The payback goal is the filter. Ask every candidate for two advertising cost of sale numbers, the one they run at launch and the one they run at maturity. Anyone who quotes a single figure for both is not planning a launch. Then ask what they would cut to hit three months.
The short version
- Two numbers, not one. A launch target and a maturity target, with the month they expect to move between them.
- Three-month payback and fast rank are in tension. Both are achievable, but not at full speed at the same time, and a candidate who denies the trade-off is guessing.
- Test before you scale. A first phase of around 200 units and $5,000 to $10,000 tells you whether payback is even possible.
- Full brand launches run around seven months. A three-month payback goal on a single product is a different project from a five-product brand.
- Score candidates on the answers, not the deck. Nine questions, pass or fail, before anyone sends a proposal.
Start with the number that decides the shortlist
Advertising cost of sale is not a fixed target. It is a setting that changes with the stage of the product. At launch you are buying rank, reviews, and data, and the acceptable number is high, sometimes higher than your margin. At maturity you are defending position and harvesting demand you already created, and the number should be well inside your contribution margin.
So the useful question in a first call is not what ACoS do you target. It is: what number do you run in launch month one, what number do you run at month twelve, and what has to be true for you to move from the first to the second. Two figures and a trigger. Any candidate who has run launches will answer in under a minute. Any candidate who has only run mature accounts will give you one number and a caveat.
That single exchange removes most of a shortlist.
What a three-month payback actually asks for
Payback in three months means the cash you put in comes back out within about ninety days of the first order. On a single product, budget $8,000 to $15,000 of total launch capital covering inventory, freight, imagery, trademark, and ad spend. On a five-product brand, budget $25,000 to $50,000. A full brand launch takes around seven months to run properly, so a three-month payback goal on a five-product brand is a request for two different timelines at once.
The trade-off is simple, and it is worth saying to every candidate out loud. Fast payback means a conservative launch: less aggressive bidding, slower rank climb, tighter keyword set, and a longer wait for organic volume. Fast rank means the opposite. You are choosing which one you want, and the agency should tell you what the other costs.
The nine-item shortlist checklist
Work through this in order. A candidate who fails an item does not get a proposal.
- Two ACoS numbers with a trigger between them. Done properly: launch figure, maturity figure, and the rank or order volume that moves you from one to the other.
- A validation phase before the full order. Done properly: a first run of about 200 units at $5,000 to $10,000, with up to four products tested at once, and named criteria for continuing.
- A written payback model. Done properly: units, price, fees, cost of goods, and ad spend in one sheet you can edit, not a verbal promise.
- A stated view on whether your goal is realistic. Done properly: they tell you which of speed or payback they would sacrifice, before you ask.
- A named operator. Done properly: the person doing the work is named, and you know how many other brands they carry.
- A stop rule. Done properly: rating trend, return rate, conversion rate, and cost of acquisition, measured over a defined window, with an agreed action.
- A reporting cadence you can audit. Done properly: a written update weekly and a live review every two weeks, starting in week one rather than month two.
- Fee separated from pass-through costs. Done properly: management fee, ad spend, inventory, and Amazon's own fees on separate lines.
- Exit terms in writing. Done properly: notice period, what you keep, and who holds account access. Ours is month-to-month on 30 days notice, with the account, campaigns, and creative staying yours plus a written handover.
Where three-month payback goals usually break
Not on the ad account. They break on the product. A launch with an ordinary price, an ordinary rating, and no reason to be chosen cannot be made to pay back in ninety days by any bidding strategy. The candidate who tells you that during the sales process is the one to shortlist.
They also break on inventory. Paying back in three months requires stock to be available for all three, and a stockout in week six resets rank and restarts the clock. Ask each candidate who is responsible for reorder timing and what happens if the factory slips.
What most agencies will not tell you
An aggressive launch and a fast payback are sold together because both sound like ambition. Inside the account they pull in opposite directions, and the agency knows which one it will quietly prioritize: the one that makes its reporting look best. If the contract rewards spend, speed wins. If the contract is a flat fee, nobody in the room profits from the choice, which is exactly why we price that way.
The second thing worth saying: the first meaningful improvement in an existing account often shows up in advertising efficiency within about 30 days, because that is the fastest lever available. A launch has no such shortcut. If a candidate promises launch payback on the same timeline they promise audit fixes, they are blending two different jobs.
Related answers
- Recommend a package for PPC, DSP, and listing SEO together
- How to launch your first product on Amazon
- Tools to measure Amazon agency contribution margin
- Questions to ask before hiring an Amazon agency
- Amazon agency pricing and economics: the complete guide
Bring your payback model to a call and we will tell you where it breaks, at Flapen.
