Adjust bids with a simple ruleset tied to conversion and ACoS bands instead of gut feel.

Weekly Bid Adjustment Framework for Amazon PPC

Most Amazon sellers adjust bids based on a feeling. ACoS looks high, so they cut bids. Sales slow down, so they raise them. A few weeks later they're back where they started, with worse data and no clear direction.

This is not bid management. It's noise generation.

Effective bid management requires a repeatable decision process applied on a fixed cadence to keyword-level data. The cadence is weekly. The data is per-keyword ACoS compared against a per-product target. The decisions follow a defined rule tree that removes opinion from the equation. This guide walks through every component of that system — from calculating your target bid to managing placement modifiers, seasonal adjustments, and the boundary between manual and automated optimization.

Launch Fast can help you structure your initial keyword strategy before you have performance data, so your first weekly reviews are based on cleaner signal rather than noise from poorly segmented campaigns.

Why Bid Management Without a System Fails

Random bid changes break your data continuity in two ways.

First, Amazon's algorithm takes time to learn from bid changes. When you alter a bid, the system gradually shifts where and how often your ad appears. That transition period typically takes 7–14 days to stabilize. If you change the same keyword's bid twice in one week, you're making the second decision on data that partially reflects the old bid, partially reflects the transition, and partially reflects the new bid. The signal is corrupted.

Second, without a decision rule, every bid change is a hypothesis with no control. If you raise a bid because ACoS looked good this week and the product sold well, you don't know whether the bid increase contributed to the next week's result or whether seasonality, a competitor stockout, or a pricing change was responsible. Systematic rules applied consistently are the only way to develop real causal understanding of what's driving your campaign performance.

The solution is a single-week review cycle, a minimum click threshold before acting, and a decision table that maps observed ACoS to a specific, bounded bid change.

The Core Principle: Bid to Target ACoS, Not Account Average

This is where most sellers make their first mistake.

Account-level ACoS is an average. Some keywords are profitable. Some are burning money. The average blends them together and hides both. Chasing an account-level ACoS target with account-level bid changes is like steering a car by looking at the average position of all your tires instead of the road.

Bid management happens at the keyword level. Every keyword has its own conversion rate, its own competition, and its own contribution to your account. A keyword converting at 15% with a $2.50 CPC might have a lower ACoS than a keyword converting at 6% with a $0.80 CPC — even though the second one appears cheaper.

The metric that ties these together is keyword-level ACoS:

Keyword ACoS = (Keyword Spend / Keyword Revenue) × 100

You need this number for every keyword with enough data to be meaningful — which means 15 or more clicks in the review window.

Calculating Your Keyword-Level Target Bid

Before you can evaluate whether a keyword's ACoS is acceptable, you need to know what bid produced it and what bid you should be targeting. The formula for maximum sustainable CPC is:

Target CPC = Target ACoS × Selling Price × Keyword CVR

Where:

• Target ACoS is the ACoS at which this product generates the profit margin you require (see the ACoS benchmark guide for how to calculate your break-even ACoS)
• Selling Price is your listed price
• Keyword CVR is the conversion rate for that specific keyword, calculated from Search Term Report data

Worked example:

Calculation: 0.25 × $35 × 0.09 = $0.79

If this keyword's current bid is $1.20 and CPC is averaging $0.95, you are overpaying by $0.16 per click. At 100 clicks per month, that's $16 in excess spend — and since the keyword has a set conversion rate, you cannot earn that back with a higher bid. You need to bring the bid down toward your $0.79 target.

Where to get per-keyword CVR:

Pull the Search Term Report from Seller Central under Reports > Advertising Reports > Search Term. Filter for the keyword you want, sum the clicks and orders over a 30-day window (not 7-day — CVR has high variance on short windows), and divide orders by clicks. For keywords with fewer than 50 clicks over 30 days, use your campaign-level CVR as a proxy until data accumulates.

The Search Term Impression Share report gives you the additional context of how often you're winning auctions for that term — useful when a keyword has low volume not because it converts poorly, but because your bid is too low to win impressions.

The Weekly Bid Review Process

This is the core of the system. Run this process every week, same day each week (Monday morning works well — you're reviewing the previous week's data before it's contaminated by the current week's spend).

