When a seller's Amazon PPC account stops performing, when the ACoS creeps up, and the ROAS drifts down without any obvious cause, the first instinct is to change bids. Or add more keywords. Or launch a new campaign type. These things might help. But the more common root cause is something less visible and less exciting: the campaign structure itself was set up in a way that makes it impossible to know where the money is actually going.
Imagine trying to run a household budget where the grocery shopping, the restaurant bills, and the takeaway coffee all go to the same account with the same label. At the end of the month, you know you overspent on food. You don't know where. You can't fix what you can't see.
Amazon PPC structure works the same way. When campaigns are organised well, every dollar has a clear jo,b and optimisation decisions are obvious. When they are not, you are making changes in the dark and hoping something improves.
This guide explains how to build the right structure from the beginning, and what to change if you didn't start that way.
Why Structure Is the Foundation, Not an Admin Detail
Campaign structure is not a naming convention exercise. It is the system that determines whether your optimisation decisions are based on real signal or muddied data.
Automatic and manual targeting are selected at the campaign level. An automatic-targeting campaign and a manual keyword-targeting campaign are separate campaigns and cannot run together in the same ad group. Broad, phrase, and exact versions of a keyword can have different bids and performance. Amazon provides keyword- and target-level performance data, so advertisers can evaluate them individually even when they share a campaign budget.
With average CPCs rising year over year, a disorganised account wastes more money per misallocated click than it did two years ago. The structure question has always mattered. It matters more now.
The Three Levels of Amazon Campaign Organisation
Level 1: The Account: What Sits Above Everything
Your Amazon Ads advertising account contains your advertising campaigns and associated advertising profiles. Access and available products depend on your advertiser type, marketplace, permissions, and account setup.
Amazon Ads provides campaign-level budgets and also supports additional budget-management capabilities, such as portfolios and account-level budget control,s where available. Portfolios can be used to organize related campaigns and, where supported, manage budget controls across campaigns grouped within the portfolio.
Level 2: The Campaign: Where Budget and Strategy Live
The campaign is where the two most important settings sit: your daily budget and your bidding strategy. Every ad group inside a campaign shares these settings. That is why the campaign is the budget container: not the ad group.
When you set a $40 daily budget at the campaign level, all the ad groups inside that campaign compete for the same $40. The daily budget is set at the campaign level and is shared across the campaign's ad groups. Individual ad groups do not independently exhaust the campaign budget. Once the campaign's available budget is exhausted, ad delivery can stop until budget becomes available again. This is why keeping unrelated products in the same campaign, with very different spend profiles, is structurally problematic.
Level 3: The Ad Group: Where Products and Targeting Meet
The ad group is where specific products and their targeting instructions live together. Products and targeting are organized within ad groups. Targets within an ad group can make the advertised products in that ad group eligible to serve, while bids can be managed at the applicable target or ad-group level. Placement bid adjustments are configured at the campaign level.
This is the critical implication of ad group structure: Products within an ad group share its targeting structure, but keyword and product-target bids can be managed individually. Grouping products with similar characteristics can still make campaign management and optimization easier. Products with significantly different prices, margins, and conversion rates can have different profitable CPC levels, which is one reason advertisers may choose to separate them. This is the price range grouping rule in its simplest form.
The Auto-Plus-Manual Pipeline: The Core of Every SP Account
Auto Campaigns: The Fishing Net
An automatic campaign is like casting a wide fishing net. You set a budget, you set a bid, and Amazon decides which searches to show your ad against: based on your listing content, your category, and its own signals about purchase intent. You don't pick the keywords. Amazon does.
Auto campaigns discover what works. They surface search terms and ASINs that convert, many of which you wouldn't have thought to target manually. Automatic targeting can be useful for new product launches because it can help advertisers discover relevant shopping queries and products. It is not mandatory for every new product.
Manual Campaigns: The Fishing Rod
With manual targeting, advertisers select the keywords or products they want to target. Keyword match types still determine how broadly those keywords can match shopper queries. You decide the bid for each target individually. You are not casting wide; you are targeting precisely.
Manual campaigns can use insights discovered through automatic campaigns, along with keyword research, historical performance data, product research, and other inputs. Once a search term converts in auto, you take it, add it to a manual campaign at a considered bid, and give it the individual attention it deserves. This is how search terms move from discovery into deliberate targeting.
The Negative Keyword Bridge That Connects Them
Here is where most beginners get the structure wrong. Automatic and manual campaigns can both be eligible for related shopper queries. Advertisers may use negative targeting when they want greater control over which campaign handles particular traffic. When a search term is added to your manual campaign at an exact bid, your auto campaign might still be bidding on the same search: potentially at a different amount, creating internal competition.
Using negative keywords after promoting successful search terms into manual campaigns is one possible campaign-structure strategy. It can be used when advertisers want greater control over how traffic is distributed between automatic and manual campaigns. When a search term converts and you move it to a manual campaign at an exact bid, you add that same term as a negative keyword in your auto campaign. This is like installing a toll gate on the route you have already optimised: it tells the auto campaign, 'don't go this way anymore: we've got that one handled.'
Running automatic and manual campaigns without cross-negation does not inherently mean they are cannibalising each other. Advertisers can use negative targeting when they want to isolate traffic more deliberately. With it, they work as a pipeline: auto discovers new converts, manual captures and scales them, and negatives keep the traffic flowing in the right direction.
