Run one when the discount buys ranking you can hold after the price returns. Skip it when the product has not yet proven its conversion rate. Then judge the result on four signals, rating trend, return rate, conversion rate, and acquisition cost trajectory, never on units sold.
The short version
- A deal buys velocity, not demand. Discounting moves units faster, and the question is whether the ranking survives the return to full price.
- The rerun is the real decision. One deal proves nothing, and a calendar of them becomes a pricing strategy nobody agreed to.
- Validate before you discount. Phase 1 puts 200 units and $5,000 to $10,000 behind a product, and a deal can consume that run before the data arrives.
- Four signals decide. Rating trend, return rate, conversion rate, and CAC trajectory say scale, fix, or kill.
- Fix returns first. Hold the return rate under 8%, because a discount brings volume to whatever the product already does badly.
When a lightning deal on Amazon earns its discount
A deal is a rank purchase with a margin cost, and that framing settles most of the decision. You pay in gross margin per unit, and velocity is what comes back. Velocity feeds ranking, ranking feeds organic sessions, and those sessions are what you keep.
So the question is never whether units move while the price is down. The question is whether position holds once the price goes back up. Sellers who skip that read repeat the deal, and each repeat trains the category to wait for it.
Our 50 operators run about 70 brands by hand from Abu Dhabi, and promotions sit inside each brand's traffic plan. A discount there is approved against a rank target and a margin floor. It is never approved against a revenue gap somebody needs to close this month.
The checklist to clear before you submit one
Work these seven checks in order, and treat a half-done check as a skipped one. Each item states what done properly looks like, so there is nothing left to interpret.
- Confirm the product cleared validation. Rating, conversion rate, and cost of customer acquisition are the three milestones that gate Phase 2. Done properly means all three are proven at full price before any discount runs.
- Write the margin floor first. Done properly means a number per unit that you will not go below, agreed in writing and dated before you choose a discount.
- Check the return rate. Under 8% is the threshold we hold a market to at entry. Done properly means returns are measured on a recent window of orders, never on the launch month.
- Name the rank you are buying. Done properly means the target keywords and the position you need on each are written down before the deal.
- Commit the inventory the window will consume. Done properly means the units are already in the network and the reorder is placed, because a stockout after a deal hands the position straight back.
- Move advertising around the discount. Done properly means bids and budgets change with the price instead of sitting still, so the discount and the bid are not both buying the same order.
- Set the line that stops the next one. Done properly means you write, in advance, the result that would make you refuse a rerun.
Reading the four signals once the price returns
| Signal | What the discount does to it | The read that matters |
|---|---|---|
| Rating trend | Discount buyers review a cheaper purchase | Direction over the following weeks, not the lifetime average |
| Return rate | Volume magnifies whatever the product does badly | Returns from the deal cohort, held under 8% |
| Conversion rate | The price flatters it while the deal runs | The rate at full price a week later |
| CAC trajectory | Deal units look cheap to acquire | Blended cost across the month, with the margin given away counted in |
Those four are the signals behind scale, fix, or kill, and a promotion does not get a scorecard of its own. It gets read through the same four the product was already judged on.
When a signal is off but fixable, the diagnosis runs across seven areas: listing quality, primary image click-through, conversion rate, ad performance, traffic channel activation, pricing, and return rate. A deal is not on that list, because a discount is a lever and not a fix.
Give the decision a defined window, because an open-ended one always resolves in favor of hope. Ours runs 60 to 90 days, and a product that has not improved inside it gets stopped.
I learned that window the expensive way. I funded a failing product for three months waiting for the advertising to turn around, and it did not. Every promotion I ran in those months looked like progress on the sales chart and changed nothing underneath it.
What most agencies will not tell you about deal weeks
Three omissions show up in almost every promotional plan I read, ours included.
- A deal report built on units sold always looks healthy. Lower the price and more units move, which proves nothing about profit. Done properly, the report sets contribution margin beside units for the same window.
- Nobody volunteers the deal they should not have run. A calendar of wins means the losses were quietly deleted. Done properly, the provider hands you the deals that lost money and what changed because of them.
- A discount hides a listing problem for a month. Volume covers a weak conversion rate while the price is down, then the problem comes back with the price. Done properly, conversion is measured at full price before and after, never during.
Hold us to the same three. If we cannot show you a promotion that lost money and the decision it changed, do not hire us.
Related answers
- What ACoS means on Amazon
- How PPC works on Amazon
- Amazon catalog management experts in Dubai
- Audit scope for inventory forecasting and IBC issues
- Amazon account measurement and audits: the complete guide
Do this before the next deal window opens. Take your three highest-volume products and write down, for each one, the margin floor per unit, the current return rate, and the keyword position a discount would have to buy. Any product missing one of those three numbers is not ready to be discounted, and the exercise costs an hour and nothing else.
For a written read on whether your catalog is ready to discount, request the free audit that comes back inside 48 hours from Flapen.
