Back to Resources

Amazon PPC management is the ongoing work of building, monitoring, and optimizing pay-per-click campaigns in Amazon Ads: Sponsored Products, Sponsored Brands, and display ads, plus Amazon DSP for brands running programmatic campaigns. Launching a campaign takes twenty minutes. Management is everything after that: reviewing search terms every week, adjusting bids on conversion data, adding negative keywords, and changing structure when a campaign stops behaving the way it should.

2026 changed the platform underneath that work. Since May, Alexa for Shopping has answered shoppers’ questions in Amazon’s search bar, and existing sponsored ads campaigns are automatically eligible to appear in those conversations. On September 29, Amazon renamed its unified ad platform Amazon Ads Agent, bringing the ads console and DSP into one workspace with AI that can build and optimize campaigns for you.

None of that removes the need for judgment. Amazon’s AI optimizes toward the goal you give it. It doesn’t know your margins, your launch plans, or which keywords bring customers who reorder. This guide covers how the auction works now, the metrics that drive decisions, the weekly cadence we use across our partners’ accounts, and how to tell when it’s time to bring in an Amazon PPC expert.

What Amazon PPC Management Covers

Running campaigns and managing them are different activities. Running means the campaigns are live and spending. Managing means someone reviews the search term report every week, moves converting terms from auto to exact match, cuts keywords that drain budget, and catches the moment a campaign’s numbers start drifting.

Most brands underinvest in that second layer, and by the time a problem shows up, it has usually cost real money. For any brand using advertising to grow on Amazon, which at scale means nearly every brand, PPC management is a permanent operating discipline rather than a launch task.

What Changed for Amazon PPC in 2026

The auction mechanics are the same on the surface. Every search triggers a real-time auction, and your ad wins placement based on your bid and how relevant and well-performing your listing is. Three things around that auction changed this year.

Relevance leans harder on the listing. Keyword relevance still decides whether you’re eligible. After that, click-through and conversion rate feed back into how often your ads show. This is why many brands see CPCs rise without proportional sales growth: the system exposes listing weaknesses that aggressive bidding used to cover. A weak main image or thin bullets now cost you at every bid level.

Ads moved into the conversation. Amazon replaced Rufus with Alexa for Shopping in May 2026. Amazon Ads says existing sponsored ads campaigns are automatically eligible to serve in Alexa for Shopping without additional setup, and that Sponsored Products and Sponsored Brands campaigns can also run as “AI-powered conversational ads that appear in shopping results and product pages.” In June, Amazon added Alexa+ Agentic Ads, which it calls “the first ad format that takes a customer from seeing an ad to completing a purchase entirely within the conversation,” launching first on Echo Show devices. The assistant can only recommend what it can explain, so listing quality now affects ad performance in two places. Our complete guide to Amazon AI optimization covers what the assistant reads.

The console became Amazon Ads Agent. On September 29, 2026, Amazon announced Amazon Ads Agent as “the new name of the unified platform,” combining the ads console and DSP. Advertisers can build campaigns manually, let Amazon’s AI optimize, or mix the two. Sponsored Display is now part of Amazon’s broader display ads offering inside that platform, and existing Sponsored Display campaigns keep running. Amazon also made Full-Funnel Campaigns available to all U.S. advertisers, and its new DVA+ format, which brings display, video, and audio into one buying experience, begins rolling out in late October.

The Metrics That Drive Real PPC Decisions

The ad console produces more data than most brands can use. A small set of metrics drives the decisions that matter.

ACoS and TACoS

ACoS (Advertising Cost of Sale) measures ad spend against ad-attributed sales. Before you judge any ACoS, you need the product’s breakeven ACoS: the point where ad spend uses up all the margin left after cost of goods and Amazon fees. Every judgment about whether an ACoS is acceptable starts there, and category benchmarks or competitor guesses can’t replace it.

TACoS (Total Advertising Cost of Sale) measures total ad spend against total sales, organic included. It’s the clearest test of whether advertising is building the business or replacing sales you used to get for free. A rising ACoS with a falling TACoS is often healthy: ads are building ranking, organic sales are growing, and the account depends less on paid traffic. Rising TACoS with flat revenue means the opposite. Our full breakdown of ACoS and TACoS covers how to read them together at each product stage, and why a lower ACoS isn’t always the goal.

