A product with 40,000 orders looks safer than one with 2,000. That can be backwards. The AliExpress order count on a listing is a running total, so it rewards products that were hot a year ago and hides the ones taking off this month. Two listings can show the same number and sit in opposite places: one adding 30 orders a day, the other adding three. What you want is the rate, and whether that rate is speeding up or slowing down.
This guide shows how to read that rate from three manual snapshots taken over two weeks, then how to turn the numbers into a test, watch or skip call with a short threshold table.
Why a lifetime total hides the trend
Lifetime totals add up every order since the listing went live. Orders from a launch spike stay in the number forever. A gadget that peaked eight months ago and now sells a trickle still looks impressive, while a newer listing sitting at 2,100 orders may be the one shoppers are buying this week.
A few caveats before any math. The figure on a listing is a count the platform chooses to display, and listings do not always explain what it includes, for instance whether cancelled or refunded orders are taken out. Treat it as a rough activity gauge rather than an audited sales figure. Some listings may also show only a rounded figure, so check this before you start. This method needs exact numbers, so pass on those listings or find another listing for the same item. Finally, the orders come from everyone buying that listing, including other dropshippers, so the rate shows how much pull the product has through that supplier and says nothing certain about the orders you would get.
If you still need to find supplier listings in the first place, the walkthrough of the AliExpress Dropshipping Center covers that. This post starts once you already have a listing to judge.
Step 1: Take three snapshots in 14 days
Pick the listings you would actually source from, up to ten. Open each at roughly the same time of day, so a late-evening rush does not skew one reading, and write down the order count on day 0, day 7 and day 14. Note the price, star rating and review count too. A change in any of them between snapshots is useful context: a mid-week price cut can explain a jump in orders.
| Day | Orders shown | Gain since last |
|---|---|---|
| Day 0 | 8,200 | Start point |
| Day 7 | 8,390 | +190 |
| Day 14 | 8,620 | +230 |
If the count falls between snapshots, or the listing disappears, the seller has probably relaunched or removed it. Mark that listing as a skip and look for its replacement, because a reset count makes the history useless.
Step 2: Turn the snapshots into three numbers
You only need three calculations. All three use nothing more than a calculator.
Daily orders
Take the total gain over 14 days and divide by 14. For product A that is 8,620 minus 8,200, which is 420, divided by 14, so 30 orders a day. Think of it as a size check, and the weakest of the three signals, because price changes what a given rate means: 30 orders a day on a $5 item is a very different business from 30 a day on a $60 one.
Share of lifetime
The 14 day gain divided by the day 0 count gives this one. For product A, 420 divided by 8,200 is 5.1%. It shows how fresh the demand is. A listing that adds 5% of its entire history in two weeks is moving, while one that adds under 1% is living off old orders.
Weekly trend
For the trend, put the week two gain over the week one gain. For product A, 230 divided by 190 is 1.21. 1.1 or more means orders are speeding up, 0.8 or less means they are slowing, and anything in between is flat. Of the three, only this one separates a product on the way up from one on the way down, and a lifetime total can never show it.
Step 3: Apply the thresholds
These cut-offs are rules of thumb, not published benchmarks. No marketplace publishes a standard for this, so adjust them for your price point and your tolerance for risk.
| Check | Test it | Skip it |
|---|---|---|
| 14 day gain as share of lifetime | 3% or more | Under 1% |
| Week 2 gain vs week 1 | 1.1x or higher | 0.8x or lower |
| Daily orders | 10 or more | Under 5 |
Two or more checks in the skip column means skip. Three in the test column means test. A check that lands in neither column counts as neutral. Anything else is a watch: take another snapshot a week later instead of guessing. When weekly gains are under about 20 orders, the ratio swings wildly on a handful of orders, so wait another week before you trust it.
Two notes on reading the table. Price matters: for items under $10, the test bar for daily orders rises to 30 or more, because each order earns little, while the skip bar stays at under 5. Size matters too: a listing at 30,000 orders needs a 14 day gain of about 900 to reach 3%, so large listings rarely clear the share row. Above roughly that size, a test call needs the trend and daily rows to pass and the share row to stay out of the skip column.