When broad, phrase, and exact match keywords all live in the same campaign, or worse, the same ad group, they each draw from the same budget, and you can't bid on them appropriately. Advertisers can test different bids across match types. An exact match may justify a different bid from a phrase or broad match, but actual bids should be based on observed performance rather than assuming an exact match will always convert at a higher rate. Campaign-level ACoS combines performance across targets within the campaign, but keyword- and target-level reporting can still be used to identify which keywords and match types are performing efficiently. You cannot optimise what you cannot isolate. Amazon allows advertisers to use broad, phrase, and exact match types within manual keyword campaigns. Separating match types into different campaigns is an optional structural strategy when greater budget or campaign-level control is required. Your broad match keywords live in one campaign. Phrase match in another. Exact in another. This means you can set different bids for each, monitor performance independently, and pause or scale each type without affecting the others. Think of it like tracking your exercise habits. If you log running, swimming, and strength training all as 'exercise' in the same spreadsheet with no labels, you cannot figure out which one is improving your fitness and which one is wasting time. The data is there. It just can't tell you anything useful. The same applies to match types. The search term report will eventually tell you which specific searches are converting. Still, campaign-level ACoS provides an overall view of campaign efficiency. For more granular optimization, advertisers should also review keyword-, target-, search-term-, and match-type performance. The instinct is to group products by how they are related to each other. Blenders in one ad group, juicers in another. All the blue products together, all the red ones separately. This feels logical from a product perspective. It does not work from a bidding perspective. Group products when their targeting, bidding strategy, and advertising objectives make sense together. Similar price points and conversion profiles can be useful considerations, alongside product relevance, category, margins, and other business factors. Products with substantially different prices or economics may benefit from being separated when they require different bidding or targeting strategies. Price difference alone does not make an ad-group structure incorrect. The profitable CPC for a product depends on factors including selling price, margin, conversion rate, advertising costs, and the advertiser's target ACoS or ROAS. It cannot be calculated from product price and margin alone. You cannot set one bid that works for both. Price range grouping means the bid you set for the ad group makes economic sense for every product in it. Grouping products with similar characteristics can simplify campaign management, but individual keyword or product-target performance should still determine appropriate bids. For high-spend, high-stakes products: your best-selling ASIN, your most profitable item, your product going into a promotional event: a single-product ad group can make product-level analysis and targeting easier to isolate. However, bids are associated with the applicable keyword, product target, or ad-group targeting configuration rather than simply being a single bid attached to the advertised ASIN. The data is entirely attributable to that one product. Single ASIN ad groups are not required across the board. For large catalogues with many similar-priced products, grouping by price range is more manageable and still gives clean enough data. Reserve single ASIN ad groups for the products that justify the additional campaign management overhead. A filing cabinet with no labels is still a filing cabinet. But when you need to find something, you will be pulling out every drawer. Your campaign names are the labels on your drawers. A consistent naming convention means you can filter, sort, and find campaigns instantly: in Campaign Manager, in spreadsheets, and in automated rules. Without it, every optimisation task starts with five minutes of orientation work to understand what you are looking at. Sponsored Products product targeting lets advertisers target specific products or categories. Eligible ads can appear in shopping results and on product detail pages. The reason is simple: Performance can differ across placements and targeting approaches. Advertisers should evaluate placement and target-level performance rather than assume one targeting type will always convert differently. A shopper on a competitor's product page is comparison shopping. Their intent is different from a shopper who typed a search term and is scanning results. Keyword targeting and product targeting may require different bids based on their observed CPC, conversion rate, ACoS, ROAS, and strategic role. Create dedicated campaigns for: direct competitor ASIN targeting (products with lower ratings or higher prices than yours) and category targeting (broad product categories). Keep these separate from keyword campaigns so you can set independent bids and monitor performance without the keyword data contaminating the product targeting numbers. Structure is not a one-time setup. It is a living system that needs a weekly routine to stay clean and effective. Think of it like a garden: planting is the work you do once. Weeding and harvesting are the work that determines whether the garden actually produces anything. The weekly cycle in a well-structured SP account looks like this: Monday: Review budget pacing: are any campaigns exhausting the budget early in the day? Check top-spend campaigns for unexpected week-over-week changes. Tuesday: Pull the search term report. Identify converting terms in auto campaigns to harvest into manual. Identify non-converting, high-spend terms to add as negatives. Midweek: Add harvested terms to the appropriate manual campaign (exact match if the term has strong conversion data, phrase if it needs more testing). Add negatives to the auto campaign. End of week: Review keyword-level performance in manual campaigns. Adjust bids for keywords that are above or below your ACoS target. Brands with Amazon Marketing Cloud access can layer attribution data on top of this weekly review to identify which keywords are driving the highest lifetime value customers beyond last-click attribution. When an account grows beyond 20-30 campaigns, this weekly cycle becomes time-consuming enough that automation starts to earn its place. Rule-based bid automation: where systems like Hector Ai apply your adjustment logic continuously rather than once a week, extends the same discipline without requiring proportionally more analyst time. The structure enables the automation. Without a clean structure, automated rules operate on ambiguous data and produce inconsistent results.Match Type Separation: One Match Type Per Campaign
Broad, Phrase, and Exact in Separate Campaigns
Why Mixing Match Types Breaks Your Ability to Analyse Spend
Ad Group ASIN Grouping: The Rules That Actually Matter
Group by Price Range, Not by Product Family
The Single ASIN Ad Group: When It Makes Sense
Naming Conventions: The Filing Cabinet Principle
Consistency matters more than the specific format. Pick a convention, document it, and apply it to every new campaign. The benefit compounds over time: every future optimisation task starts from a place of instant clarity rather than reorientation.Product Targeting as a Separate Track
The Weekly Harvest Rhythm
The Three Most Common Structural Mistakes