Core Philosophy: Context Over Rules
Every Amazon PPC account is different. Different products, margins, competition, lifecycle stages. The tactics that doubled profit for one brand may bankrupt another. Before applying any strategy in this guide, ask one question: "Why does this make sense for THIS account?"
There are "best practices" and then there are practices that get results. This guide focuses exclusively on the latter — frameworks grounded in math, not opinion.
Don't follow rigid SOPs blindly. Don't apply "rules" without context. Don't assume what worked for Account A works for Account B. Break problems down into fundamental truths, then rebuild solutions from the ground up.— Mohsin Raza, Adaptoid E-Commerce
First Principles Thinking
Break every performance issue down to the fundamental formula. Rising ACOS? Which lever moved — CPC, CVR, or AOV? Start with the math, not the assumption.
Data Windows Matter
Amazon attribution lags 2-14 days. Optimizing on yesterday's data is optimizing on incomplete data. Use 7-14 day windows minimum, 30+ for structural decisions.
Context Is Everything
A 50% ACOS on a new product launch is expected. A 50% ACOS on a mature product in month 18 is a fire. The same number means different things at different stages.
The ACOS Formula: Foundation of Everything
Every PPC optimization ultimately traces back to one formula. Understanding its components — and which lever to pull when — is the foundation of everything in this guide.
Where CPC = Cost Per Click (controlled by bids), CVR = Conversion Rate (influenced by targeting, listings, placements), AOV = Average Order Value (product price), and RPC = Revenue Per Click = CVR × AOV.
Rising ACOS is always caused by CPC increasing, CVR decreasing, AOV decreasing, or some combination. Identify which lever moved first — then optimize that lever. Cutting bids when the real problem is CVR just reduces traffic without fixing the underlying issue.
The 10-Step PPC Audit Framework
This is the exact framework I use when auditing a new account. It takes 30-60 minutes once you know the steps, 2-3 hours for a comprehensive first audit. Run a full audit quarterly; key steps weekly.
Total Sales Performance
Combine Business Reports (total sales) with Advertised Product Report (ad sales/spend). Find ASINs that are over-advertised, under-advertised, or draining budget with poor ROAS.
Campaign Placements
Check spend distribution across Top of Search, Rest of Search, and Product Pages. Product Pages often consume the most spend with the worst ROAS.
Bid Management
Categorize every keyword into one of four buckets (High ACOS, High Spend No Sales, Low ACOS, Low Visibility) and apply the correct RPC bid formula.
Search Term Reports
Mine search terms for branded spend leaks, irrelevant terms consuming budget, and high-converting terms ready for harvesting to manual campaigns.
Tactic Allocations
Verify spend distribution: 70-80% non-brand (growth), 10-20% competitor (market share), 5-10% brand defense (protection).
Campaign Structure
Assess naming conventions, goal segmentation, and whether the structure matches catalog size. SPCs for most accounts, SPAGs for 500+ SKUs.
Targeting Mix
Mature accounts should have 50%+ on Exact (scaling), 30-40% Broad/Phrase (discovery), 10-20% Auto (research). Inverted ratios signal structural problems.
Budget Management
Find profitable campaigns capping out early. A campaign at 25% ACOS (target: 30%) running out of budget at noon is leaving money on the table.
Ad Type Utilization
85-95% Sponsored Products (core driver), 5-15% Sponsored Brands (awareness + ToS dominance), 0-10% Sponsored Display (retargeting).
Dayparting Opportunities
Identify peak RPC hours (not peak sales hours) and shift bids accordingly. Optimize by Revenue Per Click, not by traffic volume.
