Amazon PPC for beginners

Your product is live on Amazon, the first reviews are in, and Amazon keeps telling you to advertise it. Most Amazon PPC for beginners advice says to start at $20 to $30 a day and expect a high ACoS, so new sellers pay for 50 clicks, get 0 orders and cannot tell if the ad, the listing or the budget was wrong. Here is a small first test instead: one automatic Sponsored Products campaign, a spending limit Amazon cannot go past, and one rule each Sunday to raise it or stop.

Do these three things

About 2 hours to set up, then 20 minutes a day on top of your outside-traffic work, and 30 minutes every Sunday. Judge the test when it has spent its whole limit, not after 50 clicks.

Show the three actions in fullHide the full list

1. 1Today: check your listing has 5 reviews at 3.5 stars or more and shows the Add to Cart box. If it does, work out your break-even ACoS, set a test limit of 10 times what one order leaves you, and lock it with a portfolio budget cap that ends in 14 days.
2. 2Every day: turn off any automatic targeting group, and add as a negative keyword any search term, that has spent 3 times what one order leaves with no order. Keep sending 100 personal messages with your Amazon Attribution link, so ads are never your only shoppers.
3. 3Every Sunday: count ad orders and ACoS. When the limit is spent, run the next test 20% bigger if ACoS stayed at or under break-even, and move search terms with 2 or more orders into a manual exact campaign. If ACoS went over, stop and fix the listing.

Get help doing these

ChatGPT or Claude reads this guide and works through it with you.

Do this with ChatGPT(opens in a new tab)Do this with Claude(opens in a new tab)

1. Check your listing is ready for Amazon ads

Start Amazon ads only when your listing already sells to strangers without them: at least 5 reviews at 3.5 stars or more, the Add to Cart box showing, stock in Amazon's warehouse, and at least 1 order for every 100 sessions. If any answer is no, your Amazon ad budget is $0 for now.

An ad on Amazon puts your listing in front of more shoppers, and you pay for each click. If shoppers who find you for free do not buy, paid shoppers will not either. Amazon Ads says the same: advertise products "that have five or more customer reviews—with a rating of 3.5 stars or higher". A seller on Amazon's forums this year started ads with 0 reviews and got 2,520 impressions, 50 clicks and 0 sales in 10 days. The replies were blunt: "That is not a PPC problem, that is a listing conversion problem."

Check four things in Seller Central:

1. Reviews: 5 or more, at 3.5 stars or more, on the listing.
2. The Add to Cart box (Amazon calls it the Featured Offer) shows on your listing. Amazon only shows ads for products that can win it and are in stock. If it is missing, check that your price is no higher than in your own store, with shipping.
3. Stock: your FBA stock covers at least 4 weeks of orders, so the test does not stop halfway.
4. Orders per 100 sessions: open Reports, then Business Reports, then Detail Page Sales and Traffic by Child Item. Over the last 4 weeks, unit session percentage, Amazon's conversion rate, should be at least 1%. That is the same 1 order per 100 visits from strangers as in how to market a D2C brand. On Amazon almost every session is a stranger.

Also check the money: your ad budget comes from money you can lose without hurting the business, never from stock or rent.

If any answer is no, keep doing the launch routine from how to launch a product on Amazon: 100 messages a day with your Amazon Attribution link, 2 short videos a day on each app and 20 answers where your buyers ask for help. Get reviews only through Vine and Request a Review. The full free routine is in how to get customers online for free. ChatGPT or Claude drafts each message in about a minute; you add one line that is true about the person.

• Tool: Seller Central (your listing and Business Reports).
• Time: 15 minutes.
• You will have: a yes or no for each check, with the real number beside it.
• It worked if: every line has a number from Seller Central, not a guess.
• Common mistake: starting ads in your first week with 0 reviews, then deciding Amazon does not work after 50 clicks and no orders.

2. Work out your break-even ACoS and your test numbers

Your break-even ACoS is what one Amazon order leaves you after every cost except ads, divided by the price. Turn that one number into four more: a test limit, a daily budget, a stop line and a click check.

ACoS (advertising cost of sales) is ad spend divided by the sales the ads brought, as a percentage. Amazon's own guide says ACoS must stay below your profit margin for ads to make money. So your break-even ACoS is your margin before ads. It is the flip side of your break-even ROAS: break-even ROAS = price ÷ what one order leaves. What one order leaves is also the most you can pay to win one buyer, your break-even customer acquisition cost.

