Amazon PPC Is Not Broken. Your Campaign Structure Is. Here Is How to Fix It.

Your ACOS is climbing, your spend is up, and your sales are flat. So you lower bids, pause a few keywords, maybe blame the platform for getting more expensive. None of it moves the number for long. The problem is almost never the bids and almost never Amazon. It is the structure underneath your campaigns, and until you fix that, every bid change is just rearranging the same leaks.

This post shows you how good Amazon PPC management actually works: how to diagnose a broken account structure, the architecture that controls spend instead of chasing it, and the advanced moves that separate a profitable account from one that quietly drains margin every month.

Why bid tweaking never fixes a structural problem

Most underperforming accounts share the same root cause. Everything is jammed into a handful of automatic campaigns, or every keyword lives in one giant ad group with a single bid. When the structure is that loose, you lose the one thing that makes paid advertising profitable: control over where each pound goes.

Here is what that looks like in practice. An automatic campaign is spending well, so it feels like a winner. But inside it, three converting search terms are subsidising forty terms that have spent money and produced nothing. You cannot bid up the winners or cut the losers, because they all share one bid and one budget. You are paying full price for traffic that will never convert, and the good keywords are capped because the budget is being eaten by waste.

Lowering the campaign bid does not fix this. It just makes you less competitive on the keywords that were actually working. This is why bid tweaking feels like running to stand still. You cannot optimise what you have not separated.

Build the structure that controls spend

The fix is an account architecture that isolates intent so every keyword can be managed on its own merits. This is the foundation we build on every account before we touch a single bid.

Run automatic and broad campaigns as discovery, not as sales engines. Their only job is to surface search terms. Give them a modest budget and mine them constantly.
Harvest converting search terms into dedicated exact-match campaigns. When a term proves it converts, it graduates out of discovery and into its own exact-match ad group where you can fund it properly and protect it.
Separate branded, competitor, and generic terms into their own campaigns. These three have completely different ACOS expectations. Branded terms should be cheap and efficient. Generic terms cost more and build rank. Competitor terms are an investment. Mixed together, the cheap branded sales make the whole account look healthier than it is and hide the waste.
Add negative keywords as a daily habit, not a quarterly clean-up. Every search term that spends without converting becomes a negative in the discovery campaigns. This is the single highest-return task in PPC and the one most accounts neglect.
Cap each campaign by its role. Discovery gets a small fixed budget. Proven exact-match winners get room to scale. Defensive branded campaigns get enough to hold the buy box and no more.

What good looks like

A healthy account has dozens of tightly themed campaigns, each with a clear job, not three sprawling ones doing everything badly. When the structure is right, the daily work becomes simple: fund what converts, negate what does not, and graduate winners up the funnel.

This architecture is the backbone of how we run Amazon growth and PPC across the accounts we manage, and it is what holds a 4.37x average ROAS at a 23.4% ACOS steady even as a catalogue grows.

Tie every campaign to a goal, not a habit

Once the structure is in place, profitability comes from managing each campaign against the right target. Most accounts apply one ACOS goal to everything, which guarantees you overspend on some campaigns and starve others.

Set the target by the job the campaign is doing:

Branded defence: very low ACOS. You are protecting traffic that was going to convert anyway, so efficiency is everything.
Proven generic winners: moderate ACOS, managed to profit. These are your scalers. Fund them to the point where another pound of spend stops returning a profitable sale.
Rank-building campaigns: higher ACOS, on purpose, for a defined window. You are buying organic position. Measure success by what happens to your organic rank and total sales, not by the campaign ACOS in isolation.
Competitor conquesting: highest ACOS, treated as a customer-acquisition cost and reviewed against lifetime value, not a single order.

Common mistake

Judging a rank-building campaign by its ACOS and pausing it the moment the number looks high. That campaign was never meant to be efficient on its own. It exists to lift your organic position so you sell more without paying for the click. Kill it early and you lose the rank you were paying to build. Measure the right campaign against the right metric, or you will keep switching off the things that are actually working.

The lever most sellers never pull: total advertising cost of sale

Here is the advanced shift that changes how you run the whole account. Stop optimising ACOS in isolation and start managing TACOS, your total advertising cost of sale as a percentage of total revenue, including organic.

ACOS only tells you how efficient your ad spend was against ad-attributed sales. TACOS tells you whether your advertising is building a business or just renting sales. When your structure is working, paid spend lifts organic rank, organic sales grow, and your TACOS falls over time even if your ad spend stays flat. That is the signal of a healthy account: advertising that compounds into organic position instead of leaking.

If your TACOS is flat or rising month after month, your ads are not building anything. They are propping up sales that collapse the moment you pause. That is the difference between an account that scales and one that traps you in permanent spend. We treat this as a reporting discipline, which is why we build it into our analytics and reporting for every account, so the trend is visible and acted on every week.

Across the brands we have taken on, the pattern is consistent. The account was never broken because Amazon got expensive. It was broken because the structure could not separate the spend that built the business from the spend that wasted it. Fix the architecture and the same budget starts to compound.

How to read your search term report in ten minutes

The search term report is where a structured account is actually managed, and most sellers never open it. Here is the ten-minute routine that keeps the structure clean once it is built.

Filter to the last 30 to 60 days so you have enough data to judge a term fairly. A term with two clicks tells you nothing.
Sort by spend, highest first. Look down the list for terms that have spent real money with no orders, or with an ACOS far above your target for that campaign type. These are your bleed points.
Negate the irrelevant spenders in your discovery campaigns. A term that has spent the price of two units with zero orders is not unlucky, it is wrong for your product. Add it as a negative exact or negative phrase.
Promote the proven winners. Any term converting consistently at or below target gets graduated into its own exact-match ad group, where you fund it properly and add it as a negative in the discovery campaign so the two never compete.
Note the close variants. Terms that nearly convert often need a listing fix, not a bid change. If shoppers search a feature you do not mention in your title or bullets, that is a conversion gap, not an ads problem.

Run this weekly and the account manages itself. Skip it and even a perfect structure silently fills back up with waste. This single habit is responsible for more recovered margin than any bidding strategy, and it is the first routine we put in place when we take over an underperforming account.

What to do next

If your Amazon PPC is underperforming, work in this order:

Stop tweaking bids and audit the structure first.
Separate discovery, branded, generic, competitor, and proven-winner campaigns.
Set an ACOS target per campaign based on its job, not one number for everything.
Make negative keywords a daily habit.
Track TACOS as the real measure of whether your advertising is building or renting.

A profitable account is mostly architecture and discipline, not clever bidding. Get the structure right and the optimisation becomes obvious.

Start with a free audit. We will tell you exactly what is holding your brand back and what a 90-day plan to fix it looks like for your specific channels. You can book yours here.