Step 1 Deep Dive: Catalog Budget Distribution
Your ad spend share should roughly align with each product's sales contribution. Here's what misalignment looks like:
| ASIN | Total Sales % | Ad Spend % | Status | Action |
|---|---|---|---|---|
| A123 | 50% | 20% | ⚠ Underspending | Increase ad support — this is your winner |
| B456 | 20% | 60% | 🚫 Overspending | Reduce spend, check ROAS, possible restructure |
| C789 | 30% | 20% | ✅ Balanced | Maintain current allocation |
Step 5 Deep Dive: Ideal Tactic Allocation
Step 7 Deep Dive: Ideal Match Type Distribution
If your Auto campaigns account for 40%+ of spend on a mature account, your campaign structure is doing research when it should be scaling. Auto campaigns discover keywords. Manual Exact campaigns scale them. The flow is always: Auto → Research → Performance.
Want Me to Audit Your PPC Account?
I'll run this exact 10-step framework on your account and deliver a prioritized action plan — no obligation, no sales pitch.
Bid Optimization: The RPC Method
RPC bidding is transparent, mathematically-based, and human-controlled. No black boxes. Every bid is rooted in what a keyword actually earns per click, adjusted for your profitability target.
Where RPC = Revenue Per Click = Sales ÷ Clicks, and Target ACOS = Your profitability goal as a decimal (e.g., 30% = 0.30).
Reduce High ACOS
ACOS exceeds your target. Apply: New Bid = RPC × Target ACOS. Example: RPC $5.00, Target 30% → New Bid = $1.50
Cautious High Spend, No Sales
Spend exceeds Target CPA with zero orders. Apply: New Bid = (AOV ÷ Clicks) × Target ACOS
Scale Low ACOS
ACOS below target minus 20% buffer. Room to grow. Apply: New Bid = Current Bid × 1.05 to 1.10
Monitor Low Visibility
Clicks below aCTC — not enough data yet. Apply: New Bid = Current Bid × 1.05 to gather more signal.
The RPC formula doesn't always reduce bids. If your current bid is $0.02 but the calculated bid is $1.50, the bid will increase. Your current bid is just a status — not a data point. The formula tells you what the bid should be, not which direction to move it.
Detailed Bid Scenarios
High ACOS Example: Keyword with $5 RPC, 30% target
Given: Target ACOS: 30%, Keyword RPC: $5.00 (Sales $50 ÷ Clicks 10), Current Bid: $2.50
Calculation: New Bid = $5.00 × 0.30 = $1.50
Result: Bid decreases from $2.50 to $1.50. This keyword was overbid relative to what it earns per click at your target profitability. The new bid ensures that if you're paying $1.50 per click and earning $5.00 per click in revenue, your ACOS lands at 30%.
High Spend, No Sales: AOV $20, 10 clicks, no conversions
Given: AOV: $20, Keyword Clicks: 10 (no conversions), Target ACOS: 30%
Calculation: Anticipated RPC = $20 ÷ 10 = $2.00. New Bid = $2.00 × 0.30 = $0.60
Logic: Since there are no sales, we estimate what the RPC would be if the keyword converted at its anticipated rate. This gives a conservative bid that limits further spend while keeping the keyword alive for potential conversions.
Low ACOS: The 20% Buffer Rule explained
Given: Target ACOS: 30%, Current Keyword ACOS: 18%
Buffer Threshold: 30% - (30% × 0.20) = 24%. Since 18% < 24%, this keyword qualifies as Low ACOS.
Action: Increase bid by 5-10%. New Bid = Current Bid × 1.05 to 1.10. The buffer prevents you from over-bidding on keywords that are only slightly below target — you only scale keywords that have clear profitability headroom.
New Keyword Starting Bid (no historical data)
Formula: Starting Bid = (AOV × CVR) × Target ACOS
Example: Product price $10, CVR 10%, Target ACOS 30%
Anticipated RPC = $10 × 0.10 = $1.00. Starting Bid = $1.00 × 0.30 = $0.30
Why not use Amazon's suggested bid? Amazon's suggested bid is based on auction market price — it has no idea about YOUR margins, YOUR target ACOS, or YOUR product economics. Always start from your own math.
Placement Optimization: The CVR Method
This is one of the most misunderstood areas of Amazon PPC. Most sellers optimize placements based on ACOS. That's wrong. Placements should be optimized based on conversion rate.
