Google Ads — Skill Manual

Built from Social Media Agent Master Playbook (March 2026)

Full Google Ads playbook for 2026. Google Ads has evolved from keyword bidding into an AI-driven ecosystem. Success requires giving the AI the right signals, maintaining control over key levers, and creating a continuous learning loop.

Campaign Types Overview

Search Campaign Strategy

Keyword Strategy (2026 Standard)

• Broad match + Smart Bidding: Google identifies search intent, not just exact phrases. This is the current standard.
• Map search intent to readiness to act: High-intent terms get aggressive bids; informational terms get capped budgets
• Negative keywords: Regularly audit search terms report and exclude irrelevant queries
• Separate brand vs non-brand: ALWAYS. Different CPC, different bidding, different purpose.

Match Type Guidance

• Broad Match: Works best with Smart Bidding — gives Google room to find intent
• Phrase Match: Good for controlling relevance while allowing variation
• Exact Match: Use sparingly for highest-intent, specific queries you must control
• Negative Match: Essential — prevents waste on irrelevant queries

Ad Copy Best Practices

• Use Responsive Search Ads (RSAs) with 8-10 unique headlines and 4 description lines
• Pin headlines sparingly — preserve algorithmic optimization flexibility
• Include primary keyword in Headline 1
• Use specific numbers and data: "Save 30%" beats "Save Money"
• Include clear CTAs
• Use ad extensions aggressively:
  • Sitelinks (show additional pages)
  • Callouts (highlight features)
  • Structured snippets (list products/services)
  • Call extensions (show phone number)
  • Location extensions (critical for any local service business)

Quality Score Optimization

Quality Score determines ad rank and CPC. Optimize all three components:

• Expected CTR: Write compelling, relevant ad copy
• Ad Relevance: Tight keyword-to-ad alignment (ad speaks directly to what was searched)
• Landing Page Experience: Fast, mobile-optimized, relevant to the ad (must match search intent)

Quality Score tiers:

• 7-10: Below-average CPC, premium positions
• 4-6: Average CPC
• 1-3: Above-average CPC, limited impressions

Performance Max (PMax) Strategy

PMax is the dominant campaign type in 2026. Properly optimized PMax delivers 30-50% better returns than traditional campaign structures.

Asset Requirements for Competitive Performance

Providing limited assets starves the AI of testing options. This is the #1 mistake. If no videos are provided, Google auto-generates them — usually poorly.

Audience Signals (Critical)

Audience signals are technically "optional" but providing none wastes weeks and thousands of dollars:

• Upload customer lists as signals
• Define custom segments based on search behavior
• Include website visitor lists
• Add in-market and affinity audiences relevant to your product
• These are SIGNALS, not restrictions — Google will go beyond them

PMax Campaign Structure

Don't run one massive PMax campaign. Segment by:

• Product category or service type
• Customer acquisition vs remarketing
• Margin levels (optimize toward profitable products)
• Geographic regions (if performance varies by location)

PMax Optimization

• Feed Google real-time conversion data (use Google Ads conversion tag, not GA4 alone)
• Implement value-based bidding when possible
• Review asset performance weekly — replace underperformers
• Check search term insights monthly
• Maintain manual Shopping campaigns alongside PMax for precise control of strategic products

Bidding Strategy Framework

Progression by Campaign Maturity

Portfolio bid strategies provide 19-27% ROAS improvement typical by dynamically reallocating budget across campaigns.

Budget Rules

• High-intent search: Aggressive funding — marginal returns stay positive longest
• Mid/upper funnel: Capped budgets with clear cost ceilings
• Rebalance weekly based on marginal CPA/ROAS
• A campaign spending 3-5x target CPA with no conversions has a structural problem — pause and audit

Never Set Aggressive Targets Early

Setting Target CPA or ROAS too aggressively before the algorithm has enough data starves it of the conversions needed to learn. Start permissive, tighten gradually.

Tracking and Measurement Hierarchy

This hierarchy is non-negotiable. Bad tracking = bad optimization.

Layer 1: Google Ads Conversion Tag (Primary)

• Real-time data for Smart Bidding
• Must be set up correctly before spending significant budget
• Tag fires on the actual conversion action (purchase, lead form submit, etc.)

Layer 2: GA4 (Complement)

• Behavioral analysis and audience enrichment
• Useful for understanding the full user journey
• Import GA4 goals as secondary conversions in Google Ads

Layer 3: Enhanced Conversions

• First-party data matching for better attribution
• Hashes customer data (email, phone) and matches to Google accounts
• Improves conversion measurement as cookies deprecate

Layer 4: Offline Conversion Import

• Connect CRM data for true ROI measurement
• Critical for lead gen campaigns (track which leads actually become customers)
• Allows bidding toward leads that convert, not just any lead

Attribution

• Use data-driven attribution (not last click)
• Track full funnel: micro-conversions (email signup, add to cart) AND macro-conversions (purchase, lead submit)
• Implement UTM parameters for cross-channel tracking

Google Ads Funnel Role

Key Metrics

Getting Started (Limited Budget)

If you're new to Google Ads or working with limited budget:

Priority 1: Install Tracking BEFORE Spending

• Add Google Ads conversion tag to your website — it collects historical data from organic traffic
• Set up GA4 with goal tracking (page views, form submissions, calls)
• Add Enhanced Conversions for better attribution as cookies deprecate
• These accumulate audience data BEFORE you spend a dollar — more data = better launch performance

\x3C!-- Google Ads Conversion Tag — add to \x3Chead> of your website -->
\x3C!-- Replace AW-XXXXXXXXX/YYYYYYY with your conversion ID/label from Google Ads -->
\x3Cscript async src="https://www.googletagmanager.com/gtag/js?id=AW-XXXXXXXXX">\x3C/script>
\x3Cscript>
  window.dataLayer = window.dataLayer || [];
  function gtag(){dataLayer.push(arguments);}
  gtag('js', new Date());
  gtag('config', 'AW-XXXXXXXXX');
\x3C/script>

Priority 2: Start with Search (High-Intent)

• Start small: $10-20/day on Search is enough to learn and generate leads
• Keywords: Target high-intent, local searches specific to your service and geography
  • Format: "[your service] + [city/region]" (e.g., "web design [city name]", "website designer [area]")
  • Avoid broad informational keywords ("what is web design") — expensive and low intent
• Use location extensions immediately — critical for local service businesses
• Separate campaigns: Brand vs non-brand from Day 1

Priority 3: Build Toward PMax

• Collect creative assets before launching: photos, videos of work, client testimonials, before/after examples
• Build a customer list (even small) to use as audience signals
• PMax performs dramatically better when seeded with real customer data
• Launch PMax only after Search campaigns have generated at least 50 conversions

Commands / Triggers

• "Create Google search ads for [product/service]" → Generate RSA headlines (15) and descriptions (4) with keyword optimization
• "Write a PMax asset group for [product/service]" → Full asset suite: headlines, descriptions, image prompts, video scripts
• "Build a keyword list for [business/service]" → Research and organize keywords by intent tier with match type recommendations
• "Review my Google Ads performance" → Audit key metrics, flag issues, recommend optimizations
• "Plan a Google Ads campaign structure" → Design full campaign hierarchy: brand, non-brand, PMax
• "Generate negative keyword list for [industry]" → Build exclusion list to prevent waste
• "Set up conversion tracking" → Step-by-step guide for installing Google Ads tag and Enhanced Conversions
• "Write ad extensions for [business]" → Generate sitelinks, callouts, structured snippets, and call extensions