Amazon PPC bidding strategies determine how Amazon uses or changes your bid in a Sponsored Products auction. Down only can reduce bids, up and down can raise or reduce them, fixed removes that conversion-based adjustment, and rule-based bidding adjusts bids around a campaign goal with a ROAS guardrail. Choose the mechanism for the campaign’s job, then review placement and audience adjustments separately.
This guide compares those mechanisms, not ad scheduling or a complete campaign launch. The platform facts come from Amazon Ads’ Bidding strategies for Sponsored Products, updated September 28, 2026; data checked 2026-11-03. These are documented controls, not results from an advertising account we tested.
Amazon PPC bidding strategies at a glance
Amazon groups the choices as dynamic, fixed, and rule-based bidding, with two dynamic variants. That gives you four practical choices to compare. The following table summarizes the official page checked 2026-11-03.
| Strategy | What Amazon does | A useful campaign fit | What it does not establish |
|---|---|---|---|
| Dynamic bids — down only | May lower bids by up to 100% across all placements when a click appears less likely to convert | Learning about traffic while limiting conversion-based bid increases | A guarantee of low ACoS or a campaign spend cap |
| Dynamic bids — up and down | May raise or lower bids by up to 100% across all placements based on performance | Established products where stronger opportunities can justify more aggressive bids | A guaranteed profitable sale or a fixed click price |
| Fixed bids | Uses the bid without dynamic conversion-based adjustment; placement increases remain available | A deliberately chosen bid baseline for evaluating traffic | A promise that every click costs the entered bid |
| Rule-based bidding | Adjusts bids up and down toward campaign goals using a ROAS guardrail, for eligible campaigns | Established campaigns with a clear return objective | Guaranteed ROAS or universal campaign eligibility |
A detail worth checking against older tutorials: the current Amazon Ads page describes the up-and-down increase as up to 100% for all placements, not just top of search. Do not build your bid review around an old non-top-placement ceiling without checking the current documentation and controls in your marketplace.
The percentage describes the bidding adjustment, not a promise that CPC will move by the same percentage. Your selected strategy also does not replace targeting, product conversion, or the amount you can afford to acquire an order. Establish the economics first with the profit-first Amazon PPC strategy guide.
Choose the mechanism for the campaign’s job
Use the strategy to answer a specific operational question. A launch campaign, a proven conversion campaign, and a large catalog with return targets need not share the same setting.
Down only: start with a bid you can defend
Amazon says down only reduces bids for opportunities it considers less likely to convert, such as less relevant queries or poorly performing placements. Its official suggested uses include new advertisers and strict budgets, checked 2026-11-03.
Use it when you want conversion-based reductions without granting that mechanism permission to increase the bid. Set the underlying bid deliberately: down only is not a substitute for deciding what traffic is worth. Also inspect placement and audience boosts before interpreting the phrase as an absolute cap at your unadjusted keyword bid.
For a new product, this is a straightforward starting point when the main question is which targets attract useful traffic. A low-converting listing still needs work on the offer; asking Amazon to reduce weak-opportunity bids cannot fix a mismatch between shopper intent and the product page.
Up and down: give stronger opportunities room
Amazon’s official guidance associates up and down with products that have good historical performance, sought-after placements, and flexible budgets. The mechanism can increase a bid for an opportunity it considers more likely to convert and lower it for a weaker one, data checked 2026-11-03.
Use it for a campaign where additional conversions have enough economic value to support more aggressive bidding. Before switching, identify what would make the change successful: more profitable orders, additional relevant reach, or a deliberate volume target within an acceptable return range.
A conversion prediction is not a margin calculation. Higher conversion likelihood can still be unattractive if the product’s contribution margin is thin. Read the resulting cost alongside sales, orders, and profitability rather than calling the strategy better because impressions increased.
Fixed: remove one automatic adjustment
Amazon says fixed bidding uses your exact bid without the dynamic conversion-based adjustment, while allowing manual base-bid changes and placement adjustments. The official page lists testing new products and working with historical data and a specific target ACoS as possible fits, checked 2026-11-03.
