The smallest Amazon PPC setup that still teaches you something is one Sponsored Products automatic campaign, one ad group holding closely related products, a $10 daily budget, no end date, and no edits for the first two weeks. Everything else — Sponsored Brands, dayparting, bid rules, third-party software — is a month-two decision at the earliest. This guide covers that minimum configuration, the arithmetic that tells you when a number is big enough to act on, and the specific settings a first-month advertiser should leave alone. Data checked 2026-09-01 against Amazon Ads’ own documentation.
This is deliberately narrower than a full playbook. If you already have campaigns running and want to lower ACoS or scale spend, read the profit-first PPC playbook instead. If you are deciding which ad products to open, start with Amazon advertising for sellers.
What You Need Before You Spend a Dollar
Three things gate your first campaign, and two of them have nothing to do with advertising.
An eligible account. Amazon Ads states that “Sponsored Products are available for professional sellers, vendors, book vendors, Kindle Direct Publishing (KDP) authors, and agencies” (Amazon Ads, Sponsored Products, checked 2026-09-01). An Individual plan does not qualify — see Individual vs. Professional plan for the cost comparison.
A product that can actually serve ads. The same page states that “Products must be in one or more eligible categories and be eligible for the Featured Offer in order to advertise.” This is the single most common reason a new campaign spends nothing: the campaign is live, but the ASIN does not hold the Featured Offer, so there is no ad to show. Confirm your status first — Featured Offer eligibility covers the requirements.
A margin number. Amazon Ads confirms that “Sponsored Products campaigns have no monthly or upfront fees” and charges on a cost-per-click basis, so the ceiling on what a click is worth comes entirely from your own unit economics. Work out your break-even ACoS before launch, not after — the method is in the ACOS guide.
The Minimum Viable Setup
One campaign. One ad group. Automatic targeting. That is the whole configuration.
Amazon Ads describes automatic targeting plainly: “With automatic targeting your ad is automatically matched with keywords and products that are similar to the product you’re advertising” (Amazon Ads, A guide to targeting with Sponsored Products, checked 2026-09-01). It sorts matches into four groups — close match, loose match, substitutes, and complements. For a first campaign that is a feature, not a compromise: you are paying Amazon to tell you which search terms your listing can win, which is exactly the information you do not have yet.
| Setting | What to choose | Why |
|---|---|---|
| Ad product | Sponsored Products | The only type that needs neither Brand Registry nor a store |
| Targeting | Automatic | Discovers search terms you would not have guessed |
| Campaigns | One | Two campaigns split the same budget and halve your data |
| Ad group | One, closely related products | Amazon Ads advises grouping “products that are closely related to each other” |
| Daily budget | $10 | Amazon Ads’ own floor for most advertisers |
| End date | None | Amazon Ads recommends “an always-on approach with your campaign by setting no end date” |
| Bidding strategy | Dynamic bids — down only | Lowers bids in real time on clicks less likely to convert |
On budget, Amazon Ads is specific: “For many advertisers, we recommend a minimum of $10 (or the equivalent for the country in which you are selling) per campaign” (Amazon Ads, Sponsored Products budget basics and best practices, checked 2026-09-01). Two mechanics on the same page matter more than the number itself. First, “Your daily budget is averaged over the course of a calendar month. That means you could spend less than or up to 25% more than your average daily budget on any given day” — so a $10 budget can bill $12.50 on a good day and still land at roughly $300 for the month. Second, when the budget runs out “your ads won’t be eligible to show until the next day, at midnight when the daily budget resets.” A campaign that empties by 10 a.m. is not a successful campaign; it is a campaign that only saw morning shoppers.
On bidding, dynamic bids — up and down can move your bid in either direction by up to 100% per impression, while down-only moves in one direction. Down-only keeps your worst case bounded to the bid you set, which is what you want while you still have no idea what a good bid is (Amazon Ads, Guide to dynamic bidding, checked 2026-09-01).
How Much Data a Decision Actually Needs
Most expensive first-month decisions are made on numbers too small to mean anything, and the fix is arithmetic rather than opinion.
Suppose your listing converts at 10%. A search term that has taken 5 clicks and produced no sale feels like a failure. Run the numbers: the chance of seeing zero sales in 5 clicks at a genuine 10% conversion rate is 0.9^5 — about 59%. A coin-flip result is not evidence.
The usable rule falls straight out of that same calculation, (1 − CR)^clicks: a search term needs roughly 3 ÷ your conversion rate clicks before “no sales” means anything. At 10% that is about 30 clicks; at 5%, about 60. Below that threshold, pausing a term is not optimisation — it is noise removal in the wrong direction, and you may be deleting your best keyword. The full probability table behind the rule, and the spend-based cut-off that runs alongside it, are in our negative keyword guide — the point where you actually start cutting terms.
Two consequences for month one. Your total click volume is capped by budget: roughly $300 ÷ your average CPC for the month, and that pool is shared across every search term the automatic campaign finds. Ten or fifteen search terms will therefore be the only ones with enough clicks to judge, and that is the honest ceiling on what one month of $10-a-day buys. And because the threshold scales with conversion rate, a weak listing needs far more spend to produce the same certainty — fixing listing conversion is cheaper than buying statistical confidence.
