Amazon PPC for Beginners: How the Auction Actually Works
Amazon PPC is a pay-per-click auction: you bid on keywords, your product shows in search results or on product pages, and you pay only on click. Beginners should launch one automatic campaign, read the search term report after two weeks, then split winners into manual campaigns.
What this looks like across the book we manage
What Amazon PPC Is, With Real Numbers
Amazon PPC is the ad system built into Seller Central and Vendor Central. You bid on a keyword. If you win the auction, your product shows as a sponsored listing in search results or on a competitor's product page. You pay only when someone clicks — not for the impression, and win or lose, you never pay more than one cent above the next highest bid.
The metric that decides whether any of this is worth it is ACoS: ad spend divided by ad revenue. Here's a worked example most guides skip. Say you bid $0.75 a click on a campaign, it runs one week, and gets 120 clicks. That's $90 spent. Of those 120 clicks, 3 people buy, at $25 each — $75 in ad revenue. ACoS is $90 ÷ $75 = 120%.
That number alone doesn't tell you if you're losing money — you need your margin. If that $25 product leaves $9 of profit after cost of goods, FBA fees, and the referral fee, your break-even ACoS is $9 ÷ $25 = 36%. At 120% ACoS, you spent $90 to earn back $27 in actual margin (3 units × $9). That week cost you $63, not $90 — but it still cost you. This is the calculation to run before you decide a campaign is working, not impressions or clicks alone.
The Ad Types You'll Actually Use First
There are more ad types than a beginner needs on day one. Three matter early:
- Sponsored Products — shows in search results and on other product pages. No brand registry required. This is where almost every beginner starts, and where the worked example above applies directly.
- Sponsored Brands — requires Brand Registry. Shows your logo and multiple products near the top of search results. Better for awareness once you have more than one SKU worth cross-promoting.
- Sponsored Display — targets shoppers by audience or by competing product, and can follow them off Amazon. Useful for remarketing once you have traffic history, not a starting point.
Inside Sponsored Products you'll choose automatic or manual targeting. Automatic hands keyword selection to Amazon's algorithm and is the right choice for a brand-new listing with no data. Manual gives you control over bids per keyword, but only pays off once you know which keywords convert — which is the entire point of the next section.
Your First 30 Days: Automatic, Then Manual
The sequence that works is the same one every experienced PPC manager runs, and it's simpler than most tutorials make it sound. Launch automatic, read what it tells you, then take control.
The Mistake That Quietly Kills Beginner Accounts
It isn't bad targeting. It's stopping the loop above after month one or two — leaving an automatic campaign running on autopilot for a year with the same bids it launched with, because nobody went back to read the search term report a second time.
Across the more than $500 million in Amazon ad spend Full Circle has managed across 100+ brands, this is the single most common failure pattern we see when we take over an account: not one catastrophic error, just a harvest loop that ran once and never ran again. We've seen it happen inside accounts we manage too, before it got caught — this isn't a mistake unique to beginners.
The discipline that fixes it doesn't require software. It requires writing down, before you change a bid or add a keyword, three things: what evidence made you do it, how you'll measure whether it worked, and at what point you undo it if it didn't. That's the same structure we hold every proposed change to — every proposed change carries three things before it runs: the evidence behind it, a measurement plan, and a rollback trigger. You can run that discipline manually with a spreadsheet. Most beginners just never write it down, so nothing gets reverted and nothing gets repeated.
When Your Numbers Look Bad
Bad numbers are normal in the first few weeks. What matters is knowing which fix to try first.
- ACoS spikes suddenly: check whether a competitor entered the same keyword, your price changed, or your organic rank dropped — PPC often absorbs the blame for problems that started elsewhere on the listing.
- Impressions but no clicks: your bid is probably too low for the placement you're chasing, or your main image and price aren't competitive against what's beating you in that search result.
- Clicks but no sales: check the actual search term report before touching the bid. A broad match keyword is often pulling in traffic that was never going to convert — that's a targeting problem, not a bid problem.
- Attribution looks off: Amazon's sales reporting lags behind the click, so don't judge a change on same-day numbers. Give any single change at least a week before you decide it failed.
If a fix doesn't move the number in a week, revert it before you try a second fix. Stacking changes on top of an unproven change is how beginner accounts turn into ones nobody can explain anymore.
