Amazon's algorithm shifted from matching keywords to matching customers. The campaign structures that worked in 2024 are already obsolete.
Developed by Amazon Growth Labs, an Amazon Advertising Agency, this strategic framework draws on real campaign data and hands-on account restructuring to outline how brands should adapt their PPC architecture for Amazon’s 2026 algorithm.
When a premium kitchen appliance brand came to us in March 2025, they were spending $47,000 monthly on Amazon PPC with a 42% ACoS. Their owner was frustrated: "We're advertising everywhere, but we can't tell what's working."
The diagnosis took 15 minutes. They were running just three campaigns:
- One automatic campaign with a $25,000 monthly budget
- One manual campaign mixing branded, competitor, and category keywords
- One Sponsored Brands campaign targeting 200+ keywords
Their campaign structure made optimization impossible. Branded searches (which should deliver 10-15% ACoS) were averaged together with competitor conquests (30-40% ACoS). High-performing exact match keywords competed for budget against wasteful broad match terms. Their automatic campaign was spending $800 daily on search terms they'd never reviewed.
Within 60 days of restructuring their campaigns, we reduced their ACoS to 24% while increasing revenue by 31%. The products didn't change. The listings didn't change. The budget barely changed. Only the structure changed.
But here's what makes this case study already outdated: the campaign structure that saved this brand in March 2025 won't be enough in 2026. Amazon's algorithm has fundamentally transformed from a keyword-matching search engine into a customer-matching recommendation engine. The rules of profitable PPC are being rewritten in real-time.
This guide shows you how to build campaign architecture for 2026's personalization-first algorithm - where success depends less on ranking for keywords and more on training Amazon's AI to understand who your customer is.
Understanding Amazon's 2025-2026 Algorithm Shift
What Changed (and Why It Matters)
The Old Model (2010-2024): Keyword Matching Customer searches "water bottle" → Amazon matches keywords in your campaign → Shows ranked results based on keyword relevance + bid → You win by bidding on the right keywords
The New Model (2025-2026): Customer Matching Customer browses products, searches, views competitor ASINs → Amazon's AI builds customer profile (interests, purchase patterns, demographics) → Shows personalized results based on WHO the customer is → You win by teaching Amazon's AI who your ideal customer is
The Three Pillars of Amazon's New Algorithm
- Personalization Dominates Keyword Matching In 2024, Amazon fully activated the backend attribute infrastructure they built 5-7 years ago. The algorithm now uses structured data (target_audience, use_case, special_features) to match products to customer profiles, not just search queries.
- AI Assistants Are Changing Discovery Rufus (conversational shopping AI) and Cosmo (visual understanding AI) launched in 2024. By 2026, estimates suggest 30-40% of Amazon purchases will involve AI-assisted discovery. Customers increasingly ask questions ("What's the best water bottle for hiking?") instead of searching keywords.
- Long-Term Value Metrics Override Short-Term Conversions Amazon's Long-Term Sales (LTS) metrics now track 12-month incremental sales impact. The algorithm increasingly prioritizes customer lifetime value over immediate conversion, meaning brand-building campaigns justify higher initial ACoS.
The Critical Numbers
- Average CPC exceeded $1.00 in 2025 (up from $0.60-0.80 in 2023)
- Personalized search results now comprise 60%+ of Amazon searches
- AI-assisted shopping reached 15-20% of purchases in late 2025, projected 35-40% by end of 2026
- Algorithm updates shifted from quarterly (2023) to weekly (2025) to potentially daily (2026)
Campaign Architecture for 2026: The Persona-Based Framework
Old vs. New Structure
OLD STRUCTURE (2024 and Earlier):
Portfolio: All Products
Campaign: Exact Match Keywords
Ad Group: "water bottle" (exact)
Campaign: Broad Match Keywords
Ad Group: "water bottle" (broad)
Campaign: Competitor Targeting
NEW STRUCTURE (2026 Recommendation Engine):
Portfolio: Gym & Fitness Persona
Campaign: Branded - Fitness
Campaign: Category - Fitness Keywords
Campaign: Product Targeting - Fitness Ecosystem
Campaign: Sponsored Display - Fitness Retargeting
Portfolio: Office & Professional Persona
Campaign: Branded - Professional
Campaign: Category - Office Keywords
Campaign: Product Targeting - Office Ecosystem
Why Persona-Based Portfolios Win
Traditional campaign structure optimizes for keyword match types and budget allocation by keyword performance. The 2026 structure optimizes for customer personas, purchase ecosystems, and intent signals.
