Amazon PPC Strategy for Beginners: Your First 90 Days

ByGulfam Haidar

A good Amazon PPC strategy for beginners starts with two campaigns per product: one auto and one manual. Run both for two weeks on a set daily budget. Then move search terms that sell into exact match. Block terms that spend but do not sell, using a click limit based on your own conversion rate. Amazon lets you start from $1 a day, but its own guide suggests $10.

Key facts

  • Minimum daily budget: $1 for Sponsored Products (Amazon Ads FAQ). Amazon’s suggested start: $10 a day.
  • Amazon has three bidding strategies: down only, up and down, and fixed bids.
  • Placement adjustments go up to 900%. With up and down bidding, a $1 bid can reach $20 at top of search.
  • There are only two negative match types: negative exact and negative phrase. There is no negative broad.
  • Since 25 March 2026, US Sponsored Products ads can serve in AI prompts. Trade press reports creator placements from 10 August 2026.
  • Amazon publishes no official average cost per click. Use the suggested bid in your own console.

This page is the plan itself. If you need the basics first (what PPC is, what ACoS means, what clicks cost), read our Amazon PPC guide and come back.

Before you spend: is your listing ready for ads?

Ads send shoppers to your listing, so a weak listing wastes every click and you should fix the page before you pay for traffic.

We run this check on every product before its first campaign:

  • Images: at least 7, including one that shows size and one that shows use.
  • Title: 75 characters or less. Amazon made this the rule for most categories from 27 July 2026.
  • Price: in line with the products that show up next to yours.
  • Reviews: a few real reviews. Zero reviews makes cold traffic hard to convert.
  • Stock: enough for 60 days or more. Running out stops your ads and hurts rank.

If two or more of these fail, fix the listing first. Our listing creation service covers the fixes, and a product launch plan covers brand new products.

Your Amazon PPC strategy, week by week

Once the listing passes, follow this schedule, and use the steps below to learn how to do each task.

Week What you do What you do not do
1 Launch one auto and one manual campaign per product. Down only bidding. Set budget with the formula below. Change bids.
2 Check daily that budget is not running out before evening. Raise budget if it is. Pause anything.
3 Pull the search term report. Move terms with 2+ orders into an exact campaign. Cross-negate them. Negate terms that are under your click limit.
4 Add negatives for terms past your click limit with 0 orders. Add obvious bad modifiers. Raise any bid more than 20%.
5 Read the placement report for the first time. Note which placement converts best. Add a big placement boost yet.
6 Harvest again. Add a first placement adjustment (for example 25% to 50%) where it converts. Launch new ad types.
7 to 8 Raise bids 10% to 20% on exact terms that sell under your target ACoS. Cut bids on the ones above it. Judge by one week of data.
9 If you are in Brand Registry, launch a Sponsored Brands campaign on your best exact terms. Spread budget across new products.
10 Check prompts and creator placements in the placement data. Pause or limit them if they spend without sales. Leave new surfaces unchecked.
11 to 12 Try Sponsored Products video or up and down bidding on your best campaign only. Test more than one change per campaign.
13 Review the full 90 days. Set your target ACoS and a monthly routine. Start over from scratch.

Step 1: Launch two campaigns per product

Your first move is two Sponsored Products campaigns for each product: one auto and one manual. The auto campaign finds search terms you did not think of, while the manual campaign tests the keywords you already believe in.

Auto campaign. Amazon picks where to show the ad, based on your listing. It uses four targeting groups, per Amazon’s targeting guide:

  • Close match: searches closely related to your product.
  • Loose match: searches loosely related to it.
  • Substitutes: product pages of items shoppers might buy instead.
  • Complements: product pages of items shoppers buy alongside yours.

Manual campaign. Add 10 to 20 keywords from your research, in broad and phrase match. Broad match catches the most variation, and phrase match keeps the words in order. Our keyword research guide shows how to build that first list.

Put one product in each ad group. Amazon allows up to 1,000, but mixing products hides which one earns the sale. With one product per group, every number in the report means one thing.