Where Dr. PPC Fits
Everything above is the actual mechanics — the auction, the loop, the discipline of measuring a change before trusting it. You can run all of it by hand. Dr. PPC exists for sellers who'd rather have that loop run against their real account economics without doing it themselves: it reads the account, proposes changes with the evidence and rollback trigger behind them, and lets you choose how much runs automatically. It's $300 a month plus 3% of ad spend, capped, month-to-month, with the first 30 days free, and the Orbit software suite — the trackers most of this list charges separately for — is included. It's built and operated by Full Circle. Whether or not you ever use it, the loop is the same one to run yourself.
| Stage | Typical Duration | What It Tells You | What To Do Next |
|---|---|---|---|
| Launch automatic campaign | 1–2 weeks | Which search terms Amazon's algorithm finds for your listing | Let it run without touching bids — you need data, not adjustments |
| Pull the search term report | Day 10–14 | Which terms actually converted vs. which just spent money | Sort by ACoS and orders per term |
| Harvest converters | Ongoing, weekly | Which keywords deserve their own controlled bid | Move them into a manual exact-match campaign |
| Negate and repeat | Ongoing, weekly | Which terms spent without converting | Add as negative in both campaigns, then re-check the report next week |
Which one you should actually pick
Beginners who want to actually understand PPC before spending money should run the automatic-then-manual loop by hand for at least one product — it's the fastest way to see the mechanics. Once the account is real, has history, and the harvest loop becomes a job nobody has time to keep running weekly, that's where a managed option like Dr. PPC or a traditional agency earns its cost. Tools like Ad Badger's automation suit sellers who want to keep the manual control but skip the repetitive steps.
Shortlist on the job, not the feature grid. Pull your search-term report for the last 90 days and total the spend against terms that produced no orders — 48.5% across the 47 brands above. Then ask each vendor on your list what they would do about it in week one, and see who answers with a process rather than a screenshot.
Common questions
How much budget does a beginner need to start Amazon PPC?
There's no universal minimum — it depends on your CPC range and how many keywords you're testing. Start small enough that a bad week doesn't hurt, but large enough to generate the 100+ clicks you need before a search term report tells you anything reliable. A budget too small just delays the data you need to make the first real decision.
What's a good ACoS for a beginner?
There's no fixed good number — it's whatever's below your break-even ACoS, which is your product margin as a percentage of price. Calculate that first. A 40% ACoS can be profitable on one product and a loss on another, depending on what's left after cost of goods and Amazon's fees.
Should I run automatic or manual campaigns first?
Automatic first, always, for a brand-new listing. It has no keyword history to draw on, so let Amazon's algorithm find converting terms for one to two weeks, then move the winners into manual campaigns where you control the bid directly.
Why do my Amazon PPC sales numbers keep changing after I check them?
Attribution isn't instant — a click today can show as a sale several days later, and Amazon's reports update retroactively. Don't judge a campaign change on same-day or next-day numbers. Give it a full week minimum before deciding it worked or didn't.
What's the single biggest mistake beginners make in Amazon PPC?
Running the harvest loop once at launch and never running it again. An automatic campaign left untouched for months keeps paying the same bids on the same mix of good and bad search terms indefinitely — the fix isn't more spend, it's going back to the report.
Dr. PPC runs your Amazon ads daily — an AI agent doing the work, operators from a $500M+ Amazon team supervising. $300/mo + 3% of ad spend, published and capped, month-to-month. Orbit is included.
Start free for 30 daysRead next
- Amazon PPC Software Comparison: Who Publishes a PriceComparison · amazon ppc software comparison
- Prestozon Pricing: The Product Stopped Selling in 2021Pricing · prestozon pricing
- Jungle Scout Pricing 2026: Both Tabs, Seats and the MathPricing · jungle scout pricing
- Helium 10 Pricing 2026: Every Tier and the 2% Ads FeePricing · helium 10 pricing
- Perpetua Pricing: The Rate Card and the Missing PercentPricing · perpetua pricing
- Pacvue Pricing 2026: No Public Rate, and How to Bracket ItPricing · pacvue pricing
Part of
- Every Amazon PPC tool we have comparedIndex of the comparison set
- Dr. PPC’s libraryEvery guide, benchmark and answer in one place
- Agency vs software vs AI-managedThe decision underneath all of these