Your campaign structure should send clear signals to Amazon's AI about three things:
- WHO Your Product Serves - Target audience attributes, demographic patterns, lifestyle indicators
- WHAT Problems You Solve - Use case specificity, purchase triggers, complementary product relationships
- WHEN Customers Need You - Purchase cycle patterns, seasonal demand, competitive timing
The 2026 Campaign Taxonomy
Level 1: Separate by Customer Persona (Portfolio Level)
Create portfolios based on distinct customer types:
Portfolio: Fitness Enthusiasts
- Budget: 40% (if this is your primary customer)
- Target ACoS: 25-30%
- Focus: Gym, workout, active lifestyle
Portfolio: Busy Professionals
- Budget: 30%
- Target ACoS: 20-25%
- Focus: Office, commute, productivity
Portfolio: Outdoor Adventurers
- Budget: 20%
- Target ACoS: 28-35%
- Focus: Hiking, camping, travel
Level 2: Segment by Purchase Intent (Campaign Level)
Within each persona portfolio, create campaigns by intent stage:
Campaign 1: Branded Defense
- Target: Customers already aware of your brand
- Keywords: Your brand name + product terms
- ACoS Target: 10-15%
Campaign 2: Problem-Aware (Category Terms)
- Target: Customers who know they need this product type
- Keywords: "[Persona] + [Product Category]"
- ACoS Target: 20-30%
Campaign 3: Solution-Aware (Competitive)
- Target: Customers viewing competitor products
- Strategy: Competitor ASIN targeting + competitor brand keywords
- ACoS Target: 30-40%
Campaign 4: Ecosystem Discovery (Product Targeting)
- Target: Customers buying complementary products
- Strategy: Target ASINs of products this persona buys
- ACoS Target: 35-45%
Campaign 5: Retargeting (Sponsored Display)
- Target: Warm traffic (viewed your product or competitors)
- Strategy: Views remarketing + similar audiences
- ACoS Target: 20-30%
Product Targeting: Training the Recommendation Engine
The 2026 Reality: Product Targeting > Keyword Targeting
Major shift: In 2024-2025, ASIN/product targeting campaigns increasingly outperformed keyword-only campaigns for customer acquisition.
Why it matters: When you target complementary ASINs (gym bags if you sell water bottles), you're explicitly teaching Amazon: "My product is relevant to people who buy THIS." The algorithm learns these relationships and starts showing your product organically to those shoppers. That’s going to improve your organic rank without ranking for keywords.
2026 Budget Allocation:
- 30% keyword targeting (down from 60% in 2023)
- 40% product/ASIN targeting (up from 15% in 2023)
- 30% Sponsored Display remarketing & lookalikes (up from 10% in 2023)
Mapping Your Customer's Shopping Journey
Think about what your customers buy before, during, and after they purchase from you. This creates three natural targeting opportunities:
Early Journey Products Target products customers buy when they're just starting out. If you sell water bottles, target gym bags, fitness trackers, and running shoes. You're reaching people who are setting up for fitness, not yet looking for hydration solutions. Expect to pay 35-45% ACoS here, you're introducing yourself early.
Complementary Products Target what customers buy at the same time as your product. For water bottles, that's protein powder, gym gloves, resistance bands, yoga mats. These shoppers are already active and building out their setup. Conversion rates are better here, so aim for 25-35% ACoS.
Advanced Purchase Products Target what customers buy after they've established their routine. Think electrolyte tablets, cleaning supplies, or upgraded gear. These are your lowest-cost acquisition opportunities at 20-30% ACoS, because you're reaching established, committed customers.
Why This Actually Matters
Here's what most sellers miss: your product targeting choices teach Amazon's algorithm about your customers.
When you target gym bags with your water bottle ad, some gym bag shoppers click and buy. Amazon notices this pattern and thinks: "People who buy gym bags also want water bottles." The algorithm starts organically showing your water bottle to future gym bag shoppers. Your organic visibility improves without ever ranking for new keywords.
We saw this with a client who targeted 50 complementary fitness products. After 90 days:
- The targeting campaign itself generated 15% of revenue at 32% ACoS
- But organic sales to that same customer segment jumped 67%
- Their product started appearing in the "customers also bought" section on 40 of those 50 product pages
- Total impact was 3.2x what the campaign revenue alone suggested
This compounding effect is why product targeting matters more now than keyword targeting.