Click-through rate

CTR measures how often shoppers click your ad when they see it, which makes it a test of how well the ad and listing fit the search. A keyword with plenty of impressions and weak CTR usually points to the main image, title, price, or review count. Raising the bid in that situation buys more impressions of an ad shoppers are already passing over.

Conversion rate

Conversion rate measures how often a click becomes a sale. When it’s low, the reflex is to tighten targeting, but weak images, an unclear title, or missing A+ Content can’t be fixed in the ad console. Conversion rate also sets what you can afford per click: at the same ACoS target, a product converting at 15% can pay three times the CPC of one converting at 5%. Improving the listing is often the fastest route to a better ACoS, and our complete guide to Amazon listing optimization covers each field.

Search Query Performance

Search Query Performance, in Brand Analytics for Brand Registry sellers, shows your share of impressions, clicks, cart adds, and purchases for each search query, paid and organic together. Low impressions point to a relevance problem. Strong impressions with weak clicks point to image, title, price, or reviews. Healthy clicks with weak purchases point to the detail page. Our Brand Analytics guide walks through the reports, and knowing how to read Amazon’s data is most of the advantage.

Campaign Structure: Build Something You Can Optimize

Campaigns that mix match types, unrelated products, or auto and manual targeting produce data you can’t act on. Good structure makes performance readable at the level where you make decisions.

Separate auto and manual campaigns. Auto campaigns let Amazon match your ads to searches based on your listing and shopper behavior, which makes them a research tool for finding queries you haven’t thought of. Manual campaigns are where proven terms get precise bids. Mixing them blurs the data and forces one bid onto terms with very different economics.

Separate match types. Broad, phrase, and exact match each have different costs and conversion rates. Keep them in separate campaigns or ad groups so you can see which one is doing the work. When you’re starting out, 10 to 20 tightly themed keywords per campaign keeps budgets controllable and results readable.

Organize by product role. Hero products driving category rank, growth products building sales history, and defensive products holding position each deserve a different share of budget and a different ACoS target. Splitting budget evenly across the catalog funds the wrong products.

Give new campaigns room. New campaigns need two to four weeks of data before optimization means much. Cutting bids hard in week one produces noisy signals and decisions you’ll reverse later.

New products need their own setup: higher acceptable ACoS, broader keyword sets, and heavier top-of-search investment while they build sales history. The handoff from launch targets to efficiency targets is one of the most consequential calls in a product’s life, and our Amazon product launch playbook covers the three-layer launch setup and how it changes over the first 90 days.

Keyword Research for Amazon PPC

Paid keyword research has a different goal than organic research. You’re looking for searches where shoppers at the right intent level will find your product and buy it. Volume without purchase intent is expensive traffic.

Your own account is the best source. Auto and broad match campaigns surface the searches real buyers used, and mining their search term reports is the highest-signal research you can do, because the data comes from people who bought or didn’t. Competitor research fills gaps in your coverage, and Brand Analytics, Search Query Performance, and Product Opportunity Explorer show which keyword clusters carry enough demand to be worth pursuing.

Treat Amazon’s AI keyword suggestions as research inputs: they surface long-tail variations you’d miss, but they don’t know your margins or priorities.

Tier the list into primary terms that describe the product directly, secondary terms for attributes and use cases, and long-tail phrases with higher intent. Our 10-step keyword research process covers how we build and rank the list, and the same research feeds your listing and backend keyword strategy.

Keywords that convert in paid campaigns build the sales velocity that lifts organic rank for the same terms. That’s the core of the PPC and organic flywheel, which our post on SEO and PPC integration covers in detail.

Bidding Strategy: Match the Strategy to the Campaign

Amazon offers three bidding strategies, and the right one depends on what each campaign is for.

Fixed bids apply the same bid to every auction. They give you full control and suit defensive branded campaigns or tests where you want predictable spend.

Dynamic bids (down only) let Amazon lower your bid when a sale looks less likely. Amazon suggests this strategy when you’re optimizing for return on ad spend, and it limits downside on campaigns still gathering data.

