellers carrying the same SKU. Recently-ordered pulse fires every 12-18 seconds.

Bowl B — A$3.90 COGS, 4.8 stars, 8,200 reviews, 180 orders in last 30 days, store opened October 2025, 94.1% positive feedback, 1 seller. Recently-ordered pulse never fires in a 90-second window.

Star rating is identical. Review count favours Bowl B. Every other signal screams Bowl A. The operator who picks Bowl B because of "more reviews" is buying a corpse from a 6-month-old store. The operator who picks Bowl A is entering active demand from a proven supplier base. Six months later: Bowl A operator has done A$84k revenue. Bowl B operator killed in week 3 after the supplier delisted half the variants.

Action items

Open three random products from your shortlist on AliExpress.
Time yourself running the seven-signal scan. Aim for under 90 seconds.
For each, write down: 30-day orders, pulse frequency, store age, supplier count.
Eliminate any product where pulse never fires in 60 seconds — dormant demand is a leading indicator of decline.

Next lesson: TikTok search for product discovery, the right way — the search strings that surface rising winners, not the same viral replays everyone else has already cloned.

Sources

AliExpress order count display — aliexpress.com product pages
AliExpress seller verification — sale.aliexpress.com seller policies
AliExpress image search — aliexpress.com image search documentation

Module 02 — Product Research — Finding Winners Before Everyone Else

The hardest skill in this business. Data-driven frameworks for spotting products at the beginning of their curve, not the end.

Lessons in this module

The 4 Types of Winning Products (and which you should pick) · 11 min
Problem-solvers, wow-factor, impulse, evergreen — the trade-offs of each.
Trend Velocity — Catching a Winner at Day 10, Not Day 60 · 13 min
How to read a velocity curve and when to pounce.
AliExpress Signals That Actually Matter (this lesson) · 9 min
Ignore reviews. Watch orders, store age, and "recently ordered" pulse.
TikTok Search for Product Discovery (the right way) · 10 min
The search strings that surface rising products, not viral replays.
Meta Ad Library — Reverse-Engineering Competitor Winners · 12 min
How to tell a test from a scale, and steal the ad angle without the copy.
The Majorka Winning Score Explained · 8 min
What goes into the score, why it beats raw order counts, how to use it.
Building a 20-Product Shortlist in Under an Hour · 15 min
Live walkthrough: from dashboard to validated shortlist, fast.