Amazon Campaign Bidding Strategy: How the Three Options Actually Work

The three strategies, in plain terms

When you set up or edit a Sponsored Products campaign, Amazon asks you to pick a bid strategy under Campaign Settings. There are three:

• Fixed bids — Amazon uses the exact bid you set, every time, regardless of how likely that specific impression is to convert.
• Dynamic bids, down only — Amazon will lower your bid in real time on impressions it judges less likely to convert. It never goes above your set bid.
• Dynamic bids, up and down — Amazon can both lower your bid on weak opportunities and raise it, by up to 100%, on impressions it judges highly likely to convert.

The thing to hold onto: none of these three change whether you win a specific auction. They change how much you're willing to pay for it, based on Amazon's real-time read of conversion probability at that placement, for that query, at that moment.

A worked example: the same $1.00 bid, three different outcomes

Say you set a base bid of $1.00 on a keyword. Here's what each strategy actually does with it, and this is where most explainer pages stop short — they define the setting but never show the arithmetic.

Under fixed bids, every auction you enter for that keyword bids exactly $1.00. Under dynamic down only, a low-probability impression — say, a broad match query far from your core intent — might get bid down to $0.60 or lower; a strong-intent query still bids at or near $1.00, but never above it. Under dynamic up and down, that same strong-intent query, especially at a top-of-search placement, can get bid up to $2.00 — double your entered bid — while the weak query still gets discounted the same way it would under down only.

That 100% ceiling is the number worth sitting with. If you set your base bid assuming dynamic up and down would rarely max out, and it turns out top-of-search wins are common for your keywords, your realized cost per click can run well above what the campaign dashboard implies from the bid field alone.

Bidding strategy vs. placement bid adjustments — the lever people confuse

Separate from the three strategies above, Amazon also lets you set manual percentage adjustments by placement — Top of Search, Rest of Search, Product Pages. These stack on top of whichever bidding strategy you've chosen. A campaign running dynamic up and down with a +50% Top of Search modifier can, in theory, bid well above double your base for that placement.

This is the single most common source of confusion we see: someone raises the placement modifier to chase Top of Search visibility, sees cost per click jump, and blames the bidding strategy — when the strategy setting and the placement modifier are two different dials, both turned up at once. If you're diagnosing a bid problem, check both settings before you touch either one.

When each strategy actually fits

Fixed bids suit testing periods — new keywords, new campaigns, or any time you want a clean read on what a specific bid does without Amazon's real-time adjustment layered on top. If you're trying to isolate the effect of a bid change, fixed is the control group.

Dynamic bids, down only suit campaigns where you want the efficiency benefit of not overpaying for long-shot clicks, but you're not ready to risk a doubled bid on the upside. This is the more conservative default for thinner-margin products.

Dynamic bids, up and down fit campaigns with a clear sales goal, healthy margin to absorb a doubled bid on the best opportunities, and enough history that Amazon's conversion-probability model has real signal to work with — proven keywords, overstock ASINs you're trying to move, or ASINs currently in a deal. It's the strategy built to chase Top of Search wins aggressively, and it will spend accordingly.

The common mistakes, including ones we've made

Flipping a brand-new campaign straight to dynamic up and down. With no conversion history, the algorithm has almost no signal about which impressions actually convert for that ASIN, so early spend can be erratic rather than efficient. Running fixed bids for the first stretch and switching once you have real conversion data is the safer sequence.

Judging a strategy change after three or four days. Amazon's own guidance is to review roughly every two weeks, and that's before you get into whether you've accumulated enough clicks to trust the comparison at all. We've made this mistake ourselves — reading a bad ACOS day three and reverting, when the sample was too small to mean anything.

Changing the bid strategy, the base bid, and the placement modifier all in the same week. If three things move at once and performance shifts, you can't attribute the change to any one of them. This is exactly why every proposed change carries three things before it runs: the evidence behind it, a measurement plan, and a rollback trigger — without that structure, you're guessing at which lever moved the number.

What to do when the change didn't work

If ACOS got worse after switching to dynamic up and down, check three things before you assume the setting is broken. First, has it run long enough — two weeks minimum, with meaningful click volume, not three days. Second, are your bids actually hitting the 100% ceiling on a large share of auctions — if so, the base bid was probably already too high, and dynamic bidding is just exposing that, not causing it. Third, confirm the setting actually saved; campaign manager occasionally holds a pending change if there's another edit queued on the same campaign, so what you think is live may not be.

If none of that explains it, roll back to fixed bids for a like-for-like period — same length, same keywords, same starting bid — and compare. That's the rollback trigger: a predetermined condition and a predetermined fallback, decided before you make the change, not after you're already annoyed at the number.

Where this fits if you're not doing it by hand

Reviewing bid strategy every two weeks across a real portfolio — dozens of campaigns, each with its own margin, inventory position, and conversion history — is a lot of the actual job of running Amazon ads well, and it's tedious in a way that invites exactly the shortcuts above. Dr. PPC, built by Full Circle, reads the account's real economics and proposes changes like these with the evidence, the measurement plan, and the rollback trigger attached before anything runs; the client sets how much runs automatically versus waits for a click. It's $300 a month plus 3% of ad spend, capped, month-to-month, first 30 days free, and it includes Orbit — the analytics and tracking suite most tools charge for separately — at no extra cost. That's one way to handle this; understanding the three strategies above is useful whether or not you ever use it.