Q: What happens if I change conversion goals mid-campaign?

Changing conversion goals mid-campaign (e.g., switching from "Email signups" Primary to "Purchases" Primary) triggers a Smart Bidding learning period—the algorithm treats it like a new campaign and needs 7-14 days to re-optimize for the new goal. During this learning period: (1) Performance may be volatile—CPAs might spike 30-50% above target as algorithm tests new bid levels; (2) Conversion volume may drop temporarily—algorithm explores different search queries and audiences; (3) Historical performance becomes less relevant—past CPA data was for a different goal, not comparable to new goal. Best practices when changing goals: (1) Make changes at natural break points—end of month, end of quarter, or after completing a major campaign test; (2) Expect 14 days of learning period volatility—don't panic and revert after 3 days of weird performance; (3) Increase budget temporarily if possible—gives algorithm more flexibility to explore during learning (20-30% budget increase for 14 days); (4) Review Search Terms report daily during learning—add negative keywords if algorithm explores irrelevant searches. When goal changes are necessary: (1) Business strategy shift (pivoting from lead gen to direct sales); (2) Initial goal was wrong (optimized for white papers when should have optimized for demos); (3) Seasonal changes (optimize for email list building Q1-Q3, optimize for sales Q4). Avoid changing goals: (1) Mid-month "just to see what happens"—wait for month-end; (2) Back-and-forth changes (Primary → Secondary → Primary in 2 weeks)—creates continuous learning periods with no stable optimization; (3) Due to short-term performance fluctuations—give current goal 30+ days before changing. Historical note: Before conversion goals feature (pre-2021), changing which conversions to optimize for required rebuilding entire campaigns. Conversion goals made this much easier but learning period impact still exists.

Question 1

What's the difference between Primary and Secondary conversion goals?

Accepted Answer

Primary conversion goals are the conversion actions that Smart Bidding algorithms optimize for and that appear in the "Conversions" column in reporting. Secondary conversion goals are tracked for reporting purposes only—they appear in the "All conversions" column but have zero impact on automated bidding. Example: You sell software ($1,200/year) and track three actions: (1) "Email signup" (500/month), (2) "Started trial" (150/month), (3) "Purchase" (40/month). If all three are Primary, Smart Bidding optimizes for maximum total conversions (500+150+40 = 690) and will heavily prioritize email signups since they're cheapest and most frequent. If only "Purchase" is Primary and others are Secondary, Smart Bidding ignores emails and trials completely and optimizes exclusively for the 40 purchases—even if email/trial CPAs are $5 and purchase CPA is $400, the algorithm will bid for purchases because that's the only Primary goal. When to mark an action as Primary: (1) It generates direct revenue (purchases, paid subscriptions); (2) It represents a highly qualified lead (demo requests with >30% close rates, consultation bookings); (3) You want Smart Bidding to optimize specifically for this action. When to mark as Secondary: (1) Top-of-funnel actions (content downloads, email signups, page views); (2) Micro-conversions that don't directly indicate purchase intent; (3) Actions you want to track for reporting but shouldn't influence bidding. Critical rule: Only mark an action as Primary if you're willing to pay your target CPA to acquire it. If your Target CPA is $200 and you wouldn't pay $200 for an email signup, don't mark email signups as Primary.

Question 2

How do campaign-specific conversion goals work?

Accepted Answer

Campaign-specific conversion goals allow you to override account-level default goals for individual campaigns, enabling different campaigns to optimize for different conversion actions. Setup: Tools → Conversions → Goals → Create conversion goal set → Assign to specific campaigns. Example use case: Account-level default goal is "Purchase" (bottom-funnel). But you run a top-of-funnel awareness campaign where purchases are rare and you want to optimize for "Email signups" instead. Create campaign-specific goal "Email Signups" and apply only to the awareness campaign. Now: Awareness campaign optimizes for email signups (its specific goal), all other campaigns optimize for purchases (account default). Common strategies: (1) Funnel-based goals—top-funnel campaigns optimize for content engagement, mid-funnel for trial starts, bottom-funnel for purchases; (2) Product-based goals—Campaign A for Product A conversions, Campaign B for Product B conversions; (3) Geo-based goals—US campaigns optimize for purchases, international campaigns optimize for lead forms (since international close rates are lower). When NOT to use campaign-specific goals: (1) All campaigns have same objective (all driving purchases)—use account default; (2) You lack sufficient conversion volume per campaign (each campaign needs 30+ conversions/month in its specific goal for Smart Bidding to work); (3) Goals are too granular (10 different campaign-specific goals creates management overhead). Best practice: Start with account-level goals for simplicity. Add campaign-specific goals only when you have clear funnel stages or meaningfully different objectives across campaigns. Most businesses use 2-4 total goal sets maximum (e.g., "Prospecting" campaigns = email signups, "Conversion" campaigns = purchases).

Question 3

Should I include micro-conversions in my Primary conversion goals?

