If you are a TikTok Shop Vietnam beauty seller at a VND 450,000 basket, hit on 3 August 2026 by the commission rising from 17.8% to 19.8% and the voucher surcharge rising from 4% to 5%, with a 13-day window from 3 to 15 November, the decision in front of you is not how deep the voucher goes. It is what share of it you fund, and how many more units that share has to sell before it pays for itself. Vietnam is the market where the voucher itself got taxed. On 3 August 2026 the beauty commission moved from 17.8% to 19.8% and the voucher surcharge moved from 4% to 5%, so every discounted order now carries a fee that is charged on the discount you gave. Stack a 6% transaction fee and a fixed VND 3,000 per order on top and the arithmetic stops being intuitive: a 20% voucher on a VND 450,000 basket leaves VND 12,120 of profit where an undiscounted order left VND 83,400.
Take a VND 450,000 basket at 55% cost of goods, so VND 247,500. Undiscounted, at 19.8% commission, 6% transaction fee and VND 3,000 processing, you keep VND 83,400. Now fund a 15% voucher, VND 67,500. The buyer pays VND 382,500; commission falls to VND 75,735, the transaction fee to VND 22,950, but a new line appears, the 5% voucher surcharge, which is VND 3,375 charged on the discount itself. Profit becomes VND 29,940, so the voucher cost you VND 53,460 rather than VND 67,500, and you need 2.79 times the volume. Push the voucher to 20%, VND 90,000: the buyer pays VND 360,000, commission VND 71,280, surcharge VND 4,500, transaction fee VND 21,600, processing VND 3,000, and profit is VND 12,120. That is 6.88 times the volume. Five more points of discount multiplied the required volume by 2.5.
Two rate changes on 3 August 2026 that made the voucher a taxed instrument
Vietnam changed two lines at once on 3 August 2026, and they interact. The beauty commission moved from 17.8% to 19.8%, which raises the cost of every order, discounted or not. The voucher surcharge moved from 4% to 5%, which raises the cost specifically of the orders you discount, because it is charged on the discount amount. Together they mean the marginal cost of going one point deeper on a voucher is no longer one point: it is one point of discount plus 5% of that point in surcharge, on top of a commission base that is already two points higher than it was in July. A plan built on last year's numbers under-states the cost of the campaign by a wide margin, and the error is concentrated exactly on the SKUs the campaign is built around.
The surcharge is the line that decides the cliff, and almost nobody prices it. At 5% of the discount, a 15% voucher carries VND 3,375 of surcharge on a VND 450,000 basket, and a 20% voucher carries VND 4,500. Those numbers look small next to a VND 67,500 or VND 90,000 discount, and on their own they are. What matters is what they do to the marginal rate. Going from 15% to 20% costs you VND 22,500 more in discount and VND 1,125 more in surcharge, but it also removes VND 7,455 of commission and VND 1,350 of transaction fee, because both fall with the price. The net extra cost is VND 14,820, and that VND 14,820 is what turns VND 29,940 of profit into VND 12,120.
Step 1 — Put the voucher surcharge on its own line, because it is new
Build the stack from the bottom up and give the surcharge its own row. On a VND 450,000 basket the order is: cost of goods VND 247,500, commission at 19.8%, transaction fee at 6%, processing VND 3,000, and then, only if you discount, the voucher amount and 5% of it. That last pair is the one missing from most seller spreadsheets, because the surcharge did not exist at 5% until August and because it is charged on a number sellers think of as money given away rather than money spent twice. Put it in and the arithmetic changes: the voucher is not a discount, it is a discount plus a fee on the discount.
Note what the surcharge does to the shape of the decision, not just its size. Because it is proportional to the discount, it is neutral between a 10% and a 20% voucher in percentage terms, and it therefore does not discourage depth on its own. The thing that punishes depth is the interaction with the fixed lines: the VND 3,000 processing fee and the cost of goods do not move when the price moves, so as the discount deepens they take a larger share of a smaller payment. At a 20% voucher, VND 3,000 is 0.83% of what the buyer pays. At a 30% voucher it is 0.95%. Small, but it is the same direction as the commission, and the two together are why the curve steepens rather than flattening out.
If you would rather not rebuild the stack by hand, the per-SKU margin sheet is on the dashboard page — register with DAMI and ask the team for the per-SKU margin sheet and it already separates discount, surcharge and commission into three rows.
Step 2 — See where the cliff is between 15% and 20%
The cliff is between 15% and 20%, and it is worth seeing why. At 15% the break-even is 2.79 times volume, which is hard but achievable in a 13-day window with real traffic. At 20% it is 6.88 times, which is not a campaign target, it is a different business. The reason the multiple explodes rather than rising in proportion is that both the numerator and the denominator move: profit before the voucher is fixed at VND 83,400, while profit after it has already fallen to VND 29,940 at 15% and then to VND 12,120 at 20%. Dividing a fixed number by a shrinking one produces a curve, not a line. This is why the same five points feel trivial in the meeting and impossible in the settlement.
There is a second reason the cliff sits exactly there, and it is specific to beauty. Consumables in this category have a repeat cycle, so the argument for a deeper voucher is lifetime value rather than the first order. That argument is legitimate but it has to be priced honestly: if the first order leaves VND 12,120 against VND 29,940 at the shallower depth, the deeper voucher is borrowing VND 17,820 from the future, and it only pays back if the repurchase actually happens inside the window where you still carry the customer. Most sellers cannot measure that, which means the deeper voucher is being approved on a story rather than on a number.
