Amazon PPC Structure for Small Catalogs

Most Amazon PPC advice is written for catalogs with hundreds of ASINs and enough conversion data to split by match type, placement and product. A small catalog does not have that luxury. Ten SKUs and a limited budget cannot support twenty campaigns, and an attempt to build them usually produces a structure that is busy, expensive to manage and statistically empty.

The goal for a small catalog is the opposite of sophisticated: concentrate spend and data where they can produce a decision.

Start with the economics

Before you create a campaign, know the numbers that decide whether a click is worth buying:

• The product's gross margin after Amazon fees and fulfillment.
• The break-even ACoS implied by that margin.
• The price and the room to discount without losing money.
• The return rate and the cost of returns.
• The repeat purchase rate if the product is consumable or part of a routine.

Break-even ACoS is not a target. It is the ceiling. A launch may deliberately spend above it to build rank and reviews, but that is a decision with a time limit, not a permanent setting.

If you do not know the margin, stop here. No campaign structure fixes a product that loses money on every sale.

A structure that fits ten to fifty SKUs

For a small catalog, start with four campaign types, not twenty.

1. One auto campaign for discovery

Run a single auto campaign across the products that share economics and a category. Auto targeting finds search terms you did not think of. Keep it on a modest budget and review the search term report on a fixed schedule.

If the catalog has two very different product lines, split auto into two campaigns, one per line. Do not split by SKU at this stage. You will starve each campaign of the clicks it needs to learn.

2. One manual exact campaign for proven terms

Move converting search terms out of auto and into a manual exact campaign. This is where you control the bid and budget for terms that already work. Keep the list short and the bids deliberate.

Do not move a term after one click. Give it enough clicks to make a judgment, and write down what "enough" means for your margin before you look at the report.

3. One manual phrase or broad campaign for controlled discovery

This is optional. If auto is producing useful terms and you want more control over match expansion, add one phrase campaign. If it is not, skip it. A second discovery campaign with no budget is not a strategy.

4. One product targeting campaign

Target complementary ASINs and, where the category allows, competitor detail pages. This is often the most efficient placement for a small catalog because the shopper is already comparing. Build the list deliberately: similar price, use case and review profile.

Add brand defense only if people search your brand. For a new brand, that campaign may have no volume, and that is fine. Do not create it because a template says to.

Why not one campaign per SKU

It fragments data. Each campaign needs its own budget, learning period and search term review. With ten SKUs, you end up with ten small campaigns that each spend too little to optimize, plus a reporting job that takes a day.

Group products by economics and shopping behavior, not by SKU count. A consumable refill and a durable accessory may belong in different campaigns even if they sit in the same category.

Naming that survives growth

Use a consistent name before you need it:

[market] | [product group] | [campaign type] | [targeting] | [budget tier]

For example: US | skincare-core | manual exact | hero terms | tier 1.

The exact format matters less than using it every time. When the catalog grows, the names are how you find a campaign, compare performance and hand the account to someone else without a spreadsheet legend.

Budget and bidding

Give the discovery campaign enough budget to run through the day. A campaign that exhausts its budget by mid-morning only learns about the morning. If the budget is genuinely small, reduce the number of products in the campaign rather than spreading it thinner.

Start bids conservatively and adjust based on impressions and clicks, not anxiety. Test top of search and product pages separately once there is enough volume to compare. Do not stack placement modifiers before you have a baseline.

Review cadence

Weekly:

• Search term report: add negatives, move winners to exact.
• Budget pacing: which campaigns run out, which never spend.
• Relevance: irrelevant terms waste a small budget quickly.

Monthly:

• ACoS and TACoS by campaign.
• Organic rank and sales for the products the ads support.
• New-to-brand share, if the brand has access to it.
• The break-even ACoS again, because fees and costs change.

Do not change bids daily. You will chase noise and never learn what a change did.

Judge the account on the business

ACoS is a ratio. It can improve because sales fell, not because the ads got better. TACoS, total ad spend divided by total sales, shows whether ads are building the product or buying sales you would have made anyway. Organic rank and review velocity tell you whether the spend is compounding.

For a small catalog, the question is not how to scale to 40 campaigns. It is which three products deserve the next dollar. The structure should make that question easy to answer.

Common mistakes

• Too many campaigns for the data available.
• No negative keyword routine, so discovery spend leaks.
• Moving terms to exact before they have enough clicks.
• Judging a launch in the first week.
• Ignoring organic rank and treating PPC as the only channel.
• Setting an ACoS target without knowing the margin.

If you want the account managed, our Amazon PPC management service covers structure, search term work and reporting, and the wider Amazon practice covers listings, A+ content and brand operations.