Your First 30 Days
Amazon Ads recommends that “In the first week, monitor your metrics at least twice a week to evaluate the effectiveness of your campaign, and adjust accordingly” — note that this says monitor, not rebuild. The two-week hold is also Amazon’s own: “We recommend that you let your automatic campaign run for about two weeks before creating a manual campaign.”
| Period | What you do | What you do not do |
|---|---|---|
| Days 1–14 | Check twice a week. Confirm the campaign is spending and impressions are accruing | Change bids, add keywords, pause search terms, or judge ACoS |
| Days 15–21 | Pull the search term report. Move terms that converted into a manual exact-match campaign. Add negatives only for clearly irrelevant terms | Add broad-match keyword lists, or negate anything that merely has a high ACoS |
| Days 22–30 | Adjust bids only on terms that have crossed your click threshold. Raise the budget only if the campaign consistently exhausts it before evening | Open a second ad type, or start editing daily |
The search term report is the single artefact that makes month one worth paying for. Amazon Ads describes it as showing “the shopping terms used by customers shopping on Amazon that resulted in at least one click on your ad” (Amazon Ads, Best practices for your Sponsored Products ads, checked 2026-09-01). Everything you do in week three comes out of that file.
There are exactly two reasons to intervene before day 14. If the campaign has served no impressions after 72 hours, that is a configuration problem, not a performance problem — check Featured Offer eligibility and ad status. If the daily budget is exhausted before midday every day, you are sampling one part of the shopping day; raise the budget or lower bids so the campaign survives to evening.
The Month-One Do-Not-Touch List
Each of these is a real feature that works. None of them works on 30 days of thin data, because every one of them either needs a baseline you do not have or splits the data you do have.
| Leave alone | Why it fails in month one |
|---|---|
| Dayparting | Needs weeks of hour-level conversion data before the pattern is distinguishable from noise |
| Budget rules and rule-based bidding | Automates decisions against a baseline you have not established |
| Sponsored Brands | Requires Brand Registry for professional sellers, and spends budget on awareness before you know what converts |
| Sponsored Display and DSP | Different objectives, different measurement windows, and they dilute a small budget |
| Bulk operations | Built for editing hundreds of campaigns; with one campaign it only adds a way to break things |
| Third-party PPC software | Optimises against data volume you do not yet generate, and adds a subscription to a $300 month |
| Placement bid multipliers | Multiplies bids on a placement split that is still statistically meaningless |
| A second campaign “to test” | Halves the click volume on both, doubling the time to any conclusion |
Sponsored Brands deserves the explicit note because sellers often try it first for the branding: “Sponsored Brands are available for vendors, book vendors, Kindle Direct Publishing (KDP) authors, agencies, and professional sellers enrolled in Amazon Brand Registry” (Amazon Ads, Sponsored Brands, checked 2026-09-01). Without Brand Registry it is not a choice you have.
What to Check, and What “Check” Means
Three numbers, twice a week, in this order.
Spend against budget. Is the campaign spending at all, and does it run past the evening? This is a plumbing check, and it is the only one that can trigger action in week one.
Clicks per search term. Not the total — the distribution. You are watching for which terms are approaching the threshold from the table above, because those are the only ones you will be allowed to act on in week three.
Orders. A raw count, not a rate. ACoS in week one has a denominator of two or three sales, which makes it a number that moves 40% on one order. Once you have a month of orders, ACoS becomes the right lens — read it against your own margin using the ACOS guide, and understand how it relates to total advertising cost of sale in TACoS vs. ACoS.
When You Have Outgrown the Minimum Setup
Three signals, any one of which means month two can begin: your automatic campaign has produced at least ten search terms above the click threshold; the campaign reliably spends its full budget without emptying before evening; and you have at least one term with a repeatable conversion rate you would defend to someone else.
At that point the structure question becomes real — separate campaigns by job, manual keyword campaigns fed by harvested terms, and negative targeting that flows both ways. That is what the profit-first PPC playbook covers, and it is also the point where keyword research starts paying for itself, because you now have real search terms to research against rather than guesses.
Frequently Asked Questions
How much should a beginner spend on Amazon PPC per day?
Amazon Ads recommends a minimum of $10 per campaign per day for most advertisers, or the local equivalent (checked 2026-09-01). Because daily budgets are averaged across a calendar month with up to 25% daily variance, $10 a day works out to roughly $300 a month. Spending less is possible, but it lengthens the time needed to reach a readable click count on any single search term.
Should I start with automatic or manual campaigns?
Automatic. Amazon Ads recommends letting an automatic campaign run for about two weeks before creating a manual campaign. Manual targeting requires you to supply the keywords, and at the start you do not know which ones your listing can actually win — that is the question the automatic campaign answers.
Why is my Amazon PPC campaign getting no impressions?
The most common cause is not bidding. Amazon requires that products be in an eligible category and eligible for the Featured Offer in order to advertise, so an ASIN that has lost the Featured Offer will not serve ads regardless of bid. Check eligibility and ad status before raising bids.
How long before Amazon PPC starts working?
Treat two weeks as the earliest point at which the automatic campaign has anything to tell you, and 30 days as the point at which individual search terms may have enough clicks to judge. “Working” in month one means producing a usable search term report, not producing profit.
Do I need PPC software as a beginner?
No. Optimisation software works on data volume, and one campaign at $10 a day does not generate enough of it — the subscription would be a meaningful share of the ad budget itself. When you do start comparing tools, independent directories that organise them by workflow, such as AMZFinder’s advertising and PPC workflow page, are a cheaper first step than starting trials you will not have time to evaluate.
Conclusion
The minimum viable Amazon PPC setup is not a beginner’s compromise — it is the configuration that produces the cleanest signal per dollar. One automatic Sponsored Products campaign, one ad group, $10 a day, no end date, and two weeks of not touching it will hand you a search term report grounded in real shopper behaviour. The discipline that makes it work is the click threshold: until a search term has roughly 3 ÷ your conversion rate clicks, “it didn’t convert” is not a finding. Month two is where structure, bids, and tooling earn their place. Month one exists to buy you the data those decisions require.