The budget was approved at four thousand ringgit and spent in six weeks on TikTok Shop Malaysia. Shipping was fine. The creators were real. Eleven of the thirty-four who received a parcel posted a video.
The campaign still lost money. At a forty-five ringgit price point, the parcels cost more than the orders they produced.
So this is for sellers running ten to thirty SKUs there, with one or two people on affiliate work. The market volume is genuine. The unit economics are what nobody modelled before the first shipment went out.
What makes this market different is not the shipping cost. It is the ratio. A parcel costs roughly the same to move whether the product retails at nine dollars or ninety. So on a low-AOV market the sample stops being a marketing expense. It becomes a fixed cost per attempt.
A low AOV turns every sample into a debt
In practice, most sample planning is done in units, not money. Send forty parcels. Expect a third to post. Expect some of those to sell. That framing works where one order pays back a meaningful share of the parcel. It falls apart when the product is cheap, because parcel and product stop being comparable amounts.
So put numbers on it. A product retailing at forty-five ringgit with a forty percent margin returns about eighteen ringgit per order. A parcel shipped cross-border costs forty to sixty ringgit, once product cost and freight are counted. So the creator has to produce roughly three orders before the sample pays for itself. Four, before it contributes anything.
But that bar is higher than it sounds. It is not “did the video get posted.” It is “did it sell three units.” A creator who posts to nine thousand followers and moves one unit has cost you money, though every step looked successful.
What one sample has to earn back
So the useful exercise is to price the parcel against the order, for the actual product, before the list is built. The table below runs the same arithmetic at four price points. Margin and freight are held constant.
| Retail price (RM) | Margin per order (RM) | Orders to repay one sample | What that implies |
|---|---|---|---|
| 25 | 10 | About 5 | Only creators with a record of multi-unit sales |
| 45 | 18 | About 3 | Mid-tier creators who have sold this category before |
| 80 | 32 | About 2 | Broader list becomes viable |
| 150 | 60 | 1 | Reach matters more than a sales record |
Then read down the third column and strategy changes with price. Under fifty ringgit, a sample is only rational when the creator has shown they can move several units of something similar. Above a hundred, one order covers the parcel, and the calculus flips towards reach.
And this is also why a creator list cannot be reused across markets without redoing the sums. What works for a ninety-dollar product in Singapore is over-invested at forty-five ringgit. Not because the creators are worse, but because the payback period doubled.
Who to send to when the margin is thin
Once the payback bar is three or four orders, the selection rule stops being about audience size. It starts being about evidence of sales. Three signals do most of the work.
The first is category history. A creator who posted about this category in the last sixty days has an audience already primed to buy. That shortens the path from video to order.
The second is whether they have sold anything at all, not how many followers they have. Ten thousand followers plus a record of moving product beats two hundred thousand with no sales history. On this market more than most.
The third is response behaviour. On a thin margin you cannot afford the parcel that sits unopened for three weeks. The budget is committed the moment it ships. So a creator who replied within two days last time is a different risk from one who took a fortnight.
But none of those three is visible in a follower count. That is why hand-built lists go wrong. The information exists. It is spread across the platform, a spreadsheet and somebody’s memory.
This is the specific place DAMI earns its cost on a low-AOV market. The creator filter runs on category, audience and follower band together. So the pool that comes back is already narrowed to people who produced content in this space. DAMI’s sample workspace then ties each parcel to a creator and a date. That turns “who is sitting on an unopened sample” into a filter rather than a guess. That single check recovers more budget than any negotiation does.
Then there is attribution. DAMI’s product sales view shows what each creator’s traffic actually converted into, after refunds and fees. So the second campaign’s list is built from the first campaign’s evidence, not its impressions. On a market with a three-order payback, that feedback loop is the whole method.
One honest limit. No amount of selection makes a forty-five ringgit product carry a sixty-ringgit parcel comfortably. The arithmetic is uncomfortable, and it stays uncomfortable. What better selection does is raise the hit rate enough that the average works. That is different from making each parcel profitable.
The second limit is that a low AOV market rewards many small creators over a few large ones. And small creators are harder to manage, not easier. More names, more messages, more parcels, each with a thinner margin. DAMI makes that manageable. It does not make it free.
For the language side, which is a separate constraint here rather than the same one, running outreach in Malay, English and Chinese covers what changes when the message itself has to be localized. And why big creators waste a sample budget is the same problem from the opposite end.
Where the creator filter, the parcel record and the per-creator sales figure sit together, the DAMI workspace is where the payback arithmetic gets run.
FAQ
Is TikTok Shop Malaysia worth running at a low price point?
Yes, if the sample list is short and evidence-based. The market moves volume, and the creators are reachable. What does not work is importing a sample strategy built for a higher price point. The payback period doubles, and the same list turns from an investment into a cost.
Should I send samples to creators with no sales history?
Rarely, and never in bulk. One parcel to test a creator is a cheap experiment. Twenty parcels to a category you have not tested is a budget spent on hope. If the product retails under fifty ringgit, keep the test to a handful of names and hold the rest of the budget until the first results land.
How many samples can a small budget carry on TikTok Shop Malaysia?
Divide the budget by the full parcel cost, not the product cost. At forty-five ringgit retail with cross-border freight, that is fifty to sixty ringgit per attempt. So four thousand ringgit is around seventy parcels, and fewer if any need replacing. Most teams find the real number is half what they assumed.