Dynamic bids (up and down) let Amazon raise or lower bids based on the likelihood of a sale. Amazon recommends it for maximizing performance, but it needs real conversion history. On new campaigns or thin catalogs it can spend fast on traffic that doesn’t convert. A practical sequence: run fixed or down-only bids for the first three to four weeks to learn the product’s natural conversion rate and true breakeven, then test up-and-down with guardrails.

Placement adjustments sit on top of the strategy and raise bids by a percentage for specific placements, such as top of search or product pages. Check your placement report first. If top of search converts at two or three times the rate of product pages for a campaign, the adjustment should reflect that. Our Amazon PPC bid strategy guide covers when to move between strategies as campaigns mature.

Negative Keywords: The Most Undervalued Task in Amazon Advertising

Negative keywords stop your ads from showing on searches that won’t convert, which cuts wasted spend without cutting coverage on terms that sell.

There are two kinds. Irrelevant-query negatives block searches that have nothing to do with your product: a brand selling premium kitchen knives doesn’t need to show for “plastic knife set.” Cross-campaign negatives block a term you already run profitably in exact match from your broad and auto campaigns, so your own campaigns don’t bid against each other and muddy attribution.

This is weekly work. Broad and auto campaigns keep opening new query territory, so the search term report surfaces new waste every week. Add any query with a meaningful number of clicks and no sales. If you’re adding dozens of negatives every week on a mature account, the match-type strategy is too loose, and that’s the thing to fix. Our guide to building a negative keyword strategy covers how to structure both kinds across a full account.

Sponsored Products, Sponsored Brands, and Display Ads

The three self-service formats do different jobs, and optimizing them with the same logic shortchanges all three.

Sponsored Products promote individual listings in search results and on product pages. They’re the highest-intent format and the biggest source of attributed sales for most brands. They also generate the search term data that informs everything else: manual targeting, Sponsored Brands keywords, and listing copy. To run them, sellers need a Professional account and listings eligible for the Featured Offer.

Sponsored Brands show your logo, a custom headline, and several products at the top of search results, with video options in search as well. They build awareness and defend your best terms, and the brand exposure they create when a shopper sees the ad but doesn’t click is value most attribution misses. Sellers need Brand Registry to run them. Paired with Sponsored Products on your most valuable terms, they create a double presence on the page that’s hard for competitors to displace. Our comparison of Sponsored Products vs. Sponsored Brands covers when each earns its budget.

Display ads (formerly Sponsored Display) target audiences and product pages rather than search queries. The most dependable use is remarketing to shoppers who viewed your product and didn’t buy, one of the highest-intent audiences on Amazon. Sponsored Display now sits inside Amazon’s broader display ads offering in Amazon Ads Agent, and existing campaigns continue to run there.

Run together, the three formats cover each other’s gaps. Our guide to Amazon advertising budget allocation covers how to split spend across them.

The Weekly PPC Optimization Cadence

Accounts that improve reliably are the ones where someone sits down with the data every week and makes small decisions. The warning signs almost always show up in the data before they show up in revenue.

Search term report. Pull the last seven days and sort terms into three groups. Converting terms not yet in a manual campaign move to exact match with a bid set to your target ACoS. Terms spending consistently without converting become negatives. High-ACoS terms that convert need judgment: worth running at a loss during a launch, a drain on a mature product.

Bids. Lower bids on keywords running well above target ACoS and raise them on keywords converting with room to scale. Move in steps of 10 to 20 percent. Large swings disrupt campaigns that were building useful history.

Budgets. Check which campaigns hit their daily cap and when. Raise budgets on efficient campaigns that run out before the end of the day.

Placements. Compare top of search, rest of search, and product page performance on your larger campaigns and adjust placement percentages to match.

Sponsored Brands and display ads. Review them separately, against their own goals: new-to-brand orders and branded search lift for Sponsored Brands, remarketing efficiency for display.

AI and Automation: What to Hand Off and What to Keep

Amazon’s AI now runs through every layer of the platform: dynamic bidding, budget recommendations, AI keyword suggestions, and with Amazon Ads Agent, campaigns it can build and optimize largely on its own. The question is which decisions to give it.