Put your product into Amazon's free FBA Revenue Calculator to get the fees. The referral fee for beauty and personal care is 15% on items over $10. Then take away what one unit costs you to make and ship to Amazon. How to price a product has the full sum. Ask ChatGPT or Claude to turn it into a sheet with formulas, and check every number against an invoice or the calculator.

Price on Amazon: $[price]
What one order leaves (price − referral fee − FBA fee − product and shipping to Amazon): $[X]
Break-even ACoS: $[X] ÷ $[price] = [%]
Test limit (X × 10, and never more than money you can lose): $[limit]
Daily budget: $[limit] ÷ 14 = $[a day]. If that is under $10, use $10.
Stop line for one targeting group or search term (X × 3): $[X × 3]
Click check: 100 clicks × $[Amazon's suggested bid] = $[A]. Is the limit bigger? [yes / no]

These are the same rules as in how much to spend on ads: 10 times your number lets about 3 targeting groups each reach the stop line, and 3 times is where one has had a fair chance. The daily budget has a floor of $10 because Amazon recommends starting at "$10 or equivalent" a day; the lowest it allows is $1. At $10 a day, a small limit just runs out a day or two early, and step 3 stops it there.

The click check tells you if the test can buy enough clicks to learn anything. On Amazon you pay per click, and the cost per click is set in an auction. You will see Amazon's suggested bid when you set up the campaign in step 4: write it down here. If the limit buys fewer than 100 clicks, keep ads at $0 and raise what one order leaves first, with a higher price or a bundle that makes a bigger order.

• Tool: the FBA Revenue Calculator (free), Google Sheets, and ChatGPT or Claude to build the sheet.
• Time: 30 minutes.
• You will have: six numbers in one sheet: what one order leaves, break-even ACoS, test limit, daily budget, stop line and click check.
• It worked if: every line holds a real number from the calculator or an invoice, and none says "about".
• Common mistake: using a "good ACoS" from a blog, such as 15 to 30%, instead of your own. A product that leaves 46% of its price can afford a 40% ACoS; one that leaves 20% cannot afford 25%.

3. Lock your test limit with a portfolio budget cap

Put your campaign inside a portfolio with a budget cap equal to your test limit and a date range that ends in 14 days. When the cap is reached, every campaign in the portfolio stops. A daily budget alone cannot do this.

Amazon averages your daily budget over a calendar month, and on any one day it can spend up to 25% more: a $10 budget can cost $12.50 on a busy day. Sponsored Products has no lifetime budget. The portfolio budget cap is the one setting that stops spend at a total. Amazon's help says that when the cap is reached, "all campaigns in the portfolio will stop delivering", and they stay stopped until you raise the cap or extend the dates.

1. 1

   Create the portfolio

   In Seller Central, open Advertising, then Campaign Manager, then Portfolios, and create a new portfolio. Name it for the test, such as test1-oct.

2. 2

   Set the budget cap

   Set a budget cap equal to your test limit, with a date range that starts tomorrow and ends 14 days later. Do not choose a monthly cap: it resets each month and keeps spending.

3. 3

   Check how you pay

   Amazon charges your seller balance or a card once you have $1 of clicks, and your first credit limit is $50. If you created your first listing in the last 90 days, the New Seller Guide matches your first $50, $200 or $1,000 of Sponsored Products spend with ad credits, if you start the campaign within 90 days of listing. The credit shows in Campaign Manager within two weeks. Count your test limit in real money anyway.

• Tool: Amazon Campaign Manager (free; you pay only for clicks).
• Time: 15 minutes.
• You will have: an empty portfolio with a cap equal to your limit and an end date 14 days away.
• It worked if: Portfolios shows your cap and both dates.
• Common mistake: a daily budget with no cap and no end date. It keeps spending every day, and a month at $10 a day is about $300.

4. Launch one automatic Sponsored Products campaign

Launch one Sponsored Products campaign for your best-selling product, with automatic targeting, inside the portfolio from step 3. Amazon chooses the searches and products to show it on, and you learn from what shoppers type.

