A brand doing $200K a month on Amazon typically has a PPC freelancer, a listing copywriter, someone handling FBA shipments, and a VA working account health cases. Every one of them is competent. The account has been flat for eight months.
This is not a skills problem. It is a structural one, and it produces the same handful of failures in every account where it exists.
The four failures, in order of cost
Advertising scales into a stockout. The PPC manager sees improving efficiency and increases budget. Nobody tells them the product has nineteen days of cover. Three weeks later the listing goes out of stock at peak velocity, organic rank decays, and the position bought over four months has to be repurchased from scratch. The full bill for that is much larger than the missed units.
A listing rewrite breaks the campaigns. New copy ships. The keyword the top-performing exact-match campaign was built around no longer appears in the title. Rank degrades over the following weeks. Nobody connects the two events because the people involved never spoke.
A price change wrecks advertising efficiency. Retail price drops 12% for a promotion. Bids, calculated from contribution margin at the old price, are now above break-even on every term. The campaigns keep running exactly as configured, quietly unprofitable.
Inventory is bought against advertising nobody plans to run. Six months of stock arrives for a launch push that was never budgeted. The units sit, age into higher storage bands, and consume capital.
Each is a coordination failure. None is fixed by any of the individuals involved doing their own job better.
Why the split feels sensible
Amazon presents its functions as separate interfaces. Advertising lives in one console, inventory in another, catalog in a third, account health somewhere else again. Dividing the work by interface is the obvious move, so most brands make it.
But the underlying business does not divide that way. Stock cover determines what bid strategy is rational. Listing quality determines what a click is worth. Price determines break-even ACOS. Advertising velocity influences organic rank, which changes what inventory you need.
These are one system with four dashboards.
Stock cover is a bidding input
The clearest example, because it is the most expensive.
There is no correct bid for a product in isolation. There is a correct bid for a product with four months of cover and a different correct bid for the same product with three weeks of cover.
Deep cover: scale. Buy placement, accept higher ACOS while acquiring rank, push velocity.
Thin cover: protect. Defend branded terms and existing rank, stop paying to acquire customers who will find you unavailable.
Getting this backwards is routine when advertising and inventory sit with different people.
Conversion rate is an advertising input
Cost per acquisition is a function of bid and conversion rate. Any listing change that moves conversion rate moves advertising economics immediately.
So listing work should be sequenced before spend increases, not run in parallel by someone who does not know a spend increase is coming. And when copy changes, the campaigns built on the old copy need reviewing in the same week.
The diagnostic: which number is lying to you
Flat revenue with healthy ACOS is the most common presentation, and it has a specific tell.
If ad spend rose while total sales stayed flat, advertising is replacing organic sales rather than adding to them. ACOS cannot see this, because ACOS only measures advertising against advertising. TACOS, spend against total sales, catches it immediately. The ACOS vs TACOS piece covers how to read the two together.
What good coordination actually requires
Not more meetings. A weekly loop where the same person looks at four things in a fixed order:
- Inventory position. What has cover, what is thin, what is aging.
- Advertising performance. What converted, what leaked, where placement moved.
- Catalog health. Suppressions, variation integrity, content that no longer matches the campaigns.
- Account health. Policy warnings, IP complaints, ODR movement.
Then one decision: what is the binding constraint this week? Usually only one thing is genuinely holding the account back, and effort spent elsewhere is effort spent optimising something that was not the problem.
The uncomfortable conclusion
Most brands would get more from consolidating marketplace execution under one operator than from hiring a better specialist for any individual function.
The specialist model works when functions are genuinely independent. On Amazon they are not. The account that grows is usually not the one with the best PPC manager. It is the one where somebody can answer “should we push this product harder right now?” with reference to stock, margin, listing quality and rank at the same time.