Amazon’s holiday deal submission window closes on September 8. That means brands are locking their Q4 event calendar over the next four weeks — which deals run, on which ASINs, at what discount.
In roughly the same week, somebody opens the ads console, sees event-based bid rules sitting under Settings, sets a blanket increase across every campaign, and moves on. Total time spent: about ninety seconds.
After managing hundreds of brands through peak, that ninety seconds is one of the most expensive decisions in the Q4 build. Not because automating bids for high-traffic events is wrong — it’s usually right — but because three completely different systems in the Amazon console are called “bid rules,” they behave differently, and one of them has a safety mechanism that is explicitly switched off during the exact events you’re setting rules for.
Three things called “bid rules,” and they are not the same
Brands conflate these constantly, including in conversations where an agency is on the call. Get the vocabulary straight before you touch anything.
Schedule-based bid rules. You set bid increases at specific times of day, days of the week, or date ranges. Deterministic. You wrote the number, Amazon applies it, nothing is being optimized on your behalf. This is dayparting with a scheduler attached.
Event-based bid rules. Per Amazon’s own documentation, these let advertisers set predefined bid increases ahead of high shopper traffic events, and are accessed from the advertising console in the Settings section when you add a new bid rule. Amazon’s stated reason for building it is that advertisers previously had to adjust bids manually before every event and keep track of an events calendar per marketplace. Individual events are available across most marketplaces; recurring events are documented as a Japan-only feature. Still deterministic — you’re pre-scheduling a multiplier against Amazon’s event calendar instead of your own.
Rule-based bidding. Entirely different animal. This is a bidding strategy, not a scheduled adjustment. You set a target ROAS guardrail and an average bid, and Amazon adjusts your base bids up and down with each impression trying to hit that guardrail. You are handing over control of the bid itself.
The first two are a calendar. The third is an algorithm. When a brand tells us “we’ve got bid rules set for Q4,” we now ask which one, because the risk profile is not remotely comparable.
The supervision gap nobody reads about
Here’s the part that should change how you set this up.
Rule-based bidding has a built-in circuit breaker. Amazon does not guarantee it will hit your ROAS guardrail, but per Amazon’s own help documentation, if the campaign is not meeting the guardrail and ROAS drops over a 21-day period, Amazon disables the rule on your behalf and reverts to your previous bidding strategy and targeting-level bids.
That’s a reasonable safety net. Then comes the line that matters:
That 21-day period does not include special days, such as Prime Day or Black Friday.
Read it twice. The platform’s own mechanism for catching a rule that is destroying your efficiency excludes the highest-spend days of your year from its evaluation window.
The logic from Amazon’s side is defensible — peak days are outliers and you don’t want a circuit breaker tripping because of one anomalous Tuesday. But the operator consequence is blunt: on the days where you spend the most money per hour, at the highest CPCs of the year, on the traffic with the most competitive pressure, the automated supervision you were relying on is not watching.
If you are running rule-based bidding into Prime Big Deal Days and Black Friday, you are running it unsupervised on exactly the days it can do the most damage in either direction. That isn’t an argument against using it. It’s an argument for a human check on a calendar, and almost nobody has one scheduled.
There is a second guardrail worth knowing: Amazon states it will not exceed your bid guardrail by more than 25% at a campaign level — set $1 and your average CPC shouldn’t exceed $1.25. Amazon also notes that providing that guardrail may limit its ability to hit your ROAS target. Both of those are true at the same time and you have to pick which one you care about more.
A bid rule is a multiplier. What are you multiplying?
The second failure is arithmetic, not documentation.
A bid rule doesn’t set a bid. It applies a percentage increase to whatever your bid already is. Set a 40% event increase and you have not made a decision about what a click during Prime Big Deal Days is worth — you’ve made a decision to trust your current bid and add 40%.
So the question that should precede any event rule is: is the underlying bid any good?
For a meaningful share of campaigns in a typical account, the honest answer is that nobody knows. The bid was set eight months ago, adjusted twice, and has been running inside a campaign that hits its daily budget at 1pm most days. A campaign that has never run unconstrained through a full day has never produced a full-day performance number, which means the bid was calibrated against a morning-only sample. Multiply that by 1.4 on the busiest shopping day of the year and you’ve scaled a number you can’t defend.
The sequence that works is unglamorous:
Do it in the other order and the rule is amplifying noise. You have four weeks before deal submissions close and roughly eight before peak CPCs start moving — that’s enough time to do this properly, and it will not be enough time in October.
The eligibility gates that catch Q4 launches
Rule-based bidding has entry requirements that catch brands out at precisely the wrong moment. Per Amazon’s documentation, the campaign must have been running at least 30 days, have a minimum of 30 conversions in the last 30 days, and meet a minimum daily budget of $10.
Do the calendar math on a Q4 product launch. A SKU that goes live in early September will not have 30 days plus 30 conversions until October at the earliest, and probably later. Your newest, most fragile, highest-uncertainty campaigns are the ones that cannot use the automated bidding strategy — which is arguably correct, but it means your Q4 launch plan needs a manual bid management plan attached to it, run by a named person, not a rule.