Choose fixed when a stable bidding mechanism helps answer your question. For example, you may want to understand how a particular base bid performs before adding conversion-based automation. Keep the other controls stable so the comparison remains interpretable.
Fixed is not synonymous with paying a fixed CPC, and it does not remove the need to manage bids. It removes one form of Amazon’s auction-time adjustment. If you increase a placement setting or change the base bid yourself, you have still changed the campaign’s bidding behavior.
Rule-based: automate against a return objective
Amazon describes rule-based bidding as automatically adjusting bids to match campaign goals, with a ROAS guardrail for Sponsored Products campaigns that meet its eligibility criteria, checked 2026-11-03. The page suggests experienced sellers, established performance history, large catalogs, and clear ROAS objectives as suitable contexts.
Check whether the campaign actually offers the option before planning around it. The cited overview does not give a universal numerical eligibility threshold, so do not assume that a particular age, spend, or order count guarantees access.
Use a return objective grounded in your economics and review actual delivery after activation. A guardrail expresses the objective guiding the automation; it is not a contractual return guarantee. Rule-based bidding also does not choose your product positioning, repair a weak listing, or turn unsuitable search terms into valuable traffic.
Separate strategy from placement and audience adjustments
Think of three distinct decisions: the base bid you enter, the bidding mechanism you select, and any additional adjustments you enable. Reviewing only the strategy name leaves part of the bidding setup unexplained.
Placement adjustments are a separate control
The official page explicitly says fixed bidding campaigns can use placement adjustments to increase bids for particular placements, data checked 2026-11-03. Therefore, fixed does not mean every eligible opportunity uses an unchanged, unadjusted base bid in every context.
Likewise, down only describes the direction of the dynamic adjustment. It does not mean you have disabled all other bid increases in the account. Review the campaign’s placement settings alongside the bidding strategy and record which adjustments are active.
Before increasing both a base bid and a placement boost, decide which question each change answers. If placement profitability is the issue, inspect placement performance rather than raising everything and hoping the blend improves. This guide does not prescribe a universal stacking formula or maximum combined multiplier: confirm the applicable controls and adjustment rules in the current documentation for your campaign.
Audience boosts do not replace strategy selection
Amazon Ads says advertisers can adjust bids by audience in addition to selecting a bidding strategy. Its page describes up to 10 audiences with independent boost percentages, while advertisers outside the US marketplace can currently select up to one audience, data checked 2026-11-03. It names Amazon audiences and Amazon Marketing Cloud audiences and points to the Audiences tab for reviewing performance.
Availability and marketplace matter. Do not copy a US campaign’s audience setup into another marketplace on the assumption that its controls are identical. When assessing a change, record the selected audiences and boosts as well as the strategy, so an audience change is not mistaken for a strategy effect.
Scheduling and budget are different questions
Rule-based bidding with a ROAS guardrail is not the same thing as scheduling bid changes by hour. For the time-of-day workflow, use the Amazon PPC dayparting guide. Keep scheduling decisions separate from the choice of auction-time bidding mechanism.
A budget concerns how much spend the campaign can support; a bid concerns how aggressively it competes for traffic. A strategy change is not, by itself, a budget increase. Use the Amazon PPC cost guide to connect bid decisions with campaign economics rather than treating the strategy selector as a cost calculator.
Change one control and read the right evidence
A useful comparison starts with a written hypothesis, not a belief that one strategy always wins. Record the campaign’s purpose, current settings, expected trade-off, and the evidence that will trigger your next decision.
Capture the baseline
Before changing the strategy, record:
- Marketplace, advertised products, campaign objective, and targeting structure.
- Base bids, placement adjustments, and audience boosts.
- Budget and any settings or rules that could change delivery.
- Impressions, clicks, spend, attributed orders, sales, and the return metric you use.
- Known stock, price, promotion, or listing changes affecting the comparison.
