Search Campaign Optimization Framework for Ecommerce

Most Search campaign optimization is reactive—fixing problems after they appear. This framework is proactive—systematically improving every aspect of your campaigns.

The 7-Layer Framework

Layer 1: Keyword Architecture - Foundation of relevance Layer 2: Match Type Strategy - Controlling reach and precision Layer 3: Ad Copy Optimization - Maximizing click-through Layer 4: Bid Strategy Selection - Balancing volume and efficiency Layer 5: Audience Layering - Enhancing targeting Layer 6: Landing Page Alignment - Completing the journey Layer 7: Continuous Testing - Always improving

Framework Benefits

Systematic optimization delivers:

• 20-40% ROAS improvement over 90 days
• Reduced wasted spend
• Scalable campaign structure
• Predictable performance

Ecommerce Search Reality

For ecommerce, Search campaigns typically:

• Contribute 20-40% of paid revenue
• Have higher CPCs than Shopping
• Capture high-intent traffic
• Complement Shopping/PMax

Search is essential for brand defense and high-intent capture.

Keywords are the foundation. Poor architecture = wasted spend and missed opportunities.

The Tiered Keyword Model

Tier 1: Brand Keywords

• Your brand name and variations
• Branded product searches
• Expected ROAS: 8-20x

Tier 2: High-Intent Product Keywords

• Specific product searches with buying signals
• "[product name] buy," "[product] price"
• Expected ROAS: 4-8x

Tier 3: Category Keywords

• Product category searches
• Generic but relevant terms
• Expected ROAS: 2-4x

Tier 4: Long-Tail Keywords

• Specific, lower-volume searches
• Problem/solution queries
• Expected ROAS: 3-6x

Keyword Research Process

1. Mine existing data: Export search terms from Shopping/existing Search
2. Competitor research: What do competitors bid on?
3. Keyword tools: Google Keyword Planner, SEMrush, Ahrefs
4. Customer language: Reviews, support tickets, surveys

Keyword Qualification Criteria

Include keywords that are:

• Relevant to products you sell
• Have commercial intent
• Have reasonable search volume
• Have manageable CPC

Exclude keywords that are:

• Purely informational
• Too competitive for ROI
• Irrelevant to your catalog
• Already captured by Shopping

Ad Group Structure

Group keywords by:

• Theme (product type, feature)
• Intent (research vs. purchase)
• Performance (high ROAS vs. testing)

Target: 5-15 keywords per ad group for manageability.

Match types control the balance between reach and precision.

Match Type Overview

Match Type	Trigger	Control	Reach
Exact [keyword]	Exact meaning	Highest	Lowest
Phrase "keyword"	Contains phrase	Medium	Medium
Broad keyword	Related queries	Lowest	Highest

Ecommerce Match Type Strategy

For Brand Campaigns

• Exact match for pure brand terms
• Phrase match for brand + product combinations
• Avoid broad match (leaks to competitors)

For High-Intent Product Campaigns

• Exact match for proven converters
• Phrase match for exploration
• Broad match with strict negatives only

For Category Campaigns

• Phrase match as primary
• Broad match for discovery (with heavy negative management)
• Exact match for top performers

The Match Type Ladder

1. Start keywords as phrase match
2. Collect data for 2-4 weeks
3. Top performers → add as exact match
4. Poor performers → pause or negative
5. Expand winners with phrase/broad variations

Broad Match in 2026

Google's AI has improved broad match significantly. Consider broad match when:

• Smart Bidding is active (gives algorithm more signals)
• You have strong negative keyword lists
• You want to discover new queries
• You have budget for testing

Avoid broad match when:

• Manual bidding (too much variance)
• Limited budget
• New campaigns (not enough data)

Search Term Mining

Weekly process:

1. Export search terms report
2. Identify high-performing terms → add as exact
3. Identify irrelevant terms → add as negatives
4. Identify patterns → create new ad groups

Ad copy directly impacts Quality Score, CTR, and conversion rate.

RSA Best Practices

Headlines (15 available)

• Include primary keyword in 3+ headlines
• Include benefits in 3+ headlines
• Include offers/promotions in 2+ headlines
• Include brand name in 2+ headlines
• Include numbers/specifics in 2+ headlines
• Include CTAs in 2+ headlines

Descriptions (4 available)

• Lead with unique value proposition
• Include supporting benefits
• Add social proof if possible
• Clear call to action
• Include offer details

Pinning Strategy

Pin strategically:

• Pin keyword-focused headline to position 1 (relevance)
• Pin brand or CTA to position 2 (recognition)
• Leave position 3 unpinned (let Google optimize)

Don't over-pin—it limits Google's ability to optimize.