Three products, three calls
Say you shortlist three listings that each sell for around $20: a pet grooming glove (A), LED strip lights (B) and a mini waffle maker (C). After 14 days the numbers look like this.
| Listing | Share added / trend | Call |
|---|---|---|
| A: grooming glove | 5.1% / 1.21x | Test |
| B: LED strips | 0.4% / 0.71x | Skip |
| C: waffle maker | 16.7% / 1.50x | Test, verify first |
Product B has the biggest total, 41,000 orders, and looks like the safe pick. It gained only 180 orders in two weeks, which is 0.4% of its history, and week two (75 orders) was weaker than week one (105). The daily rate of 13 is a middling result, and the direction fails it. Two of its three checks land in the skip column, so the call is skip, and at 41,000 orders it would have needed a 14 day gain of about 1,230 to reach the 3% test bar anyway. Most of those 41,000 orders were placed long ago.
Listing A passes all three checks on a moderate base. Its gains rose from 190 to 230, so it is growing steadily on a base big enough to trust, which makes it the cleanest candidate for a small test. Watch for distortions, though: if a sale event had lifted that second week from 230 to 380, the trend would read 2.0x instead of 1.21x and A would look better than it is.
Then comes C, which posts the strongest rates: 350 orders in two weeks on a base of 2,100, with week two at 210 against 140 in week one. A small base also means little history to lean on, and a 16.7% share can come from a single viral video. It deserves a test, but only after the cross-checks below confirm that the interest is real.
Step 4: Decide what each call means
A test call means a small, cheap test. Order one or two samples to check quality and delivery time, build one product page, and run ads for 3 to 5 days. Fix the total budget and the stop rule before you start. One simple rule is to stop once you have spent two to three times your profit per sale without a single purchase. At $8 profit per sale, that means stopping at $16 to $24 of ad spend. If sales do arrive, keep going only while your ad cost per sale stays below your profit per sale. A watch call means one more snapshot on day 21, then recalculate the three numbers on days 7 to 21, using the day 7 count as your starting point. A skip call goes in your log with the date and the three numbers, since a product that is slowing out of season may deserve another look a few months later.
Keep a simple watch list
A spreadsheet is enough. Give each listing a row with the supplier name, the listing link, the quoted shipping time, and one column for each weekly order count. Pick a fixed day, say Sunday evening, to update every row, so the habit takes ten minutes and not an afternoon.
Cap the list at ten listings. More than that and the snapshots stop happening. Log every skip call, but retire a listing from the sheet only after three skip calls in a row, and promote anything that moves into the test column to a proper product test. Over a few months the sheet also becomes a record of how quickly products in your niche rise and fade, which helps you judge timing on the next one.
Cross-check before you spend
Order counts are a supplier-side signal. Confirm them with demand-side ones before you commit money.
Start with search interest. Google scales each Trends data point by the total searches in that region and period, then maps it to a 0 to 100 range, so the chart shows relative direction and cannot tell you search volume (Google Trends Help, how data is adjusted). Look up the product's plain name on Google Trends over the past 90 days. A rising line supports a rising order count. A falling line contradicts it, and you should believe the falling line.
Next, check ads. Search the TikTok ad library for the product. Several advertisers pushing the same product in recent weeks supports real demand. Rising orders with no ads at all may mean an organic or very niche product, which is worth knowing before you plan to buy traffic.
Finally, see who else is selling it. Orders on a supplier listing can come from several Shopify stores, and the more stores that sell the same item, the sooner it saturates. If you want store-level data to set beside your snapshots, the DropshipTool sales tracker lets you follow what Shopify stores are selling and how their sales move. Once a listing passes these checks, move on to the demand tests for a new product.
Where the numbers fool you
Sale events
Big platform promotions pull orders forward. A jump in one week followed by a normal week is a one-off spike rather than a trend. Note any sale event that overlaps your snapshots, or start the log just after one ends.
Split listings
The same product is often listed by many sellers, and demand gets divided between them. If your listing looks weak, check the next two or three listings for the same item and add their gains together before you call the product dead.
Price drops
A seller who cuts the price will see orders climb for a while. Check that your landed cost still leaves a real margin at that lower price, because a trend built on a price war is a trap for anyone who copies the offer.
Review timing
Reviews are dated, so you can count how many appeared in the last few weeks as a rough second opinion. Only a fraction of buyers leave one, so use reviews to confirm direction and leave volume estimates to the order count.