Choose a review window long enough to contain meaningful traffic and allow the applicable attribution reporting to mature. A universal number of days is less useful than understanding whether the campaign has enough data to support a decision.
Write a testable hypothesis
For example: switch a proven product’s campaign from down only to up and down while leaving base bids and other adjustments unchanged, then evaluate whether additional attributed sales justify the added spend. That is a proposed comparison, not a claim that we ran it or that it will win.
For a fixed-bid comparison, the hypothesis might be that removing dynamic reductions exposes more traffic at a deliberately chosen bid. Review the traffic’s relevance and return rather than concluding that more impressions alone are success.
Record exactly when you change the setting. If a promotion begins or the product goes out of stock during the comparison, note it before attributing the result to bidding. A before-and-after campaign comparison can guide operations without proving that the strategy alone caused every change.
Read volume and efficiency together
A higher return with almost no delivery may not meet a volume objective. More sales with sharply worse economics may not meet a profit objective. Put spend, sales, orders, and your profitability requirement in the same review.
If CPC rises, inspect whether order volume and conversion quality improved enough to justify it. If delivery falls, look at target relevance, bids, adjustments, and the offer before blaming the strategy selector. For rule-based bidding, compare the actual return and delivery with the chosen objective instead of assuming the guardrail means the result must already be acceptable.
Keep the next intervention small enough to interpret. Change the base bid, adjust a placement setting, revise targeting, or switch strategy based on the issue you found—not all of them simply because the first result was disappointing.
Common mistakes and activation checklist
Mistake: treating down only as an all-settings bid ceiling. Review additional boosts before deciding what the unadjusted base bid represents.
Mistake: importing an outdated up-and-down limit. The official overview checked 2026-11-03 describes increases up to 100% across all placements. Use that current scope when assessing how much room you grant the mechanism.
Mistake: assuming fixed produces a fixed CPC. It describes the bidding mechanism, not an invoice rate, and placement increases can still apply.
Mistake: reading a ROAS guardrail as guaranteed profitability. The target guides the automation; your product economics and observed results still determine whether the campaign meets its objective.
Mistake: switching strategy, bids, boosts, and budget together. That can change delivery, but it obscures which lever helped or hurt. Separate the decisions when learning is the goal.
Before activation, confirm the ad type and marketplace, the campaign’s objective, the selected strategy’s direction of adjustment, and any eligibility restrictions. Review base bids, placement settings, audience boosts, and budget together. Save the baseline, state the hypothesis, and schedule a review after the relevant data is available.
The practical choice is not the most automated setting. It is the setting whose behavior matches the campaign’s job and whose results you can evaluate without confusing it with other controls.
Frequently Asked Questions
Which Amazon PPC bidding strategy is best for a new campaign?
Start with the constraint you need to control. Down only is a straightforward option when you want conversion-based reductions without dynamic increases; fixed can suit a deliberate baseline test. Amazon’s official page lists new advertisers under down only and testing new products under fixed, checked 2026-11-03. Neither label guarantees the better result for your campaign.
Can up and down raise bids outside top of search?
Yes. Amazon’s bidding-strategies page, updated September 28, 2026 and checked 2026-11-03, says bids may increase or decrease by up to 100% for all placements. Older placement-specific limits should not override that current description.
Does fixed bidding disable placement adjustments?
No. The official page says fixed-bid campaigns can use placement bid adjustments, checked 2026-11-03. Fixed removes the dynamic conversion-based adjustment; it does not erase other controls you enable.
Is rule-based bidding the same as dayparting?
No. The rule-based strategy described here adjusts bids toward campaign goals with a ROAS guardrail. Dayparting concerns time-based changes. Decide which mechanism you need before configuring either.
Does rule-based bidding guarantee my ROAS target?
No. Amazon describes a guardrail-based objective for eligible campaigns, not guaranteed results. Check that the option is available, choose a target consistent with the product’s economics, and review actual spend, sales, and delivery after activation.