Sponsored Display: The 2026 Full-Funnel Strategy
The October 2025 Optimized Targeting Launch
Amazon's Optimized Targeting uses machine learning to analyze shopping behavior patterns and identify high-intent prospects. Early data suggests 20-40% better conversion rates than manual audience targeting.
Three-Tier Structure
Tier 1: Warm Audience Recapture
- Campaign: Views Remarketing (viewed your product, didn't purchase)
- Campaign: Competitor Views (viewed competitor products)
- ACoS Target: 15-35%
Tier 2: Intelligent Prospecting
- Campaign: Optimized Targeting (let Amazon's AI decide)
- Campaign: Similar Audiences (lookalike of your converters)
- ACoS Target: 30-50%
Tier 3: Ecosystem Contextual
- Campaign: Contextual Category (browsing complementary categories)
- ACoS Target: 40-55%
Budget Allocation for the 2026 Algorithm
Portfolio-Level Allocation (by persona):
- Primary Persona: 40% (proven best-converting segment)
- Secondary Persona: 30% (strong conversion, smaller market)
- Tertiary Persona: 20% (seasonal or growing segment)
- Exploratory Persona: 10% (testing new segment)
Campaign-Level Allocation (within each persona):
- Conversion Campaigns: 55% (branded, exact match, product pages)
- Consideration Campaigns: 30% (category keywords, competitive, Sponsored Brands)
- Discovery Campaigns: 15% (automatic, ecosystem targeting, SD prospecting)
ACoS Targets by Campaign Type (2026 Update)
For HIGH-LTV Personas:
- Branded campaigns: 12-18% ACoS
- Category/Generic campaigns: 25-35% ACoS
- Competitor campaigns: 30-45% ACoS
- Product targeting: 25-40% ACoS
- SD Retargeting: 20-30% ACoS
- SD Prospecting: 35-50% ACoS
- Optimized Targeting: 40-60% ACoS (learning phase)
Calculating Break-Even with LTV
Traditional formula: Break-even ACoS = (Profit Margin ÷ Selling Price) × 100
2026 Formula (includes LTV): Break-even ACoS = ((Profit Margin + Expected Repeat Revenue) ÷ Selling Price) × 100
Example:
- Product: Water Bottle at $35
- First-Purchase Margin: $14 (40%)
- Average Repeat Purchases: 1.5x within 12 months at $10 margin each
- Traditional Break-Even: 40% ACoS
- 2026 LTV-Adjusted: 83% ACoS
For products with strong repeat purchase rates, persona-based acquisition campaigns can run at much higher ACoS than traditional break-even suggests, IF you're accurately tracking LTV by customer segment.
Leveraging AI and Automation
Amazon Marketing Cloud (AMC): The Game-Changer
AMC became accessible in Ads Console (September 2025) with no-code templates, democratizing cross-campaign attribution insights.
The critical insight: Traditional reporting says "This campaign generated X sales at Y ACoS." AMC reveals "This campaign was viewed by 70% of customers before they converted via branded search."
Real example: One of our client's Sponsored Display prospecting campaign showed 45% ACoS in standard reporting (barely profitable). AMC revealed that 68% of customers who eventually converted via branded search or organic had first seen the SD prospecting ad. When properly attributed, the SD campaign was their most valuable channel.
The AMC-Informed Campaign Structure
Discovery Layer (SD prospecting, automatic, ecosystem targeting)
- Judged on: Reach + early-stage engagement
- Typical apparent ACoS: 40-60%
- Actual contribution: Customer acquisition ignition
Consideration Layer (category keywords, competitive, Sponsored Brands)
- Judged on: Engagement + move-to-next-stage
- Typical apparent ACoS: 30-40%
- Actual contribution: Education and preference-building
Conversion Layer (exact match, branded, retargeting)
- Judged on: Conversion efficiency
- Typical apparent ACoS: 15-25%
- Actual contribution: Closing deals (but not generating them)
After implementing AMC tracking, we typically shift 15-20% more budget to discovery and consideration layers, because their true impact is 2-3X what standard reporting shows.
The Weekly Maintenance That Prevents Decay
Campaign structure only works if you maintain it. Thirty minutes per week prevents the slow degradation that kills profitability. Here's what that looks like:
Mondays (10 minutes): Check if your personas are still working Look at impression share by persona. If it's declining, the algorithm is shifting away from you. Check conversion rates: are they trending up or down? Make sure no campaigns are hitting budget caps and missing sales.
