Smart Bidding: Common Problems and How to Fix Them
Smart Bidding strategies use machine learning to optimize bids based on conversions or conversion value. When performance drops, the cause is almost always data quality, incorrect settings, or unrealistic expectations.
1. Ensure Enough Conversion Data
Problem
- Smart Bidding depends on historical conversion data
- Low volume prevents the algorithm from learning patterns
Fix
- Aim for at least 30–50 conversions per month per campaign
- If volume is low:
- Combine campaigns with similar goals
- Use Maximize Conversions temporarily
- Import offline conversions if applicable
2. Check Conversion Tracking Accuracy
Problem
- Incorrect or misfiring tags cause optimization toward wrong signals
Fix
- Audit all conversion actions (forms, purchases, calls, leads)
- Verify “Include in Conversions” is set correctly
- Confirm conversion values for Target ROAS campaigns
- Test tags to ensure each fires once per action
3. Avoid Sudden Major Changes
Problem
- Frequent changes reset the learning phase
- Volatile performance follows constant adjustments
Fix
- Limit bid or budget changes to 10–15% at a time
- Allow 7–14 days after changes before evaluating
- Avoid pausing and reactivating campaigns frequently
4. Align Bidding Strategy with Goals
Problem
- Wrong strategy selection leads to poor ROI
Fix
- Target CPA: best for stable lead generation
- Target ROAS: best for revenue-focused campaigns
- Maximize Conversions / Value: best when budget and volume are sufficient
- Avoid mixing low- and high-volume campaigns under one strategy
5. Optimize Target CPA or ROAS
Problem
- Targets set too aggressively restrict bidding
Fix
- Start with targets based on historical performance
- Adjust targets gradually, not dramatically
- Wait 1–2 weeks before making further changes
6. Use Audience Signals
Problem
- Algorithm lacks context on high-value users
Fix
- Add first-party data: remarketing lists and customer match
- Use in-market or similar audiences
- Helps guide machine learning toward higher-intent users
7. Segment High-Value Campaigns
Problem
- Mixed goals confuse bidding algorithms
Fix
- Split campaigns by:
- Product or service category
- Conversion type (lead vs sale)
- ROI or lifetime value potential
- Ensure each campaign has a single, consistent goal
8. Optimize Ad Quality and Landing Pages
Problem
- Poor ads or landing pages limit Smart Bidding effectiveness
Fix
- Improve CTR with stronger headlines and ad extensions
- Ensure landing pages are fast, relevant, and mobile-friendly
- Higher conversion rates help the algorithm bid more efficiently
9. Monitor Auction Insights
Problem
- Competitors outbidding you limits impression share
Fix
- Review impression share and outrank metrics
- Adjust bids or loosen targets when needed
- Improve Quality Score to reduce CPC pressure
10. Allow Learning Time and Monitor Trends
Problem
- Judging performance too quickly leads to bad decisions
Fix
- Allow 7–14 days after any major change
- Monitor trends in conversions, CPA/ROAS, and impression share
- Make small, data-driven adjustments only
Quick Smart Bidding Troubleshooting Framework
- Low conversion volume: Combine campaigns or switch to Maximize Conversions
- CPA/ROAS too high: Check tracking accuracy and relax targets
- Performance volatility: Reduce frequency of changes
- Low CTR: Improve ad copy and extensions
- Budget exhausted early: Increase budget or prioritize top campaigns
Smart Bidding Optimization Checklist
- ✔ Sufficient conversion volume per campaign
- ✔ Accurate conversion tracking and values
- ✔ Minimal frequent bid or budget changes
- ✔ Correct bidding strategy aligned to goals
- ✔ Realistic CPA or ROAS targets
- ✔ Audience signals applied
- ✔ Campaigns segmented by consistent objectives
- ✔ High-quality ads and landing pages
- ✔ Auction Insights monitored regularly
- ✔ Learning phase respected before optimization