Ad Copy Testing

Create multiple RSAs per ad group:

• RSA 1: Benefit-focused messaging
• RSA 2: Offer-focused messaging
• RSA 3: Social proof-focused

Let Google find winners, then iterate.

Ad Extensions

Essential extensions:

• Sitelinks: 4+ linking to categories, bestsellers, offers
• Callouts: Free shipping, guarantees, differentiators
• Structured snippets: Product types, brands, features
• Price extensions: For product-specific campaigns
• Promotion extensions: Active sales and offers

Extensions increase ad real estate and CTR 10-15%.

Quality Score Optimization

Quality Score = f(Expected CTR, Ad Relevance, Landing Page Experience)

Improve:

• Expected CTR: Better ad copy, stronger hooks
• Ad Relevance: Keywords in headlines, tight ad groups
• Landing Page: Fast, relevant, good UX

The right bid strategy depends on your goals and data maturity.

Bid Strategy Options

Manual CPC

• Full control over bids
• Best for: New campaigns, learning phase
• Requires: Active management

Maximize Clicks

• Optimizes for click volume
• Best for: Traffic generation, new keywords
• Requires: Budget cap

Maximize Conversions

• Optimizes for conversion volume
• Best for: Campaigns with 30+ conversions/month
• Requires: Accurate conversion tracking

Target CPA

• Optimizes for cost per acquisition
• Best for: Predictable CPA goals
• Requires: 30+ conversions, stable CPA history

Target ROAS

• Optimizes for return on ad spend
• Best for: Ecommerce with varying product values
• Requires: 50+ conversions, stable ROAS history

Maximize Conversion Value

• Optimizes for total revenue
• Best for: High-value focus
• Requires: Value tracking

Strategy Selection Matrix

Conversions/Month	Goal	Recommended Strategy
<15	Learning	Manual CPC or Max Clicks
15-30	Volume	Maximize Conversions
30-50	Efficiency	Target CPA
50+	Revenue	Target ROAS

Strategy Transition

Move through strategies as campaigns mature:

1. Start: Manual CPC (control and learning)
2. 15+ conversions: Maximize Conversions (build volume)
3. 30+ conversions: Target CPA (efficiency)
4. 50+ conversions: Target ROAS (revenue optimization)

Bid Strategy by Campaign Type

Campaign Type	Recommended Strategy
Brand	Target Impression Share (90%+)
Non-Brand High-Intent	Target ROAS
Non-Brand Discovery	Maximize Conversions
Competitor	Target CPA (often lower)

Audiences enhance Search campaigns with behavioral signals.

Audience Types for Search

First-Party Audiences

• Website visitors
• Cart abandoners
• Past purchasers
• Customer lists

Google Audiences

• In-market segments
• Affinity audiences
• Detailed demographics
• Similar audiences

Application Methods

Observation Mode Add audiences to see performance, no targeting change.

• Use for: Data collection, bid adjustments

Targeting Mode Only show ads to audience members.

• Use for: RLSA campaigns, remarketing

RLSA Strategy (Remarketing Lists for Search Ads)

Layer remarketing on Search campaigns:

1. Add audiences as observation to existing campaigns
2. After 2 weeks, analyze performance by audience
3. Set bid adjustments based on data

Typical adjustments:

• Cart abandoners: +30-50%
• Product viewers: +20-30%
• Past purchasers: +10-20% (or exclude)
• All site visitors: +10-15%

Audience Exclusions

Exclude to prevent waste:

• Recent purchasers (7-14 days) from acquisition campaigns
• Existing customers from new customer campaigns
• Converters from remarketing

In-Market Layering

Add relevant in-market audiences:

• Observe performance difference
• Increase bids on high-performing segments
• Consider targeting mode for best segments

Customer Match

Upload customer lists for:

• Cross-sell campaigns (target existing customers)
• Lookalike creation (similar audiences)
• Exclusion (prevent paying for existing customers)

Audience Expansion Caution

Google offers "audience expansion." Approach carefully:

• Can broaden reach beyond your lists
• May reduce efficiency
• Test with limited budget first

Landing pages complete the journey. Misaligned pages waste ad spend.

