Amazon advertising reports are read three ways, and each way answers a different decision. A raw export from the console tells you what happened, and a weekly written summary from whoever runs your account tells you what changed and why. A live review decides what happens next.
The short version
- Three habits, not three reports. The export, the written summary, and the live review close different decisions, and buying one does not cover the other two.
- A raw export answers what happened. It carries no owner and no next action, so nothing on it moves until a person writes a sentence about it.
- About 1.4 brands per operator is our published ratio, 50 operators against about 70 brands, and caseload is what decides whether a written summary can arrive every week.
- Our cadence is a written update in Slack every week and a live review every two weeks. Ask any provider for both cadences in writing before you sign anything.
- Read every number against your own margin per product. An account average hides the one product losing money on every advertised sale.
Three reporting habits, and what each one can decide
Most sellers searching for Amazon ads reports already have the exports sitting in a folder. What they lack is a habit that turns an export into a decision with a name attached to it. Three habits exist, and they are not interchangeable.
| The habit | The decision it can close | What it cannot do |
|---|---|---|
| The console's own reports, read raw | Confirm what a campaign spent and returned across a window you select | Explain why a number moved, or say who is accountable for moving it back |
| A weekly written summary from whoever runs the account | Catch drift in the week it starts, because a person names the change and the result | Settle budget, catalog, and pricing questions that need a conversation |
| A live review every two weeks | Decide budget, catalog, and whether a product keeps its advertising at all | Replace the weekly record, because week one is forgotten by week three |
Our own cadence as of September 2026 is the third and second rows together, with Slack access in between.
Here is the decision rule, in one sentence. Buy the habit that matches the decision you are behind on, and never accept a reporting cadence slower than the decision it is meant to feed.
The caseload sitting behind a written report
A weekly written summary is not a software feature you switch on. It is an hour of a named person's attention, and attention divides by caseload before it divides by anything else.
Fifty operators run about 70 brands by hand here, which works out at about 1.4 brands per operator. We publish that ratio because it is the number that makes a weekly written update and a live review every two weeks physically possible.
So run the same arithmetic on anyone before you read a sample of their reporting. Divide the accounts under management by the people who manage them, and insist on both figures in the same sentence.
An account manager carrying twelve brands can still send you something every week. What they cannot do is read your campaigns, your listing, and your stock position before writing it, so the file arrives full of numbers and empty of judgment.
That division of labor is also why our clients spend about 2 hours a month with us after onboarding, and 4 to 6 hours a week while a product is launching. The reading happens on our side, and a measurable ACoS improvement typically shows inside the first 30 days.
Match the decision to the habit, never the reverse
Sellers usually inherit a reporting habit instead of choosing one. So work backwards from the decision actually sitting on your desk this week, then pick the report that can close it.
| The decision on your desk | The habit that closes it | The cost of reading the wrong one |
|---|---|---|
| Whether last month of spend earned its keep | A raw export read against your own margin, product by product | An account average that quietly hides the product bleeding money |
| Whether the person running the account is still working | A written summary that names the change, the reason, and the result | Months of steady spend with nobody able to name a single decision |
| Whether this product keeps its budget at all | A live review reading the four signals across 60 to 90 days | Capital committed to a product the data had already rejected |
The four signals are rating trend, return rate, conversion rate, and cost of customer acquisition trajectory.
The same rule holds in reverse. Name the decision first and then pick the report, because a report chosen before the decision only ever confirms what you already spent.
What most agencies will not tell you about the reports they send
Three gaps sit inside almost every reporting pack sold in this market, and on a careless month that includes ours. Each one is the same matrix turned around and pointed at the provider.
| What the report shows | What it leaves out | The question that closes the gap |
|---|---|---|
| Spend and sales side by side, both rising | Profit per product after landed cost, fees, and returns | Show me contribution margin next to advertising cost, by product |
| A list of optimizations made this week | Which of them were reversed, and what the reversal cost | Name the last three things you turned off and what followed |
| A cadence promised in the proposal | The caseload of the person who has to write it every week | How many other brands does the person writing my report carry |
One rule settles all three. A line that cannot change a decision you own is decoration, so ask for it to be cut and replaced with a line that can.
Hold us to exactly that. Our updates name an ASIN, a number, a fix, and the person who owns it, and if yours reads like an export with a logo on top, stop paying us. The contract runs month to month on 30 days' notice, so nobody has to wait out a year to act on that judgment.
Related answers
- Amazon ad types
- Amazon ads case studies
- How to track TACoS without losing margin sanity
- Rank providers for DSP and PPC combined audits
- Amazon account measurement and audits: the complete guide
Here is the free step for this week. Export the last 60 days for one product, then add two handwritten columns beside it, your contribution margin before ad spend and the decision that number changes. Every row that changes no decision comes off next month's report.
For a written read of that account across seven areas, ranked by the money each fix moves and back inside 48 hours at no cost, ask Flapen.