Optimizing by ACOS
- Look at placement ACOS
- Increase adjustments on LOW ACOS placements
- Decrease adjustments on HIGH ACOS placements
- Use target ACOS in placement calculations
- Results: unpredictable bid behavior
Optimizing by CVR
- Compare conversion rates across placements
- Set lowest-CVR placement to 0% (baseline)
- Calculate relative CR lift for adjustments
- Control ACOS through keyword bids instead
- Results: mathematically sound allocation
| Placement | Conversion Rate | Calculation | Adjustment |
|---|---|---|---|
| Product Pages | 5.0% | Baseline (lowest) | 0% |
| Rest of Search | 7.0% | (7 - 5) / 5 = 40% lift | +40% |
| Top of Search | 9.45% | (9.45 - 5) / 5 = 89% lift | +89% |
Target ACOS is NOT part of the placement calculation. Whether your target ACOS is 15% or 30%, the placement adjustments are the same (0%, +40%, +89% in the example above). Target ACOS controls your keyword bids. Placement adjustments control where those bids get amplified based on conversion efficiency.
A higher adjustment on a high-ACOS placement can be correct. If Top of Search has the highest ACOS but also the highest conversion rate, the high ACOS is a bid problem, not a placement problem. Fix the keyword bid — don't reduce the placement modifier.
Search Term Analysis & Keyword Harvesting
Don't harvest every search term that converts. One sale doesn't mean you should promote it to a manual campaign. Use these six criteria to filter signal from noise.
| # | Criterion | Threshold | Why It Matters |
|---|---|---|---|
| 1 | Orders > 1 | At least 2 sales | Weeds out one-off flukes — most important filter |
| 2 | CVR ≥ Benchmark | Product or tactic CVR | Only harvest strong performers |
| 3 | Clicks ≥ aCTC | 1 ÷ CVR | Ensures sufficient traffic signal |
| 4 | ACOS (Optional) | Can optimize after | Don't use as a hard filter — bids fix ACOS |
| 5 | Query Matches Type | Campaign segmentation | Keeps structural integrity clean |
| 6 | Same SKU Orders | Same product, not cross-ASIN | Validates true keyword-product fit |
The Keyword Flow: Auto → Research → Performance
Negative Keyword Strategy
Irrelevant Non-Converters
- Wrong product category entirely
- Product variation you don't sell
- Consistently high ACOS (>100%)
- Zero sales with significant spend
Relevant Non-Converters
- Harvest to exact match for bid control
- Negate from research campaigns only
- Don't negate relevant terms too fast
- Seasonal patterns may be at play
N-Gram Analysis: Finding Hidden Waste at Scale
Individual search terms look harmless — $0.50 to $1.00 each. But if a word like "prescription" appears in 200 search terms at 400% ACOS, that's $150+ wasted. Individual term analysis won't reveal the pattern. N-gram analysis will.
Run n-gram analysis at account onboarding (biggest cleanup opportunity), monthly for the first few months, quarterly for mature accounts, and ad hoc after major changes. Break search terms into monograms (single words) first, filter for zero sales, sort by spend descending. If the word is irrelevant, add it as a negative phrase — one word can eliminate waste across dozens of campaigns instantly.
Drowning in Search Term Data?
I'll run an n-gram analysis on your account, find the hidden waste, and build a clean negative keyword strategy — all included in the free audit.
Campaign Structure & Match Types
The right campaign structure depends on your catalog size and management capacity. Here's the spectrum from simple to granular:
The Standard Launch Campaign Stack
Every product should launch with this 5-campaign stack that covers all traffic sources:
| Campaign | Type | Purpose |
|---|---|---|
| Auto | SP Auto | Keyword and product targeting discovery |
| Research — Broad/Phrase | SP Manual | Keyword research and expansion |
| Performance — Exact | SP Manual | Proven converting keywords, maximum control |
| Product Targeting — Expanded | SP Manual | Category-level and broad product targeting |
| Product Targeting — Individual | SP Manual | Specific competitor and complementary ASIN targeting |
Amazon's "Exact" match is NOT exact. It includes misspellings, plurals, other languages, abbreviations, acronyms, and reordered words. "Whey protein" matches "protein whey." Always check search term reports — you may be matching queries you didn't expect, even on Exact match.