The Alignment Principle

Ad promises must match landing page delivery:

• Keyword searched → relevant landing content
• Ad message → consistent page message
• Offer in ad → offer on page

Mismatch = bounces, low Quality Score, wasted spend.

Landing Page Types

Product Pages

• Best for: Specific product searches
• Key elements: Product info, pricing, add-to-cart

Category Pages

• Best for: Category/browsing searches
• Key elements: Filter options, product grid, navigation

Collection Pages

• Best for: Interest-based searches ("gifts for runners")
• Key elements: Curated selection, lifestyle framing

Promotional Pages

• Best for: Offer-driven campaigns
• Key elements: Deal details, urgency, featured products

Page Selection by Keyword Intent

Keyword Intent	Landing Page Type
"[product name] buy"	Product page
"[category] products"	Category page
"[brand] sale"	Sale/promo page
"best [product type]"	Collection or category
"[product] near me"	Store locator or category

Landing Page Optimization

Speed

• Target: <3 second load time
• Mobile-first optimization
• Compress images, minimize scripts

Relevance

• Keyword in page title and content
• Consistent messaging with ad
• Clear path to desired action

Trust

• Reviews/ratings visible
• Security badges
• Clear policies (returns, shipping)

Conversion

• Prominent CTA
• Easy add-to-cart
• Mobile-optimized checkout

Testing Landing Pages

A/B test page variations:

• Different headlines
• Layout variations
• CTA button colors/text
• Trust element placement

Use Google Optimize or similar tools.

Optimization never ends. Continuous testing compounds gains over time.

Testing Framework

What to Test

1. Ad copy elements

   • Headlines
   • Descriptions
   • CTAs
   • Offers

2. Keywords

   • New keyword additions
   • Match type variations
   • Negative keywords

3. Audiences

   • New segments
   • Bid adjustments
   • Exclusions

4. Bid strategies

   • Different targets
   • Strategy switches
   • Portfolio strategies

5. Landing pages

   • Page variations
   • CTAs
   • Layout

Testing Methodology

1. Hypothesis: What do you expect to improve and why?
2. Control: Current state for comparison
3. Test: Single variable change
4. Duration: 2-4 weeks minimum
5. Analysis: Statistical significance check
6. Decision: Implement or iterate

Testing Calendar

Maintain a testing calendar:

• Week 1-2: Ad copy tests
• Week 3-4: Keyword expansion tests
• Week 5-6: Audience tests
• Week 7-8: Bid strategy tests
• Repeat

Statistical Significance

Before declaring winners:

• Use A/B test calculator
• Require 95% confidence minimum
• Have sufficient sample size
• Account for time-based variance

Documenting Tests

Record every test:

• Hypothesis
• Start/end dates
• Variables tested
• Results (with confidence level)
• Decision made
• Learnings

Build institutional knowledge.

Common Testing Mistakes

Avoid:

• Testing too many variables at once
• Declaring winners too early
• Ignoring statistical significance
• Not documenting results
• Testing during unusual periods (holidays, sales)

Implement the 7-layer framework systematically.

Month 1: Foundation

Week 1-2: Layer 1-2 (Keywords & Match Types)

• Audit existing keywords
• Create tiered keyword architecture
• Implement match type strategy
• Build negative keyword lists

Week 3-4: Layer 3 (Ad Copy)

• Audit current ad copy
• Create new RSAs per ad group
• Set up ad extensions
• Launch ad copy tests

Month 2: Enhancement

Week 5-6: Layer 4 (Bid Strategies)

• Analyze conversion volume
• Select appropriate bid strategies
• Implement by campaign type
• Set initial targets

Week 7-8: Layer 5 (Audiences)

• Create audience lists
• Add as observation to campaigns
• Collect performance data
• Plan RLSA strategy

Month 3: Optimization

Week 9-10: Layer 6 (Landing Pages)

• Audit landing page alignment
• Fix misalignments
• Start landing page tests
• Optimize page speed

Week 11-12: Layer 7 (Testing)

• Establish testing calendar
• Launch first round of tests
• Create documentation system
• Review and iterate

Ongoing Optimization Cadence

Daily (5 min)

• Check spend pacing
• Monitor for anomalies
• Address urgent issues

Weekly (30 min)

• Search term review
• Performance by campaign
• Active test monitoring
• Negative keyword updates

Monthly (2 hours)

• Full account review
• Test result analysis
• Strategy adjustments
• Next month planning

Quarterly (Half day)

• Complete audit
• Structural changes
• Strategy evolution
• Goal setting