Step 3 — Find the SKUs that go negative, and take them out of the campaign
The negative-margin SKUs are the ones where the fixed lines dominate, and they are identifiable before the campaign rather than after. Take each SKU's price, subtract 55% for goods, 19.8% for commission, 6% for the transaction fee and VND 3,000 for processing, and you have the undiscounted profit. Divide it by the price and you have a contribution rate. Any SKU whose contribution rate is below the voucher depth you plan to run will go negative when that voucher applies, because the voucher reduces the payment without reducing any of the fixed lines. On the numbers above the undiscounted contribution is 18.5%, so a 20% voucher puts every SKU below roughly VND 500,000 into loss.
This is the practical argument for capping the voucher by SKU instead of by store. A store-wide 20% voucher applies to the cheap SKUs that cannot carry it and to the expensive ones that can, and the cheap ones are usually the volume drivers, so the loss is concentrated exactly where the orders are. Cap the depth at 15% store-wide and run 20% only on the SKUs whose contribution rate clears it, and the average funded depth falls while the conversion on the campaign pages stays intact. The rule is not about generosity, it is about which SKUs can pay for the discount.
Step 4 — Thirteen days from 3 to 15 November, so cap the stack, not the discount
Thirteen days, 3 to 15 November, is long enough that the stack matters more than the depth. Vouchers, platform campaigns and shop coupons can apply to the same order, and each one adds to the funded amount without adding to the volume. The cap should be on the total funded percentage, not on any individual voucher: decide that no order will carry more than, say, 22% of total discount, and then allocate that 22% across the instruments available. Once the cap is set at the order level, a new platform campaign cannot quietly push you past the cliff, because you have already decided what the maximum is.
Keep one column for realised depth, because the planned depth and the realised depth are never the same number. Vouchers that do not get used, coupons that expire, platform campaigns that do not fill: after thirteen days the realised average is usually below the planned one, which is good news for margin and bad news for the measurement, because a campaign that under-delivers on discount also under-delivers on volume. The only fair comparison is realised depth against realised volume, not planned against planned.
Where DAMI fits: three moves that keep the cliff out of the campaign
DAMI brings an 8-million-plus creator database, competitor and competitor-store creator search, automated outreach tasks, AI-assisted outreach, AI video creation, sample management, campaign management, social media management, store analytics, and multi-store team dashboards into one place. Three moves cover this, and each one leaves you with a file rather than a view.
Move 1 — Rebuild the cost stack with the surcharge as its own row
Cost of goods, commission, transaction fee, processing, discount, surcharge. The output is a six-line stack where the two lines changed in August are visible rather than folded in, and it replaces the four-line version most spreadsheets still use.
Move 2 — Compute contribution rate per SKU and mark the ones under your planned depth
Divide undiscounted profit by price for every SKU, then flag anything below the depth you plan to run. The output is a list, usually shorter than expected, of the SKUs that must not carry the campaign voucher.
Move 3 — Set the cap on total funded percentage per order
Decide the maximum total discount any single order can carry, then allocate it across vouchers, shop coupons and platform campaigns. The output is one number that survives the campaign even when a new platform offer appears.
Questions sellers ask after the rate change
Did the August change apply to all categories or only beauty?
The commission change is category-specific and beauty moved to 19.8%, so check your own category rather than assuming the number applies to you. The voucher surcharge moved from 4% to 5% and is charged on the discount, so it applies wherever a voucher is funded. The two changes are separate, and a seller outside beauty may have only one of them.
Is the surcharge charged on the discount or on the order?
On the discount. That is what makes it a fee on generosity rather than a fee on sales, and it is why it does not fall when your voucher does less work than planned. If you plan a 20% voucher and only 12% of orders use it, the surcharge is charged on the 12%, not on the plan.
Why does the break-even multiple jump so sharply rather than climbing steadily?
Because profit after the voucher is the denominator, and it is approaching zero. At 15% it is VND 29,940; at 20% it is VND 12,120. Halving the denominator doubles the multiple, so each extra point of depth costs more than the one before. This is not a modelling artefact, it is what happens when fixed costs are a large share of a small payment.
Can I recover the lost margin with volume alone?
Only if the volume is real. A 6.88 times requirement on a 13-day window means the campaign must deliver roughly seven times your normal daily order count, sustained, with stock to match. Where that traffic actually exists the deeper voucher can work; where it does not, the campaign borrows from the following month by pulling forward demand that would have arrived anyway.
The window opens on 3 November and I have already committed the depth. What can still be changed?
The SKU list the voucher applies to, and the cap on total funded percentage per order. Both change the realised cost without touching the committed rate. On a catalogue with a wide price range, the SKU list is usually the larger lever of the two.
The one number to check on 16 November
Every voucher in this piece is a ratio question wearing a percentage sign. The percentage tells you what the buyer sees; the ratio tells you how much more you have to sell. DAMI brings an 8-million-plus creator database, competitor and competitor-store creator search, automated outreach tasks, AI-assisted outreach, AI video creation, sample management, campaign management, social media management, store analytics, and multi-store team dashboards into one place.
Open your SKU list, add one column for undiscounted profit, and divide it by price. Any SKU under 20% cannot carry a 20% voucher without going negative, and on a VND 450,000 basket with these fee lines most of your catalogue will be under 20%. Twenty minutes gives you a list of what must not go into the campaign, and that list is worth more than any optimisation of the discount depth.
Start with DAMI and put the ratio next to the discount before the campaign puts it on your P&L for you.