Where AI does well. Bid automation makes thousands of adjustments a day on conversion patterns, seasonality, and live competition, and budget pacing shifts spend toward what’s working. For brands with clear time-of-day patterns, AI dayparting adjusts bids at a frequency manual work can’t match.

Where AI falls short. It optimizes toward the metric you give it, usually sales or return on ad spend, and it can’t weigh what it doesn’t see. AI doesn’t know you’re launching a product and willing to run a high ACoS to build rank. It can’t tell that some keywords bring customers with higher lifetime value, or that two products have very different margins. It makes mathematically sound calls on limited inputs.

What works. People set the structure, match types, negative keyword strategy, and ACoS targets around business goals, and the AI optimizes bids inside those guardrails. Override it during launches, promotions, and ranking pushes. Review what it did every week and adjust the constraints. Our advertising automation framework lays out which tasks to automate.

Amazon DSP: What It Is and When It Makes Sense

Amazon DSP buys programmatic placements across Amazon’s own properties, partner sites, and streaming environments, reaching shoppers beyond the search moment: category browsers, competitor viewers, and past buyers. It now lives in the same Amazon Ads Agent workspace as sponsored ads, but it’s a different tool, with different targeting, measurement, and goals.

DSP makes sense once your sponsored ads are optimized, your listings convert well, and your budget can support enough spend to produce reliable data. Brands that start DSP before that foundation is solid usually pay for top-of-funnel traffic their listings can’t convert. Our guide to Amazon DSP readiness for mid-market brands covers the questions to answer first.

PPC Across Walmart and TikTok Shop

The core disciplines (keyword research, match-type structure, search term reviews, negatives, and bid management) carry over to Walmart Connect and TikTok Shop ads. The auctions, traffic volume, and creative formats differ enough that each platform needs its own cadence, and brands on several marketplaces do better coordinating keyword strategy and budget across them. Our Walmart vs. Amazon PPC strategy guide covers the differences that matter.

What an Amazon PPC Expert Does, and How to Choose One

An experienced Amazon PPC expert or agency does the same work this guide describes, with more data and more hours behind it. That means keyword research grounded in what converts, structure built for clean decisions, bids set from margins rather than benchmarks, and weekly optimization that never gets skipped because something else came up. The best teams back their PPC managers with data analysts and engineers who build tools for the repetitive work, so the people on your account spend their time on strategy, creative testing, and the business questions AI can’t answer.

Signs it’s time to bring in help

In-house management works with a simpler catalog and someone who can protect four to six hours a week for optimization. It breaks down when PPC is the first thing dropped in a busy week, when the catalog outgrows the hours, or when you’re moving into display, DSP, or other marketplaces. The data usually shows it first: rising TACoS, widening gaps between strong and weak campaigns, and search term reports nobody has opened in a month. If you’re unsure whether Amazon PPC is worth the investment at your stage, start there.

What to ask an agency or consultant

• Who runs the account day to day? Senior strategists who sell the deal and junior staff who do the work is a common gap.
• How often are campaigns optimized? Weekly is the standard to hold them to.
• How do they use AI? An agency that leans entirely on Amazon’s automation or a third-party tool without strategic oversight will optimize your account toward the platform’s goals instead of yours.
• What does reporting tie to? Look for TACoS, organic rank, and margin, rather than clicks and ACoS alone.
• Can they show proof? Ask for case studies with specific numbers, and read independent reviews such as Canopy’s on Trustpilot.

Our list of questions to ask an Amazon PPC consultant goes deeper, and our comparison of the best Amazon PPC agencies helps with the shortlist. If PPC is only one of several functions you’re short on, our guide to Amazon account management covers the full scope.

How Canopy Approaches Amazon PPC Management

We run Amazon PPC as part of the whole account. The same dedicated brand manager owns your account for the life of the engagement, so advertising decisions are made alongside listings, inventory, and organic rank instead of in a separate silo. We use AI to execute and people to set strategy, and our reporting ties to TACoS, organic rank, and margin. That approach has helped our partners generate more than $3.3 billion in revenue, with an average 84% year-over-year profit increase and 99.1% partner retention.

Our Amazon PPC management service page covers how the work is set up.

Frequently Asked Questions