Wednesdays (10 minutes): Mine your automatic campaigns Export the last 7 days of search terms. Find anything with 2+ conversions under 30% ACoS and add it to your manual campaigns. Cut the junk with negative keywords. This is how you discover what's actually working.
Fridays (10 minutes): Adjust based on what you learned Review how different ad placements perform for each persona (top of search vs. product pages). If you have two weeks of consistent data, adjust your bids accordingly. Rebalance budget between campaigns. Check inventory levels.
Skip this routine and your campaigns degrade 15-25% within a month. We've measured this across hundreds of accounts.
Your 90-Day Transition Plan
Phase 1: Foundation (Days 1-14)
- Mine 12 months of order data for patterns
- Use Brand Analytics to identify demographics
- Analyze reviews for use case patterns
- Create 3-5 distinct customer personas
- Audit existing campaign structure
- Calculate persona-specific LTV metrics
Phase 2: Core Build (Days 15-45)
- Create persona-specific branded campaigns
- Build high-intent exact match campaigns by persona
- Launch ecosystem product targeting campaigns
- Set up automatic campaigns by persona
- Launch Sponsored Display prospecting with Optimized Targeting
Phase 3: Optimization & Expansion (Days 46-90)
- Analyze persona performance data
- Optimize budget allocation to best performers
- Refine placement modifiers by persona
- Add Sponsored Brands campaigns
- Implement event-based bid rules
- Launch AMC analysis
- Document learnings for next iteration
Realistic 90-Day Outcomes
- Overall ACoS: 15-25% improvement
- New customer rate: 40-60% of total orders
- Search term efficiency: 20-30% fewer wasted impressions
- Total revenue: +25-40%
- Profit margin: +5-10 percentage points
- Customer LTV: +20-30%
- Time spent managing: -40%
The Bottom Line: Structure as Competitive Advantage
Let's return to the kitchen appliance brand. Their transformation came from structural changes that made optimization possible: splitting campaigns by customer intent, separating match types, implementing search term harvesting, and launching ecosystem-based product targeting.
The result: ACoS dropped from 42% to 24%, revenue increased 31%.
But here's what we didn't tell them initially: Those structural changes weren't just about organization and control. They were training Amazon's algorithm to understand who their customers were and where to find more of them.
Six months later, organic sales to their PPC-targeted personas had increased 67%. They had fundamentally improved how Amazon's algorithm understood their product.
The Seven Structural Principles That Win in 2026
- Structure campaigns by customer persona, not keyword match type - Amazon's algorithm matches products to people, not keywords to searches.
- Product targeting is now more important than keyword targeting - ASIN targeting trains the algorithm about customer relationships.
- Judge campaigns on customer acquisition, not just ACoS - A 45% ACoS campaign acquiring valuable new customers may be more profitable than a 20% ACoS campaign converting existing demand.
- Build discovery → consideration → conversion funnels within each persona - Stop judging campaigns in isolation.
- Use AMC data to understand true campaign value - Standard reporting undervalues top-of-funnel campaigns by 2-3X.
- Set persona-specific ACoS targets based on LTV - Different customer segments have different lifetime values.
- Optimize weekly, not monthly - The algorithm moves too fast for monthly optimizations.
What's Coming in 2026
Based on Amazon's public roadmap and algorithm trends:
Rufus Integration with Sponsored Ads (Q2 2026) When customers ask Rufus questions, sponsored ads will appear in recommendations. Expected impact: 15-25% of PPC spend will shift to Rufus placements by the end of 2026.
Video-First Sponsored Products (H2 2026) Sponsored Products will support video creative. Video may deliver 30-50% higher CTR than image-only ads.
The Death of Broad Match as We Know It By the end of 2026, "broad match" will essentially become "let Amazon's AI choose who sees this ad based on customer profiles" rather than "show this ad for related search terms."
LTV-Based Bidding Becomes Standard Within 12-18 months, Amazon will offer LTV-optimized bidding where you input expected customer lifetime value and the algorithm bids accordingly.
Ready to Restructure?
In 2026, Amazon PPC Strategy must optimize for Amazon's AI understanding your customer profiles, not just your keyword bids. The brands that structure campaigns to train the recommendation engine will compound their advantages over time.
The kitchen appliance brand made their choice. They're now on track to hit $15M on Amazon in 2026, up from $7M in 2024.
The difference between a $47,000 problem and a scaling success story isn't creativity, listing optimization, or marketing genius.
It's structure.
And in 2026, the right structure means persona-based portfolios that teach Amazon's AI who your customer is, where they shop, and why they need your product.
Everything else is tactics.