Amazon Ads tells new advertisers to start with Sponsored Products, and says automatic targeting "is the easiest and quickest way to get started". Automatic targeting has four groups, and each one can be switched on or off during or after setup:

The four groups do the job that 3 ads do on Meta: each reaches different shoppers, so each gets its own stop line in step 5. Guessing keywords yourself comes later, once real searches have shown which words sell.

Campaign: Sponsored Products, Automatic targeting
Name: [product]-auto-[month], for example shampoo-auto-oct
Portfolio: the one from step 3
Daily budget: $[from step 2, at least $10]
Start: tomorrow. End: the portfolio end date
Bidding strategy: Dynamic bids – down only
Placement adjustments: 0% for top of search and product pages
Ad group: one, with one product
Default bid: Amazon's suggested bid (write it in your step 2 sheet)
Targeting groups: all four on

"Down only" lets Amazon lower your bid on searches less likely to sell, and never raise it. Leave placement adjustments at 0%: raising them means paying more per click before you know which clicks sell. Pick the product with the most reviews and the highest unit session percentage, because the ad sends shoppers to its listing, and its photos have to make a stranger trust a new brand. Your listing does the selling, so its title, photos and bullets should already follow the launch guide and a product description that sells.

• Tool: Amazon Campaign Manager.
• Time: 30 minutes.
• You will have: one live automatic campaign with one product, inside the capped portfolio.
• It worked if: the campaign shows as delivering within a day and has impressions the next morning. If it has 0 impressions, check the Add to Cart box and stock first.
• Common mistake: three campaigns for three products at $3 a day each. None gets enough clicks to tell you anything. Put the whole limit on one product.

5. Check targeting groups and search terms every day and turn off what loses money

Every day, at the same time, write down each targeting group's spend and orders. Turn off any group that has passed your stop line with no order, and add any search term that has passed it, or that has nothing to do with your product, as a negative keyword.

The groups. Open the campaign, then the ad group, where the four groups are listed with their switches. Copy each group's spend, clicks and orders into its own row in your sheet. Then:

• Turn off any group that has spent 3 times what one order leaves and has 0 orders.
• Leave the rest alone. Do not change bids, the daily budget or the dates during the 14 days. Amazon's own staff tell sellers to allow 2 to 3 weeks between changes, and Amazon Ads says to check "at least twice a week" in the first week. Turning off losing groups and adding negatives are the only changes you make.

The search terms. Every day, open the search term report: in Seller Central, Reports, then Advertising Reports, ad type Sponsored Products, report type Search term. It lists every search that got a click on your ad. Sort it by spend. Then add, as a negative exact keyword in the campaign:

• any search term that has spent 3 times what one order leaves with no order;
• any search term that is plainly not your product, such as another pet, a different use or "free", on the first click.

Use negative exact, which blocks only that search. Negative phrase blocks every search that contains the words, so a short one can block searches that sell.

Keep the outside traffic going. Ads bring shoppers who are already searching. Your 100 messages a day with tagged links bring shoppers who are not, and every order from them helps your listing in Amazon's search too. Amazon counts ad orders itself, so ads need no conversion tracking setup; your messages need their own links. Keep the routine from the launch guide: people you know first, then small creators, each with their own Amazon Attribution link. The link works like UTM parameters on your own site, and it is how you see which channel brought each order. Ask each creator if you may reuse their video as customer content on your listing. Post your 2 videos a day on Instagram and TikTok, and follow up with every creator who does not answer, twice, 3 days apart. Never ask anyone for a review.

• Tool: Campaign Manager, the search term report and Google Sheets.
• Time: 20 minutes a day, on top of the outside-traffic work.
• You will have: one row per group for each day, and a list of negative keywords.
• It worked if: no group or search term is still running past the stop line with 0 orders.
• Common mistake: judging by clicks or click-through rate. A group with cheap clicks and no orders still loses money. The other mistake is raising bids on day 3 because nothing has sold yet.

6. Every Sunday, count ad orders and ACoS, then raise 20% or stop

Every Sunday, write one row for the test: spend, ad orders, ad sales and ACoS. When the portfolio cap has been reached, decide: run the next test 20% bigger if ACoS stayed at or under break-even, or stop and fix the listing if it went over.

Take spend, orders and sales from Campaign Manager. Then open Business Reports and write down all orders for the same week. On Amazon the ads and the store share one set of orders, so ad orders are real orders. Still, if all orders stayed flat while ad orders rose, the ads may be taking orders you would have had anyway.

