Amazon PPC Management: The Complete Guide for Sellers

Key takeaways: Amazon PPC management is a weekly cycle, not a setup task. Start with one objective and one number, audit before you spend, give every campaign a single job, set bids by the job each keyword does, and read the search term report every week. Judge the whole program on profit, not on ACoS alone.

What Amazon PPC management is

Amazon PPC management is the ongoing work of deciding where your advertising money goes on Amazon and checking what it brought back. It covers Sponsored Products, Sponsored Brands and Sponsored Display, and for some brands Sponsored TV and Amazon DSP. The work splits into three parts: building the account so that each campaign has one purpose, changing bids, budgets and targets every week on evidence, and reporting the result in numbers the business cares about.

It matters because Amazon advertising is an auction that never stops. Bids that were right in March are wrong in July. Search terms that converted last quarter attract new competitors this quarter. An account that is set up well and then left alone drifts, usually toward higher ACoS and lower rank. Management is what stops the drift.

How the cycle works

Every well-run account follows the same loop. The details change with the product, the budget and the marketplace, but the order does not.

The rest of this guide walks through each box in that loop.

Step 1: Pick one objective and one number

Most PPC accounts fail at the first step. They are asked to launch a product, lower ACoS, grow sales and protect margin all at once. Those goals pull bids in opposite directions. A launch needs high bids and a tolerated loss. A profit goal needs bids trimmed to the break-even point. You cannot run both on the same campaign.

Pick one primary objective per product or product group. The common ones are launch, revenue growth, ranking on a keyword set, ACoS reduction, TACoS control and profit. Then pick the one number that tells you if it is working. For a launch, that is organic rank on a short keyword list. For profit, it is profit per unit after ad cost. For a mature product, it is usually TACoS: total ad spend divided by total sales, paid and organic together.

Write both down before the first campaign is built. The ACoS, TACoS and Profit: The Amazon Metrics That Actually Matter guide covers how the three headline metrics relate and which one to lead with.

Step 2: Audit before you spend

An audit is cheaper than a month of misdirected spend. Before changing a bid, we want to know seven things about an account:

1. Wasted spend. In the search term report for the last 60 days, how much spend sits on terms with zero orders? Anything over a fifth of total spend is a structural problem, not a bidding one.
2. Overlap. Are two or more campaigns bidding on the same search term for the same ASIN? Overlap makes you compete against yourself and hides which campaign actually earned the sale.
3. Placement performance. In the placement report, how does top-of-search ACoS compare with rest-of-search and product pages? A large gap means the placement modifiers need attention.
4. Match type balance. How much spend is in broad and auto versus exact? Discovery is necessary; discovery as the majority of spend is not.
5. Conversion rate by ASIN. Ads cannot fix a page that does not convert. If unit session percentage sits below the category norm, listing work comes before bid work.
6. Budget utilization. Are campaigns hitting their daily cap by mid-afternoon, or leaving budget unspent? Both are signals.
7. Attribution and profit. Are sales being read on the right attribution window, and does anyone know the margin per ASIN after ad cost?

The Amazon PPC Audit Checklist: 20 Checks Before We Change a Bid article lists the full set of checks. The point of the audit is a written plan with an expected saving, not a list of observations.

Step 3: Structure campaigns so each one has a job

Campaign structure is where accounts become manageable or unmanageable. The rule we use is simple: every campaign has exactly one purpose, and no two campaigns share a purpose for the same ASIN.

The four purposes are:

• Discovery. Auto campaigns and broad or phrase match, run to find new search terms. Their job is to produce candidates, not to be efficient. Their budget is capped, usually to a minority share of the account.
• Control. Exact match campaigns holding the keywords that have proven they convert. This is where most of the spend and most of the sales should sit. Each keyword gets its own bid and, for the most important terms, its own campaign.
• Defense. Campaigns on your own brand name and your own product pages, so a competitor cannot take the top slot on searches for you. Cheap clicks, high conversion.
• Brand. Sponsored Brands and Sponsored Display running to a Brand Store or a product collection, measured on new-to-brand orders and branded search growth rather than direct ACoS.

Search terms move between these campaigns on rules. A term found in discovery that converts twice at an acceptable ACoS graduates to a control campaign as an exact match and is negated in discovery so the two never compete. That single mechanism is most of what "optimization" means. The Amazon PPC Campaign Structure That Scales With Your Catalog article shows the layout in detail, including portfolios and naming.

Step 4: Set bids by the job the keyword does

There is no correct bid for a keyword. There is a correct bid for a keyword given the job it is doing.

• A ranking keyword, one you are trying to push onto page one organically, is bid above its break-even because organic sales later will pay for the paid loss now.
• A profit keyword is bid at or below its break-even: target ACoS multiplied by the average order value multiplied by the conversion rate gives the most you can pay per click.
• A discovery target is bid low, because you are paying for information, not sales.
• A defense keyword is bid enough to hold the placement and no more.