Dayparting Strategy
The core principle: don't chase traffic volume. Chase high-converting traffic volume. Optimize by RPC (Revenue Per Click), not by sales volume. High sales volume with low CVR = wasted spend.
Conversion Rate by Time of Day (Example Account)
| Metric | 6-9am | 9am-12pm | 12-3pm | 3-6pm | 6-9pm | 9pm-12am |
|---|---|---|---|---|---|---|
| CVR | 8.2% | 14.1% | 13.8% | 11.4% | 9.2% | 6.8% |
| RPC | $1.64 | $2.82 | $2.76 | $2.28 | $1.84 | $1.36 |
| Traffic | Low | High | High | Highest | Highest | Medium |
| Action | +0% | +30% | +25% | +10% | -10% | -30% |
Notice: evening has the highest traffic but the lowest RPC. Most sellers increase bids during peak traffic hours because "more shoppers = more sales." That's wrong. More shoppers browsing aimlessly = more clicks that don't convert. Increase bids during high-RPC hours, decrease during low-RPC hours.
Common Dayparting Mistakes
- Chasing total sales volume (high volume ≠ profitable)
- Pausing/unpausing campaigns (zero chance of sales)
- Adjusting budgets hourly (budgets are daily settings)
- Adjusting day-parting weekly (patterns are stable)
Proper Implementation
- Calculate RPC by hour (use last 30 days)
- Identify peak RPC hours, not peak sales hours
- Increase bids 20-40% during high-CR hours
- Decrease bids 20-40% during low-CR hours
Give dayparting adjustments 2-4 weeks to stabilize before evaluating. Expected result: sales increase 10-25% with ACOS same or better. Dayparting patterns are extremely stable — once set, they rarely need more than quarterly review.
Troubleshooting Performance Issues
When ACOS rises or sales decline, follow this structured diagnostic — not panic adjustments.
Check Attribution Delays First
Is the issue only in the last 1-2 days? If yes — it's likely attribution noise. Amazon conversion data can lag 2-14 days. Always use 7+ day windows for reliable analysis.
Identify the Primary Issue
Did spend increase more than sales? → ACOS being "pulled up." Did sales drop while spend stayed flat? → CVR problem. Did CPC spike? → Bid or competition problem.
Find the Biggest Movers
Sort by DELTA (change vs prior period): Spend Delta, ACOS Delta, Sales Delta. Focus on the campaigns that moved the needle, not the ones with the worst absolute numbers.
Lower CVR → Worse placement → Lower CVR → Worse placement → Death spiral. If you see this pattern, don't cut bids — that makes it worse. Instead: run a deal or promotion to boost short-term velocity, regain Top of Search placement, higher placement = higher CVR, and the cycle reverses. Sometimes you need to spend more to fix a spending problem.
Don't sort campaigns by ACOS blindly. A campaign with 1,000% ACOS but $100 spend is inconsequential in a $50k account. Always weight by spend volume — fix the big leaks first, not the dramatic-looking small ones.
Optimization Workflow & Frequency
The biggest misconception in PPC management: more frequent optimization = better results. The opposite is true. Over-optimization leads to analysis paralysis, wasted effort, and decisions based on incomplete data.
Daily — Monitor Only
Check for anomalies, respond to clients. No optimizations. Not enough data has changed in 24 hours to make meaningful bid adjustments.
Weekly — Bid Optimizations
Apply RPC formulas across all 4 keyword categories. Use 7-14 day data windows. This is your core optimization rhythm.
Monthly — Harvesting & Placements
Keyword harvesting, placement optimizations, reporting. Use 30-day windows. These require more data to be statistically meaningful.