Step 2: Choose your bidding strategy

Start with dynamic bids, down only. It protects a new account while you learn which terms sell. Amazon offers three bidding strategies, and its Sponsored Products guide describes them this way:

  • Dynamic bids, down only: Amazon will “reduce your bid if your ad is less relevant… or is on a placement that doesn’t perform well.”
  • Dynamic bids, up and down: Amazon says your bid can be “increased by up to 100%” at the top of the first page. For all other placements, the limit is “up to 50%.”
  • Fixed bids: this “uses your exact bid for all opportunities.”

Down only never pays more than your bid, which makes it the safe choice while your conversion data is thin. Switch a campaign to up and down only after it has sold at your target ACoS for a month.

Step 3: Set a budget you can learn with

Your budget needs to buy enough clicks to make a decision. Here is the formula we use:

Daily budget = (keywords × clicks needed per keyword ÷ days) × your suggested bid

Amazon publishes no average cost per click, so take the suggested bid shown in your ad console for your own keywords.

Worked example. Say your manual campaign has 10 keywords. You expect a 10% conversion rate, so you want about 20 clicks per keyword to judge it (more on that below). You give it 14 days. Your console suggests $0.90 a bid.

  • Clicks needed: 10 × 20 = 200 clicks.
  • Clicks per day: 200 ÷ 14 = about 14.
  • Daily budget: 14 × $0.90 = about $12.60.

Give the auto campaign a similar amount, and your test costs about $25 a day, or $350 over two weeks. If that is too much, cut the keyword list to 5. Do not cut the clicks per keyword, because then you learn nothing.

Amazon’s minimum is $1 a day, and its best practices guide suggests starting at $10. A $1 budget buys about one click a day at most bids, which is too little data to act on.

Step 4: Harvest winning search terms (weeks 2 to 4)

After two weeks, open the search term report. It lists the real words shoppers typed before they clicked your ad, and it is where your plan gets sharper.

Sort the report by orders. Any term with 2 or more orders is a winner, so add it as an exact match keyword in a separate exact campaign. Exact match shows your ad only for that search and close variants of it.

Then do the step most guides skip: cross-negation. Add the same term as a negative exact in the campaign it came from. Now the auto or broad campaign stops bidding against your exact campaign for the same shopper. Your exact campaign gets the traffic, at the bid you chose.

Without cross-negation, your own campaigns compete for the same search and split the data in two.

Step 5: Add negative keywords

Negatives stop you paying for clicks that will never turn into sales, and they are the fastest savings in any new account. Add them in weeks 3 and 4, then every two weeks after that.

Negative exact vs negative phrase

Amazon has two negative match types: negative exact and negative phrase. There is no negative broad, per Amazon’s targeting guide.

Negative type What it blocks Example: you add “dog bed” Use it for
Negative exact That exact search and close variants Blocks “dog bed”. Still shows for “large dog bed”. One bad search term, and cross-negation
Negative phrase Any search that contains the phrase in that order Blocks “dog bed”, “large dog bed” and “dog bed cover” Words that are always wrong for you, like “free” or “used”

Negative phrase reaches wide, so use it with care. If you sell dog beds and negate the phrase “dog bed”, you block your own best traffic. Use negative phrase for modifiers, not for your core product words.

You can add negatives to auto campaigns too. Many sellers miss this, and their auto campaigns keep paying for the same bad terms week after week.

When to negate: use your own conversion rate

Most guides say “negate after 15 clicks,” but that rule ignores your product. A term with 15 clicks and no sale is normal at a 5% conversion rate. At a 20% rate, it is a clear loser.

Use this rule instead: the clicks you would expect per sale equal 1 ÷ your conversion rate. Negate a term when it reaches 2 times that number with zero orders.

Your conversion rate Clicks you expect per sale (1 ÷ CVR) Negate at 0 orders after (2 × that)
3% 33 67
5% 20 40
10% 10 20
15% 6.7 14 (13.3 rounded up)
20% 5 10

Find your conversion rate in the campaign manager: orders ÷ clicks, across the whole product. Use at least 100 clicks of data before you trust it.