Step 1: Pull the Search Term Report

In Seller Central: Reports > Advertising Reports > Search Term. Set the date range to the previous 7 days. Download as a CSV.

Do not use the Campaign Manager dashboard for this — the graphs are fine for orientation, but the raw Search Term Report gives you the keyword-level granularity needed for per-keyword bid decisions.

Step 2: Filter for Keywords with 15+ Clicks

In your spreadsheet, filter the "Clicks" column for rows with 15 or more clicks. This is your minimum significance threshold. Fewer than 15 clicks produces ACoS numbers that are too noisy to act on reliably — a single order can swing ACoS from 80% to 12% on a 5-click sample.

For keywords below 15 clicks: do nothing this week. Let data accumulate. An exception is keywords that have zero orders and 20+ clicks over multiple consecutive weeks — those are safe to reduce or add as negatives regardless of click count.

Step 3: Calculate Actual ACoS per Keyword

For each keyword above the threshold:

Actual ACoS = (7-day Spend / 7-day Revenue) × 100

Record this alongside your pre-calculated target ACoS for this product. The comparison between actual and target drives all bid decisions.

Step 4: Apply the Decision Tree

Examples using a 25% target ACoS:

• Keyword ACoS = 15% (60% of 25%): raise bid by 10–15%. You're getting profitable sales but likely missing volume due to low bid position.
• Keyword ACoS = 22% (88% of 25%): leave unchanged. This is healthy — you're within the acceptable range.
• Keyword ACoS = 35% (140% of 25%): reduce bid by 10–15%. The keyword is running above target but not catastrophically — a controlled reduction.
• Keyword ACoS = 55% (220% of 25%): reduce bid by 25–30% or add as a negative if this pattern has persisted for 2+ weeks.

Apply these changes one at a time per keyword. Do not apply changes to keywords that received a bid change within the last 14 days (explained in the next section).

Step 5: Review Placement Modifiers

Amazon lets you set bid multipliers for three placement types: Top of Search, Rest of Search, and Product Pages. These are percentage adjustments on top of your base bid.

If your top-of-search placements have significantly better ACoS than rest-of-search (common for exact match brand terms and high-intent category terms), add a top-of-search multiplier. If your product page placements (PDP traffic) are running high ACoS, reduce the product pages modifier — or set it to 0% to effectively eliminate PDP traffic.

Find placement data in Campaign Manager > Campaigns > select campaign > Placement tab. Compare ACoS by placement type over 30 days. If top-of-search ACoS is 18% and rest-of-search is 38% on the same campaign, your campaign-level bid is set to rest-of-search economics. A +40–50% top-of-search modifier would shift spend toward the more profitable placement without changing your baseline keyword bids.

The 14-Day Rule

Never change the same keyword bid more than once every 14 days.

Amazon's algorithm recalibrates ad delivery after a bid change. In the immediate aftermath of a change, impressions and CTR can fluctuate in ways that do not reflect the new bid's equilibrium performance. The data you collect in the first week after a bid change is transitional — acting on it with another change before the system has stabilized cascades instability.

In practice: when you make a bid change, flag that keyword in your tracking spreadsheet with a "next eligible" date 14 days out. During the following week's review, skip those keywords regardless of what the data shows. Review them the week after.

This also means you need to spread your bid changes over time if you have a large keyword list. In a 200-keyword account, you might change 20–30 bids per week — always the ones that are most out of range first (highest spend, furthest from target ACoS).

The 4-Tier Bid Priority

With limited time each week, review campaigns in this order:

Tier 1: Exact match campaigns, highest spend keywords first. These drive the most volume and have the most accurate performance signal. A 20% ACoS improvement on a $500/month keyword matters more than the same improvement on a $30/month keyword.

Tier 2: Phrase match campaigns, sorted by spend. Phrase match terms have moderate signal quality. Review them second and focus changes on the top 20% of spend concentration.

Tier 3: Broad match campaigns. Broad match data is noisier because Amazon maps your bid to a wider range of queries. Be conservative with raises here — broad match ACoS can look good for a week because Amazon happened to serve your ad on high-converting queries, then deteriorate when it pushes into irrelevant territory.

Tier 4: Auto campaigns last. Auto campaigns are discovery tools. The bids on auto campaigns are less important than the search terms you're harvesting from them. When reviewing auto campaigns, the primary action is adding converting terms as negatives in the auto campaign (to prevent it from eating volume you should own in manual campaigns) and harvesting them into exact match bids in your manual campaigns.