Change bids in steps of 10 to 15%, on at least two weeks of data, and one variable at a time. Changing bids and placement modifiers in the same week means you cannot tell which change did what. Placement modifiers themselves deserve care: top-of-search usually converts best and costs most, so raise the modifier only where the placement report shows it is earning its premium. The Amazon Bid Optimization: How Often and How Much to Change Bids article covers cadence and step sizes.

Dynamic bidding settings matter too. "Down only" is the safe default for control campaigns. "Up and down" gives Amazon permission to raise your bid on placements it expects to convert, which is useful when you trust the data and dangerous when the campaign is new.

Step 5: Budgets and pacing

Daily budgets do two things: they cap the loss on a bad day and they tell Amazon how much you are willing to spend. A campaign that runs out of budget at noon is telling you either that its bids are too high for its budget, or that it is working and deserves more money. The ACoS tells you which.

Allocate budget by the objective. In a launch, discovery and ranking campaigns get the majority. In a profit phase, control campaigns get the majority and discovery is held to a fixed share. Check pacing weekly: which campaigns are constrained, which are underspending, and whether the split still matches the plan.

Budget rules can raise budgets automatically on high-traffic days and events. They are useful for Prime Day and Q4, and risky as a permanent setting, because they remove the daily cap when it is most needed. The Amazon PPC Budgets: Out-of-Budget Campaigns Without Overspending article goes into detail.

Step 6: The weekly optimization cycle

The weekly review is where management actually happens. Every week, in this order:

1. Pull the search term report for the trailing 30 and 60 days. Sort by spend. Anything with meaningful spend and no orders becomes a negative, exact or phrase depending on how the leak happened.
2. Find the graduates. Terms with two or more orders at or below target ACoS move to exact match in a control campaign and are negated where they came from.
3. Adjust bids on the control keywords that have crossed a threshold: above target for two weeks, bid down; below target with headroom on impression share, bid up.
4. Check placements and adjust modifiers only where the evidence is clear.
5. Check budgets against pacing and the plan.
6. Log every change with the reason. Next week you will need to know what you did and why.

The report from this cycle is what you should expect from anyone managing your ads. Not "we optimized bids", but a list: these terms were negated because of this spend, these were graduated because of these orders, these bids moved by this much. The How to Read the Amazon Search Term Report (and Act on It) article explains how to read the report itself.

Step 7: Choose ad types for the goal, not the menu

Sponsored Products carries most of the weight in most accounts, because the shopper typed a search and your ad appears next to organic results. Sponsored Brands adds the headline placement, video in search results and a landing page on your Store; it needs Brand Registry and works best on terms Sponsored Products has already proven. Sponsored Display reaches shoppers on product pages and off Amazon, and is at its best for retargeting people who viewed but did not buy.

The mistake is to run every format because it exists. A product with a $1,000 monthly budget should be almost entirely Sponsored Products. Brand formats earn their place when the Sponsored Products layer is efficient and there is budget left over. The Sponsored Products vs Sponsored Brands vs Sponsored Display comparison sets out when each one pays.

Step 8: Report on what the business cares about

ACoS is a campaign metric. The business runs on profit. A good monthly PPC report answers four questions: what did we spend, what did it sell, what did it cost per unit after fees, and what changed in organic rank as a result. Weekly reports are about changes and pacing. Monthly reports are about trend and profit per ASIN. Quarterly reports are about strategy: which products to scale, hold, fix or retire.

TACoS falling while sales hold is the sign of a program working. ACoS falling while total sales fall is the sign of one that has simply been cut.

Common mistakes

• Optimizing a listing problem with bids. If conversion rate is below the category norm, lowering bids lowers spend and sales together. Fix the page first. The Amazon ACoS Too High? What to Fix First (and What to Leave Alone) article gives the checking order.
• Letting discovery run the account. Auto and broad campaigns that take more than a quarter of spend for months on end are not discovering; they are leaking.
• Account-wide bid changes. A blanket 20% cut loses the placements that were working and keeps the ones that were not. Bids are a keyword-level decision.
• Pausing campaigns to save money. Paused keywords lose rank within days, and organic sales go with them. Reduce bids in steps instead.
• Judging on too little data. One week of a low-volume keyword tells you almost nothing. Wait for orders, not days.
• No change log. If you cannot say what changed last Tuesday and why, you cannot learn from the result.
• Ignoring stock. Scaling spend into a stock-out is the most expensive mistake in PPC. Rank built with ad money disappears when the listing goes out of stock.

How WeSpark does it

Our PPC engagements start with the seven-part audit, produce a written structure and bid plan before spend changes, and then run the weekly search-term decision engine with every change logged in the report. The three processes below are the backbone.

B11PPC management — start to finishan advertising programme built around one objective, audited before we spend, and optimised every week.

B12How we audit your advertising accountseven audits that show where money is wasted, which placements convert and how much we expect to save — before we change anything.

B15The weekly search-term decision enginea weekly, rule-based clean-up — scale what works, cut what does not — with every change logged in your report.

Bids follow the job of each keyword rather than a single account-wide target, and reports show profit per ASIN, not just ACoS. If you want to see what that looks like on your own account, the free growth plan starts with the audit.