Event-based and schedule-based rules don’t carry those thresholds, so they remain available on a new campaign. That’s a mixed blessing: you can absolutely apply a 50% event increase to a two-week-old campaign whose baseline bid is a guess. The platform will let you. It shouldn’t be the plan.
What to automate, and what to keep your hands on
Set rules by campaign job, never account-wide. The same 30% event increase is a smart move in one campaign and a donation in another.
Good candidates for event rules:
- Core converting keywords with 12+ months of history. You know what these are worth, the bid is calibrated, and traffic genuinely spikes. This is what the feature was built for.
- Branded defense. CPCs on your own brand terms climb during events because competitors conquest harder. A modest increase here protects cheap, high-converting traffic and is the closest thing to a free decision in this article.
- Deal-supported ASINs. If you’ve got a Prime Big Deal Days deal running on a SKU, the conversion rate on that SKU rises during the deal window, which genuinely justifies paying more per click for the same ACOS. This is the one case where the multiplier has an economic reason behind it rather than a traffic reason.
Keep manual:
- Discovery and research campaigns. These exist to find terms cheaply. Raising bids during the most expensive traffic window of the year to discover keywords is the single worst use of an event rule we see.
- Conquest and competitor targeting. Already the tier most likely to run high ACOS while feeling strategic. Automating it upward during peak is how you find out in January.
- Anything launched in the last 60 days. No baseline, no rule.
And one blunt rule of thumb on magnitude: brands routinely set event increases of 50–100% because peak feels like it deserves a big number. Start at 15–25%. CPCs during peak already run meaningfully above baseline before you touch anything — you are stacking your increase on top of a market-wide increase. A 75% rule during a window where market CPCs are already up 30–50% is not aggressive bidding, it’s a different business.
The Q4 setup, in order
Working backward from a Prime Big Deal Days event that Amazon has not officially dated but which has landed in early-to-mid October in recent years:
Now through late August. Fix budgets first. Pull the out-of-budget report, fund the chronically capped core campaigns, and let them run unconstrained for a full week. Change one lever at a time.
Late August. Take your baseline. Screenshot ACOS, CPC and impression share by campaign. You cannot evaluate a Q4 rule against a memory, and every brand that tells us “peak was rough this year” without a September baseline is describing a feeling.
Early September. Set event and schedule rules by campaign job, at 15–25%, on campaigns with real history. Deal submissions close September 8, so you’ll know which ASINs are deal-supported and can weight those separately.
Late September. Decide the rule-based bidding question deliberately. If you’re using it, put a human check on the calendar for the first and second day of every peak event, because that’s the window where Amazon’s own auto-disable is not evaluating. Somebody looks at spend and ROAS by campaign, by hand, and has the authority to pause a rule.
Do not restructure. A structural change made in September stabilizes in October, which lands your learning period on the year’s most expensive traffic. That window closed. Rules, bids and budgets are cheap; rebuilds are not.
Early January. Turn the event rules off. This is the step everyone skips. A 25% increase that made sense on Black Friday is still running on January 14 against a completely different conversion rate, and nobody notices because ACOS drift in January reads as post-holiday softness.
FAQ
Is an event-based bid rule the same as dayparting?
No. Dayparting is about which hours of the day you want exposure in and is a year-round efficiency question. Event rules are a calendar-driven multiplier for a specific high-traffic window. They stack, which is worth knowing — if you already run an aggressive evening daypart and then add a 40% event rule, Black Friday evening is running both.
Should I use rule-based bidding at all?
It works well on mature campaigns with stable conversion volume and a clear target. It works badly on thin data, and the 30-day/30-conversion threshold exists precisely because of that. If you use it, know that the auto-disable protection doesn’t evaluate peak days, and staff a manual check for those days.
How do I know whether my event rule actually helped?
Compare the event window to the same event last year and to the two weeks either side of it this year, at the campaign level, not blended. And be honest about what moved — a deal, a rank change and a bid rule all firing in the same 48 hours means you cannot attribute the result to any one of them. If you want a clean read, run the rule on half your eligible campaigns.
Can I set one rule across the whole account?
You can, and it’s the most common mistake. A rule is a statement about what an incremental click is worth, and that value differs by campaign job by a factor of several. Account-wide rules overpay in discovery and underpay in branded defense simultaneously.
What about budgets during peak?
Separate lever, and the one more likely to cost you. A budget set as a dollar figure in July becomes a much tighter cap in November when CPCs are 30–50% above baseline, on the year’s highest-intent traffic. Set Q4 budgets as a percentage of contribution margin rather than carrying a dollar figure forward, and diary an early-October review.
—
The feature itself is good. Amazon built it because manually adjusting bids before every event across every marketplace was genuinely tedious, and automating tedium is what automation is for.
The problem is that a rule feels like a decision when it’s actually a deferral — you’re deciding today what you’ll pay in October, on the basis of a baseline you probably haven’t audited, with a platform-side safety net that stops evaluating on the days that matter most. Four weeks of preparation closes all three of those gaps. Four weeks in October closes none of them.
If you’re looking for a team that manages every lever — creative, advertising, and operations — Velocity Sellers works with brands doing $100K+/month on Amazon. Contact us for a free account audit.