Dayparting and Budget Pacing

If your daily budget is consumed by noon, you are invisible to shoppers for the entire afternoon and evening — often the highest-converting hours of the day for most product categories.

Check budget pacing in Campaign Manager. If campaigns are consistently reaching their daily budget before 3pm, you have two options: increase the daily budget, or restructure campaigns to concentrate spend on your highest-performing hours.

Amazon's budget rules (Campaign Manager > Budgets > Rules) let you automate budget increases during specific time windows. A 50% budget boost rule from 6pm to 11pm prevents evening budget exhaustion without permanently inflating your daily spend.

For diagnostic purposes: download your hourly data from Seller Central to see when your orders are concentrated. If 40% of your sales happen between 7pm and midnight, and your campaigns are going dark at 2pm, that is a significant recoverable revenue leak.

Seasonal Bid Multipliers

Amazon's auction CPCs shift dramatically around major shopping events. Historical CPC increases compared to baseline:

The strategy: raise bids proactively before these windows, not reactively during them. If you wait until Prime Day to increase bids, you're responding to an auction that already inflated around you — your original bid gets outcompeted, impressions drop, and your campaign goes dark exactly when traffic volume peaks.

Pre-raise exact match bids on your top 10 keywords by 20–30% about two to three weeks before major events. Post-event, reduce bids back to pre-event levels within 7 days. Monitor daily ACoS during these windows — the higher CPCs require higher CVR to maintain target ACoS, and if your listing CVR hasn't changed, your ACoS will spike despite elevated sales volume.

Automating vs. Manual Management

Seller Central's native bid rules let you set automatic bid adjustments based on ACoS thresholds and budget caps. For accounts with fewer than 50 active keywords, manual weekly review is faster and more precise than configuring rules correctly.

For accounts with 500+ active keywords, manual review becomes untenable. At that scale, third-party tools — Perpetua, Scale Insights, Quartile, and others — apply rules-based automation continuously rather than weekly. These tools can react faster and handle volume that would take a full-time employee to manage manually.

Regardless of automation level, you should always review manually:

• Your top 10 spend keywords every week. No algorithm should have unreviewed control over your highest-spend terms.
• Any keyword that has undergone a major bid change in the past 30 days. Verify the automation moved it in the right direction.
• Seasonal pre-raise and post-raise decisions. Automation tools apply rules against historical data; they don't anticipate upcoming events unless you configure them explicitly.

Automation amplifies a good system and amplifies a bad one equally. Build the manual process first, validate that it works, then automate the repetitive components.

Building Your Bid Tracker

Maintain a simple spreadsheet with the following columns for every keyword you manage:

This tracker serves two purposes: it prevents accidental double-changes within the 14-day window, and it creates an audit trail you can review month-over-month to understand which keyword-level decisions drove account-level results.

The Weekly Bid Review in 45 Minutes

Once the system is built, a full weekly review for an account with 100–150 keywords should take 30–45 minutes:

1. Download Search Term Report (5 min)
2. Paste into tracker, filter for 15+ clicks (5 min)
3. Calculate actual ACoS per keyword, compare to target (10 min)
4. Apply decision tree, flag ineligible keywords (10 min)
5. Review placement data for top 5 campaigns (5 min)
6. Check budget pacing for campaigns with high spend (5 min)
7. Implement bid changes via bulk upload (5 min)

The process compounds over time. Better bids produce cleaner data. Cleaner data produces better bid decisions. Accounts managed with this system for 12+ months consistently outperform those managed by intuition — not because the bids are perfect in any given week, but because the cumulative direction is always toward efficiency.

Summary

Bid management is not a creative act. It is a data processing discipline. Define your target ACoS per product. Calculate the maximum CPC each keyword can justify given its conversion rate. Review actual keyword ACoS weekly against that target. Apply bounded adjustments according to a consistent decision tree. Never change the same keyword twice in 14 days. Prioritize by spend concentration. Adjust for seasonality in advance, not during.

Everything else in PPC optimization — keyword harvesting, negative keyword management, listing improvement — feeds into what this weekly process measures. Get the bid management system right, and the feedback loop from all your other work becomes legible.