Most brand owners cannot answer a simple question: is my agency actually growing my business, or just keeping it busy? The monthly report looks full. There are charts, percentages, a paragraph about "continued optimization." Sales are flat or up a little. Spend is up too. You sign off because nothing looks broken. That is exactly how an underperforming agency survives for a year longer than it should.
The problem is that activity is easy to show and results are easy to hide. An agency can adjust a hundred bids, add fifty keywords, and "refresh" your A+ content every month while moving none of the numbers that decide whether your brand makes money. Real management and busywork produce the same-looking report. The difference is in what the report measures, what it leaves out, and whether anyone can connect the work to a dollar figure you keep.
The metrics that flatter, and the ones that tell the truth
Start by sorting your reporting into two buckets: vanity and decision.
Vanity metrics go up and to the right and tell you almost nothing. Total revenue is the worst offender. Revenue climbs when you discount, when you spend more on ads, when a competitor goes out of stock, and when the category has a good season. None of those mean your agency did anything. Impressions, total clicks, and "keywords ranked" belong in the same bucket. They describe motion, not progress.
Decision metrics change what you would do next. Contribution margin is the anchor. If your agency is not reporting profit after product cost, Amazon fees, and ad spend, they are managing a number that does not pay your bills. We have written before about why contribution margin and not revenue should drive every Amazon decision, and the same logic applies to judging the people you pay. Ask for margin by product, not blended across the catalog, because blended numbers let one strong SKU hide three that are bleeding.
If your monthly report cannot tell you which product made you the most money last month, you are paying for activity, not management.
The other decision metrics worth demanding: ACoS and TACoS by product, organic rank for your priority keywords, conversion rate on your hero listings, and unit velocity on the SKUs you are trying to grow. Five numbers, tracked per product, beat a fifty-slide deck.
What a real report actually contains
A report from an agency that manages, rather than maintains, has three things busywork reports lack.
A decision log, not an activity log
An activity log says "optimized campaigns, updated negatives, adjusted bids." A decision log says "moved budget from Sponsored Brands to Sponsored Products on the kitchen line because Sponsored Products was converting at twice the rate, expected to recover roughly two points of ACoS." One is a list of chores. The other is a thesis you can check next month. If you want to see whether your team is reading data or guessing, ask how they decide what to negate and harvest. A strategist runs a repeatable system for reading the search term report every week, and they can describe it in plain language.
A target for every product, set on purpose
Good agencies do not chase a single account-wide ACoS goal, because the right number is different for a new launch than for a mature cash cow. Each product should have a target derived from its own economics. If your agency cannot tell you how they set a target ACoS for each product from its contribution margin, they are using one number for everything and calling it strategy.
Honesty about what went wrong
Every account has a bad month. A listing gets suppressed, a launch stalls, a competitor undercuts you on price. The tell is not whether problems happen. It is whether your agency surfaces them before you notice, names the cause, and shows the fix. An agency that only reports wins is hiding the losses, and the losses are where your money goes.
The questions that expose busywork
You do not need to read a spend report line by line to test your agency. You need four questions, asked on a call, watching how fast and how specifically they answer.
- "Which product made the most contribution margin last month, and why?" A real manager answers in seconds with a number and a reason. Hesitation means they track revenue, not profit.
- "What did you change in the last 30 days, and what did you expect it to do?" You want a short list with a predicted outcome attached to each item. "Ongoing optimization" is not an answer.
- "Which of our keywords are we winning organically, and which are slipping?" This separates teams who treat ads and listings as one connected system from teams who run PPC in a silo and never look at rank.
- "What is the single biggest thing holding this account back right now?" A team that is actually in your account has an opinion. A team that is coasting will give you a generic answer about "increasing ad spend."
If the answers are slow, vague, or always point to spending more money, the work behind the report is thinner than the report suggests.
Activity that is real work, and activity that is theater
To be fair, some repetitive work is the job. Pulling the search term report weekly, maintaining negatives, watching for hijackers, and refreshing creative when conversion dips are all real management. The difference is whether the activity is tied to a result.
Theater looks like this: bid changes with no stated reason, keyword counts reported as progress, A+ content "updates" that move no conversion number, and reports that grow longer as performance flattens. Real work looks like this: fewer changes, each with a reason and a measured outcome, and a report that gets shorter and clearer over time because the account is under control. When the lever being pulled is conversion, a strong team can show you the specific moves they made to lift conversion without touching price and the rate change that followed.
What to do this week
Pull your last three monthly reports side by side. Do one thing: try to trace any single decision from "we changed X" to "here is the dollar result." If you cannot draw that line on any of the three reports, your agency is producing activity, not management, and the conversation you need to have is overdue.
Then ask the four questions on your next call. Write down how long each answer takes and how specific it is. You are not looking for perfect numbers. You are looking for a team that knows your account, thinks in contribution margin, and can connect their work to your profit. If they can, keep them. If they stall, you now know exactly what you are paying for.