Worked example. Your product converts at 10%. A term has 22 clicks at $0.90 and no orders. That is $19.80 spent, and it passed the 20-click limit. Add it as a negative exact. If it had 12 clicks, you would wait.

Negative product targets and brand exclusions

Auto campaigns also show your ad on other product pages through the substitutes and complements groups. Some of those pages never convert, so add those ASINs (Amazon’s product ID numbers) as negative product targets.

Use brand exclusions when a whole brand’s pages drain spend, such as a premium brand whose buyers will not switch.

Where to find negative keywords

The search term report is your main source. Sort by spend, then filter for zero orders. Look for these patterns:

  • Wrong buyer words: “free”, “used”, “cheap”, “diy”, or “kids” when you sell adult sizes.
  • Wrong specs: sizes, colors or materials you do not sell.
  • Competitor brand names that spend without sales.
  • Wrong product type: “dog bed” traffic on a cat bed listing.

Add obvious bad modifiers as negative phrase on day one, since you need no data to know “used” is wrong for a new product.

Step 6: Read your placement report

The placement report shows where your ads ran and what each place earned. Check it from week 5, when it splits results into three placements: top of search, rest of search, and product pages.

You can add a percentage boost to each placement. Amazon allows up to 900%, per its best practices guide. Amazon added rest of search adjustments on 9 January 2024.

Beginners should go slowly here, because adjustments stack with dynamic bidding.

Worked example. Your keyword bid is $1.00. You set a 900% top of search adjustment. That takes the bid to $1.00 × (1 + 9) = $10.00. Now switch to up and down bidding, which lets Amazon raise it by up to 100% at top of search. That is $10.00 × 2 = $20.00 for one click.

A $1 bid you thought was safe can cost $20, so start with down only and small steps. Add 25% to 50% on the placement that converts best, then check again in two weeks.

Days 30 to 90: scale what works

By day 30 you have winners in exact match and fewer wasted clicks, so now you grow one change at a time.

  • Raise bids on proven exact terms. Go up 10% to 20% on terms that sell under your target ACoS. Wait a full week between bid changes on the same term.
  • Cut bids on terms above target. Lower them 10% to 20% before you pause them.
  • Launch Sponsored Brands once you are in Brand Registry. Use your best exact terms, not new ones.
  • Test video. Amazon announced Sponsored Products video at unBoxed 2025. Test it on one proven product.

Judge every step by profit, not ACoS alone. ACoS (ad spend ÷ ad sales) tells you ad cost, not whether the product makes money. Our PPC optimization guide covers the monthly routine in detail. In one account, we cut ACoS from 35% to 10% in six months, and you can read how in our ACoS case study.

2026 changes every beginner should check

New campaigns in 2026 can show in places older guides never mention, so check both of these in your first 90 days.

Prompts. Sponsored Products and Sponsored Brands prompts reached general availability in the US on 25 March 2026. They are AI questions shown on search results and product pages. They can open Amazon’s shopping assistant, which Amazon renamed from Rufus to Alexa for Shopping in May 2026. Prompts are billed within your normal cost per click, and you can pause them in the console.

Creator placements. Trade press reports that from 10 August 2026, Sponsored Products ads also run in Amazon Influencer content. According to PPC Land, which cites an Amazon help update of 4 August 2026, campaigns are auto-enrolled. The reported controls are “Increase reach” and “Limit off-Amazon spend”, plus exclusion of single creators.

Our advice: check both in week 10. If a surface spends without orders, pause prompts or choose “Limit off-Amazon spend”. Amazon has not yet published full reporting guides for creator placements, so confirm the setting in your own console.

Beginner mistakes, and what they cost

Most wasted spend in a new account comes from a short list of errors, shown here with the fix for each.

Mistake What happens Fix
Running auto only You never control bids on your best terms Move 2+ order terms to exact by week 3
Many products in one ad group You cannot tell which product earns the sale One product per ad group
Up and down bidding from day one Bids rise before you know what converts Down only for the first month
Pausing after 3 days Too few clicks to judge anything Wait 14 days or your click limit
No negatives Spend leaks to “free”, “used” and wrong sizes Add obvious negatives on day one, data-based ones in week 4
No cross-negation Your own campaigns bid against each other Negative exact in the source campaign
Big placement boosts early A $1 bid can cost $20 Start at 25% to 50%
Judging by ACoS alone You cut ads that grow total sales Check profit and total sales too

Where this advice could be wrong

This plan fits a typical new product with steady demand, and it fits less well in the cases below.

