Start With the Product’s Stage, Not a Target ACOS

The first thing we decide on any account is not the bid. It is what each product is trying to do right now.

A product that launched last month and a product that has held its rank for two years are not the same advertising problem, and a single account wide ACOS target treats them as if they were. The launching product needs impressions more than it needs efficiency, because it is buying rank and reviews that pay back later. The mature product has nothing left to buy but efficiency, so it should run tight. Force the same ACOS on both and you overspend on the one that is already winning and starve the one that still needs the runway.

So we sort every ASIN into a stage before we touch a campaign. Launch, growth, mature, and defend. Each stage gets its own ACOS band and its own budget logic. This is the single decision that most account overhauls skip, and it is the one that makes every later decision easier. We break the full framework down in ACOS targets by intent tier, because a single account average hides more than it tells you.

How to run it this week:

• Tag every ASIN by stage. Mark each one launch, growth, mature, or defend based on how long it has held rank and whether reviews are still building.
• Set a separate ACOS band per stage. As a starting reference, Amazon’s own guides put a healthy ACOS in the 15 to 30 percent range and needs work above 35 percent, but a launch product should sit above that band on purpose.
• Confirm inventory before you buy demand. Winning clicks on a launch product and then going out of stock wastes the spend and drops your rank on the way down.

Build the Structure So Every Campaign Has One Job

Once each product has a stage, the campaigns follow. The mistake we see most is one campaign carrying every keyword on one budget. It reads simple and it kills your ability to optimize anything, because you can never see which term is doing the work.

We build so every campaign has a single job. For a typical mid tier account, that means a research layer, a scaling layer, and a defense layer running at the same time.

The research layer is where you discover what actually converts. An auto campaign surfaces terms you would never have guessed, and a broad campaign expands around the ones that work. The scaling layer takes the proven terms out of research and gives them their own exact match campaign at a slightly higher bid, because a proven term deserves its own budget and its own bid, not a share of a research pool. The defense layer holds your own ground. A branded exact campaign keeps competitors off your brand searches, and a product targeting campaign protects your own detail pages while conquesting the competitors worth taking.

Sponsored Brands and Sponsored Display sit on top of this once the foundation holds. Sponsored Brands puts your logo and a video across the top of the results page and lifts click through in most categories. Sponsored Display retargets the shoppers who viewed and did not buy. Run Sponsored Products alone and you leave the brand undefended, which is exactly where competitors pick off your own customers. The full build for scaling accounts lives in the PPC architecture behind 7 figure brands.

How to run it this week:

• Break up the catch all campaign. Split any single campaign that carries every keyword into a research layer, a scaling layer, and a defense layer so each one carries a single job.
• Add a branded exact campaign. Without it, competitors bid on your brand term and take customers who were already looking for you.
• Put a video on your hero. If you are Brand Registry enrolled and not running Sponsored Brands video, add one on your strongest product. Amazon’s own guides call a missing hero video one of the biggest click through lifts left on the table.

Read the Search Term Report Every Week and Negate

The single biggest source of wasted spend we find on a new account is the same one every time. Search terms that take clicks and never convert, running week after week because nobody read the report.

The Search Term Report is the account telling you exactly where the money is leaking. We pull it weekly and sort every term into three buckets. Profitable terms get moved to their own exact campaign at a slightly higher bid so they can scale. Marginal terms stay under watch. Wasteful terms, the ones that have spent well past your target with no conversion, get negated at the campaign level so you stop paying for them.

This is not a one time cleanup. The terms shift every week as competitors change bids and new products enter your category, so an account left alone for a month drifts every time. Running without negatives is what pushes 20 to 40 percent of spend onto irrelevant queries, per the sourced industry range in our internal Amazon reference. The teams that win are the ones who read the report every week, not the ones who set it and check back in Q4.

How to run it this week:

• Pull the last 14 days. Open your Search Term Report for the trailing two weeks and sort by spend descending.
• Negate the leaks. Negate every term that has spent more than twice your target cost per acquisition with no conversion in that window.
• Promote the winners. Move your top converting terms into their own exact match campaign at a slightly higher bid so they get the budget they have earned.

Manage to Contribution, Not to the ACOS Line

Here is where most strategies quietly go wrong. They manage ACOS as if it were the goal.

ACOS is ad spend divided by ad revenue. It is a constraint, not a goal. The goal is profitable contribution per product per week. The number that actually tells you whether your ads are building the business is TACOS, ad spend divided by total revenue across organic and paid. When ACOS holds steady while TACOS falls over time, your organic rank is climbing and your ads are pulling the whole listing up. That is the pattern you are managing toward, and you cannot see it if you only watch the ACOS line.

Real numbers make this concrete. A US beauty brand we worked with cut ACOS from 42 percent to 18 percent inside 90 days, and the lever that drove most of that lift was a product level negation pattern most teams skip, not bid management. That is the whole point of this section. The win did not come from chasing a lower ACOS directly. It came from cutting the spend that was never going to convert, which lowered ACOS as a result rather than as a target.

On the multi marketplace accounts we manage, the same logic holds across geographies. A European brand we run across Germany, France, Italy, and the United States saw US sales jump 96 percent week over week after we reverted from over focused campaigns back to auto and rebuilt the image set. You can see how that engagement ran week by week in the SafePro X case study.

ACOS is a constraint, not a goal. The goal is profitable contribution per product per week, and TACOS is the number that tells you whether your ads are building the business or just renting sales.

Point the Budget at 2026, Not 2022

The surfaces where your ads show up are changing, and a 2026 strategy has to account for that instead of running the same placement mix from three years ago.

Two shifts matter right now. First, ads inside Amazon’s AI assistant. Sponsored Products Prompts and Sponsored Brands Prompts, the conversational ad units inside Amazon’s AI shopping assistant (Alexa for Shopping in the US, formerly Rufus), reached general availability in the United States on March 25, 2026 per Amazon Ads, and eligible auto and broad campaigns auto extend into these AI surfaces. A growing share of impressions is coming from AI answers rather than the classic results page, which means your research layer is quietly buying placement in a new surface whether you planned for it or not. Second, Sponsored TV is now fully self serve with no minimum spend and open to any Brand Registry seller, running across Prime Video, Fire TV, Twitch, and Freevee per Amazon Ads. The old story that streaming ads required a large managed service budget is no longer true.

You do not need to chase every new surface. You need to make sure your budget is pointed at where your buyers actually are. For most brands that is still Sponsored Products closest to the moment of purchase, but the placement adjustments now carry real weight. Amazon’s own placement controls let you push Top of Search, which is often worth a plus 50 to 100 percent bid on hero ASINs, so the highest converting placement gets the budget instead of an even spread.

How to run it this week:

• Check your AI surface eligibility. Confirm your auto and broad campaigns feed the AI assistant surfaces with terms you actually want representing the brand in an AI answer.
• Push Top of Search. Turn on placement bid adjustments and weight Top of Search on your hero ASINs, where the highest intent traffic sits.
• Price out Sponsored TV. If you are Brand Registry enrolled and have been told streaming ads are out of reach, price it directly. There is no enforced minimum.

Action Checklist: Five Moves This Week