Quarterly — Deep Analysis
N-gram cleanup, dayparting review, market conditions assessment. Use 90-day+ data. This is your strategic review cycle.
Yearly — Full Account Review
Comprehensive audit, seasonal analysis, strategic planning for the year ahead. Everything in this guide, start to finish.
Don't include Prime Day, holiday, or promotional data in normal optimization windows. Event-driven data distorts conversion rates, CPCs, and AOV. Exclude event periods or analyze them separately.
Common Mistakes to Avoid
After auditing hundreds of accounts, these are the mistakes I see most often — organized by category so you can use this as a checklist.
Bid Optimization Mistakes
✕ Daily bid optimizations (not enough data change in 24 hours)
✕ Bulk bid changes without analyzing the underlying cause
✕ Using Amazon's suggested bid (based on auction price, not YOUR margins)
✕ Negating relevant keywords instead of reducing bids (killing potential winners)
✕ Jumping to conclusions on 1-2 day data (attribution lag makes this meaningless)
Placement Mistakes
✕ Optimizing placements based on ACOS (should use conversion rate)
✕ Using target ACOS in placement calculations (it's not part of the formula)
✕ Weekly placement adjustments (need 30-60 days of data minimum)
✕ Jumping to +200% modifiers overnight (work up gradually, validate data)
Search Term Mistakes
✕ Harvesting every converting term (one sale doesn't validate a keyword)
✕ Instant negation of relevant non-converters (seasonal patterns exist)
✕ Ignoring n-gram analysis (missing cumulative waste across campaigns)
✕ Not filtering query text by campaign type (breaks structural integrity)
Structural Mistakes
✕ Over-segmentation (too many campaigns with insufficient data per campaign)
✕ Over-aggregation (can't see product-keyword relationships)
✕ Inconsistent naming conventions (makes analysis painful at scale)
✕ Not separating brand / non-brand / competitor campaigns (mixes signals)
The goal isn't to tell you definitions. It's to show you how to think critically so that you can do this on your own. Every formula in this guide is a tool — but the judgment to know which tool to use, when, is what separates operators from strategists.— Mohsin Raza, Adaptoid E-Commerce
Quick Reference: All Formulas & Thresholds
Core Calculations
| Metric | Formula |
|---|---|
| ACOS | Spend ÷ Sales |
| RPC | Sales ÷ Clicks |
| CVR | Orders ÷ Clicks |
| CPA | Spend ÷ Orders |
| Target CPA | Target ACOS × AOV |
| aCTC | 1 ÷ CVR (or Total Clicks ÷ Orders) |
| Break-Even ACOS | (Price − COGS − Fees) ÷ Price |
Bid Formulas
| Situation | Formula |
|---|---|
| High ACOS | New Bid = Keyword RPC × Target ACOS |
| High Spend, No Sales | New Bid = (AOV ÷ Clicks) × Target ACOS |
| Low ACOS | New Bid = Current Bid × 1.05 to 1.10 |
| Low Visibility | New Bid = Current Bid × 1.05 |
| New Keyword | Starting Bid = (AOV × CVR) × Target ACOS |
| Placement Adjustment | ((Placement CR − Lowest CR) ÷ Lowest CR) × 100 |
Every formula in this guide exists to remove guesswork. Stop managing campaigns by feel. Start managing them by math. The accounts that compound growth quarter over quarter are the ones where every bid, every placement modifier, and every structural decision is grounded in data.
- ACOS = CPC ÷ (CVR × AOV). When ACOS rises, identify which lever moved. Don't cut bids if the problem is conversion rate.
- Bids should be based on RPC, not Amazon's suggestion. Your margins determine your bids. The auction determines what you actually pay.
- Placements are optimized by CVR, not ACOS. Set the lowest-converting placement to 0%, calculate relative lift for the rest.
- Most sellers optimize too frequently. Weekly bids. Monthly harvesting. Quarterly deep analysis. That's the cadence.
- The biggest waste hides in aggregate. N-gram analysis reveals patterns that individual search term review misses entirely.