Launches. A brand new product may need higher bids to earn its first sales and reviews. Your early conversion rate will be low, so the click limits will look too strict. Plan a set launch budget and judge it on rank, not ACoS.

Seasonal products. A 14-day test in the off season tells you little. Run the test 4 to 6 weeks before peak demand, and expect conversion to change as the season moves.

Very low or very high prices. A $9 product may never earn back a $1.50 click. A $300 product may convert at 2%, which means 100 clicks before you negate. In both cases, work out your break-even ACoS (your profit margin before ads) before you spend. The formula still works, but the numbers get very large or very small.

The 2x rule is our operating rule, not an Amazon rule. Some sellers use 1.5x to save money faster, while others use 3x for high-margin products.

Frequently asked questions

What is the best Amazon PPC strategy for beginners?

The best start is one auto and one manual campaign per product, on down only bidding. Run both for 14 days. Then move terms with 2 or more orders to exact match, block them in the source campaign, and negate terms past your click limit. Scale bids only after 30 days of data.

Should I start with auto or manual campaigns?

Start with both at the same time. The auto campaign finds search terms you missed, and the manual campaign tests your own keyword list. Running auto alone leaves you with no control over bids on proven terms. Running manual alone means you only learn about words you already guessed.

Which bidding strategy should I use first?

Use dynamic bids, down only. Amazon lowers your bid when a sale looks less likely and never raises it above what you set. That keeps early costs in check while your data is thin. Move a campaign to up and down only after a month of sales at your target ACoS.

How much should I spend on my first Amazon PPC campaign?

Budget enough clicks to judge each keyword. Multiply your keywords by the clicks each needs, divide by 14 days, then multiply by your suggested bid. For 10 keywords at $0.90 and 20 clicks each, that is about $12.60 a day, plus a similar amount for auto.

How do I find negative keywords on Amazon?

Open the search term report and sort by spend. Filter for terms with zero orders. Look for wrong buyer words like “free” or “used”, sizes you do not sell, competitor brands, and the wrong product type. Negate a term once it passes 2 times the clicks you expect per sale.

When should I add a negative keyword?

Add one when a term reaches twice the clicks you expect per sale with no orders. Expected clicks per sale equal 1 divided by your conversion rate. At 10%, that means negating at 20 clicks. Obvious mismatches, like “used” on a new product, can go in on day one.

What is the difference between negative exact and negative phrase?

Negative exact blocks one search and its close variants. Negative phrase blocks every search that contains the phrase in order, so it reaches much wider. Amazon offers no negative broad match. Use exact for single bad terms and cross-negation, and phrase for modifiers that are always wrong.

Can I add negative keywords to auto campaigns?

Yes. Auto campaigns accept negative exact and negative phrase keywords, plus negative product targets. This matters because auto campaigns find new terms and also repeat bad ones. Add negatives there the same way you would in a manual campaign, and cross-negate any term you move to exact.

Why am I getting clicks but no sales?

Most often the listing or the traffic is wrong. Check price, images, reviews and title against the products shown beside yours. Then read the search term report for off-target searches. If traffic looks right and the listing is weak, fix the listing before you raise any bid.

How long should I wait before changing a new campaign?

Wait 14 days before any bid change. That gives Amazon time to show your ads and gives you enough clicks to read. The exception is spend: if a campaign runs out of budget by midday, raise the budget early. Negatives for obvious mismatches can also go in right away.

What is a good ACoS for a beginner?

A good ACoS is one below your break-even point, which is your profit margin before ads. If your margin is 30%, an ACoS under 30% makes money on ad sales. During the first 30 days, expect ACoS to run higher while you gather data and block wasted terms.

If you would rather hand the 90 days to a team, see our Amazon PPC management service.

Sources

Similar Posts