Week | Test | Spent | Ad orders | Ad sales | ACoS (spent ÷ ad sales) | Break-even ACoS | All orders (Business Reports) | Unit session % | Reviews and stars | Decision

When the test has spent its whole limit:

• ACoS at or under your break-even: raise the portfolio cap by 20% with a new 14-day date range, and the daily budget by the same 20% (never under $10). Keep raising 20% a test while ACoS stays at or under break-even.
• ACoS over your break-even: stop. Put the hours back into outside traffic. Then find out why shoppers click but do not buy with why ads get clicks but no sales. On Amazon, compare your first photo, price and review count with the top 5 results for your main search, as in a quick competitor analysis, and fix the weakest. Then ask 5 people who looked at it what stopped them. Do not fix it with a deep coupon: it teaches buyers to wait for the next sale.
• No orders at all: check the Add to Cart box and stock first. If both are fine, the listing or the product is the problem: read why no one is buying.

Move the search terms that sell. Amazon Ads' advice for search terms that sell is to "increase the bid on those and add them as exact match type". When the test passes, take every search term with 2 or more orders at or under your break-even ACoS. Put them in a second campaign, Sponsored Products with manual targeting, keyword targeting, exact match, in the same portfolio, at your current bid plus 20%. Add the same terms as negative exact in the automatic campaign, so the two campaigns do not bid against each other. The automatic campaign keeps finding new searches; the manual one pays for the ones you know sell.

Paste the week's numbers into ChatGPT or Claude and ask it to fill the sheet and apply the rules. Check its sums against Campaign Manager.

• Tool: Campaign Manager, Business Reports and Google Sheets.
• Time: 30 minutes every Sunday.
• You will have: one row a week, and a raise or stop decision for every test that has spent its limit.
• It worked if: the decision follows the rule, and every ad order is also in your Business Reports orders.
• Common mistake: raising the budget on day 4 because one day's ACoS looked good. Wait until the limit is spent. The other mistake is moving a search term with 1 order into exact match: 1 order can be luck.

When Amazon ads pay back test after test, use the same number to choose your next channel. For your own store, the same rules run Meta ads for a Shopify store and Google Shopping or Performance Max. Before a second paid channel, list your products on Google for free, and keep testing new ad versions where you run them. Leave retargeting for later: at a few hundred visits a month there are too few people to show it to. Once buyers arrive, work on repeat orders, so a buyer you paid for once comes back free. Your ecommerce marketing budget shows how ads fit next to creators and email. Once a month, step back for a marketing review.

Your Amazon PPC checklist for your first 14-day test

0 of 11 done

Work through this list with an AI assistant

It drafts the work and keeps track of what is done.

Do this with ChatGPT(opens in a new tab)Do this with Claude(opens in a new tab)

Step 1 · Check your listing is ready for Amazon ads

• Check 5 reviews at 3.5 stars or more, the Add to Cart box, 4 weeks of stock and at least 1 order per 100 sessions.
• Every dayWhile your ad budget is $0, send 100 personal messages, post 2 short videos on each app and give 20 answers.

Step 2 · Work out your break-even ACoS and test numbers

• Work out what one order leaves, your break-even ACoS, test limit, daily budget, stop line and click check.

Step 3 · Lock your test limit with a portfolio budget cap

• Create a portfolio with a budget cap equal to your test limit and a 14-day date range.

Step 4 · Launch one automatic Sponsored Products campaign

• Launch one automatic Sponsored Products campaign for your best-selling product inside the portfolio.

Step 5 · Check targeting groups and search terms every day

• Every dayWrite down each targeting group's spend and orders, and turn off any group past the stop line with no order.
• Every dayAdd search terms past the stop line, or plainly not your product, as negative exact keywords.
• Every daySend 100 personal messages with your Amazon Attribution links.

Step 6 · Every Sunday, count ad orders and ACoS, then raise or stop

• Every SundayFill in one row of the Sunday Amazon ads sheet.
• When a test spends its limitRaise the next test by 20% if ACoS stayed at or under break-even, or stop and fix the listing.
• When a test spends its limitMove search terms with 2 or more orders into a manual exact campaign.

Copy listDownloadReset