Accepted Answer

Generally no—micro-conversions (email signups, content downloads, page views, add-to-cart, video views) should be Secondary, not Primary, unless they're your actual business objective. Primary goals should represent qualified demand that generates revenue or has high probability of converting to revenue. The test: Would you pay your Target CPA to acquire this conversion? If you use Target CPA bidding at $150 target and you wouldn't actually pay $150 for an email signup or white paper download, don't mark it as Primary. Smart Bidding will literally bid up to $150 to acquire those actions if they're Primary. When micro-conversions CAN be Primary: (1) Lead generation business where email is the product (email marketing agency building their list); (2) Content business where page views = ad revenue (publisher optimizing for traffic volume); (3) Early-stage campaigns with <30 macro-conversions/month—temporarily optimize for micro-conversions to build data, then switch to macro-conversions once you have volume. Better approach than marking micros as Primary: Use conversion value rules to weight macros higher. Example: "Purchase" = $100 value, "Email signup" = $5 value. Both are Primary but algorithm strongly prioritizes purchases (20x value). This gives Smart Bidding visibility into full funnel without over-optimizing for low-value actions. Real-world impact: E-commerce business marks "Add to cart" as Primary alongside "Purchase" (both Primary). Result: Smart Bidding drives 10,000 cart adds at $8 CPA and 500 purchases at $120 CPA, spending 60% of budget on cart adds. Revenue is poor because cart adds don't generate revenue. Fix: Mark only "Purchase" as Primary, "Add to cart" as Secondary. Result: Smart Bidding drives 1,200 purchases at $95 CPA, spending 100% of budget on actual revenue. Cart adds drop to 3,000 (still tracked in "All conversions" for funnel analysis) but revenue increases 140%.

Question 4

What happens if I change conversion goals mid-campaign?

Accepted Answer

Changing conversion goals mid-campaign (e.g., switching from "Email signups" Primary to "Purchases" Primary) triggers a Smart Bidding learning period—the algorithm treats it like a new campaign and needs 7-14 days to re-optimize for the new goal. During this learning period: (1) Performance may be volatile—CPAs might spike 30-50% above target as algorithm tests new bid levels; (2) Conversion volume may drop temporarily—algorithm explores different search queries and audiences; (3) Historical performance becomes less relevant—past CPA data was for a different goal, not comparable to new goal. Best practices when changing goals: (1) Make changes at natural break points—end of month, end of quarter, or after completing a major campaign test; (2) Expect 14 days of learning period volatility—don't panic and revert after 3 days of weird performance; (3) Increase budget temporarily if possible—gives algorithm more flexibility to explore during learning (20-30% budget increase for 14 days); (4) Review Search Terms report daily during learning—add negative keywords if algorithm explores irrelevant searches. When goal changes are necessary: (1) Business strategy shift (pivoting from lead gen to direct sales); (2) Initial goal was wrong (optimized for white papers when should have optimized for demos); (3) Seasonal changes (optimize for email list building Q1-Q3, optimize for sales Q4). Avoid changing goals: (1) Mid-month "just to see what happens"—wait for month-end; (2) Back-and-forth changes (Primary → Secondary → Primary in 2 weeks)—creates continuous learning periods with no stable optimization; (3) Due to short-term performance fluctuations—give current goal 30+ days before changing. Historical note: Before conversion goals feature (pre-2021), changing which conversions to optimize for required rebuilding entire campaigns. Conversion goals made this much easier but learning period impact still exists.

Question 5

How do conversion goals interact with conversion values and Target ROAS bidding?

Accepted Answer

Conversion goals determine WHICH conversion actions are optimized (Primary vs Secondary). Conversion values determine HOW MUCH each conversion is worth. Target ROAS bidding uses both: optimizes for Primary conversion goals while maximizing total conversion value. Setup hierarchy: (1) Conversion actions—what you track (purchase, signup, demo); (2) Conversion values—how much each action is worth ($100 purchase, $20 signup, $500 demo); (3) Conversion goals—which actions are Primary (purchase + demo) vs Secondary (signup); (4) Target ROAS—optimize for max revenue from Primary actions at specific efficiency (e.g., 400% ROAS target). Example: E-commerce store sells $50-500 products. Setup: Conversion action "Purchase" = Primary (included in optimization), Conversion value = dynamic transaction value ($50 for yoga mat, $500 for bike trainer). Target ROAS bidding at 400%. Smart Bidding behavior: Optimizes for purchases (Primary goal) but bids more aggressively for searches likely to produce $500 purchases vs $50 purchases because higher value helps hit 400% ROAS target. If you set "Add to cart" as Primary alongside "Purchase" (both Primary) with values: Purchase $100, Cart $5, Target ROAS 400% bidding will optimize for maximum total conversion value across both actions—but cart adds are only worth $5, so algorithm still heavily prioritizes purchases ($100 value). This is better than marking both as equal Primary without values, but still not ideal because algorithm wastes some budget on cart adds. Best practice for Target ROAS: (1) Set only revenue-generating actions as Primary (purchases, paid subscriptions); (2) Assign actual transaction values (dynamic e-commerce values or fixed lead values); (3) Mark non-revenue actions as Secondary regardless of value; (4) Use Target ROAS bidding to maximize revenue at desired efficiency. Common mistake: Setting micro-conversions as Primary with arbitrary values ($10 for email signup) hoping to give Smart Bidding "signals." This backfires—algorithm will genuinely bid for $10-value email signups, wasting budget on low-value actions instead of focusing on actual revenue-generating purchases.

Conversion Goals

Quick Answer

What is Conversion Goals?

Example

Why Conversion Goals Matters

Common Mistakes to Avoid

Best Practices for Conversion Goals

Related Terms

Conversion Tracking

Target CPA (Cost Per Acquisition)

Smart Bidding

Data-Driven Attribution (DDA)

Learn More

Frequently Asked Questions

See How Your Conversion Goals Stacks Up