Smart Bidding Strategies Explained: Which One Should You Use?

Picking the right Smart Bidding strategy is one of the most consequential decisions you'll make in Google Ads. Choose correctly, and the algorithm works for you. Choose wrong, and you're fighting your own campaigns.

The problem? Google's documentation makes them all sound great. And most "expert" advice is frustratingly vague: "Use Target ROAS if you care about return on ad spend."

Thanks, that's incredibly helpful.

Let's actually dig into when each strategy works, when it fails, and how to decide for your specific situation.

The Four Smart Bidding Strategies

Google offers four conversion-focused Smart Bidding strategies:

Each strategy optimizes for a different goal, and they behave very differently in practice.

Maximize Conversions: The Volume Play

What it does: Spends your entire budget to get as many conversions as possible, with no regard for cost per conversion.

How it bids: The algorithm bids aggressively on auctions it thinks will convert, less aggressively on others. But it will always try to spend your full budget.

Best for:

Watch out:

Real-world example:

An insurance lead gen campaign launches with Maximize Conversions. Week 1 delivers 50 leads at $45 CPA. Week 2 delivers 65 leads at $62 CPA. The algorithm is finding more conversions, but at increasing cost.

This is normal. Maximize Conversions doesn't care about CPA—it's doing exactly what it's supposed to.

Pro tip: Use Maximize Conversions to build initial data, then switch to Target CPA once you have 30+ conversions and know your acceptable cost.

Maximize Conversion Value: The Revenue Focus

What it does: Spends your entire budget to generate as much conversion value (revenue) as possible.

How it bids: Similar to Maximize Conversions, but weighted toward high-value conversions. A $500 sale is prioritized over a $50 sale.

Best for:

Watch out:

Real-world example:

An online retailer uses Maximize Conversion Value. The algorithm starts prioritizing high-ticket items ($200+ products) while ignoring accessory products ($20-50). Revenue increases, but unit volume drops. This might be fine—or it might mean you're losing customers who buy accessories today and return for big purchases tomorrow.

Pro tip: If your product mix has very different margins, consider using conversion value rules to weight by profit, not revenue.

Target CPA: The Efficiency Constraint

What it does: Gets conversions at your specified cost-per-acquisition target, adjusting volume to maintain efficiency.

How it bids: The algorithm calculates the probability of conversion for each auction and bids such that, on average, your cost per conversion equals your target.

Best for:

Data requirements:

Setting the right target:

Watch out:

Real-world example:

A SaaS company sets Target CPA at $100 for free trial signups. Historical CPA was $95. The algorithm maintains $98-102 CPA for months. Then a competitor enters the market. CPA pressure increases. The algorithm, constrained to $100, cuts spend by 40% rather than pay more per conversion.

This is correct behavior—but you might prefer spending more to maintain volume.

Pro tip: Check impression share regularly. If you're targeting $100 CPA but losing 60% of auctions to budget, the algorithm is working but you may need to raise your target or budget.

Target ROAS: The Profitability Optimizer

What it does: Gets conversion value at your specified return on ad spend, balancing efficiency with volume.

How it bids: For each auction, the algorithm estimates expected conversion value and bids accordingly. If your target is 400% ROAS, it bids roughly 25% of expected conversion value.

Best for:

Data requirements:

Setting the right target:

Watch out:

Real-world example:

An apparel brand sets Target ROAS at 500%. The algorithm achieves this by focusing almost exclusively on sale items (where ROAS is naturally higher) and ignoring new season products. Revenue is efficient but growth stagnates.

Pro tip: Use campaign-level ROAS targets to balance efficiency goals with product strategy. Give new product campaigns lower ROAS targets to allow competitive bidding.

The Decision Framework

Here's how to actually choose:

Question 1: Do you have enough conversion data?

Question 2: Are all conversions valued equally?

Question 3: Do you have an efficiency constraint?

Question 4: Is this a new or mature campaign?

Common Mistakes (And How to Avoid Them)

Mistake 1: Setting Aspirational Targets

The problem: You want $50 CPA but historically achieve $80. You set Target CPA at $50.

What happens: The algorithm can't find enough $50 conversions. Volume drops 70%. You get some conversions at $50 but far fewer than before.

The fix: Start at or above historical performance. Improve incrementally (5-10% at a time). Let the algorithm optimize down to better performance rather than demanding it upfront.

Mistake 2: Switching Strategies Too Often

The problem: Performance dips, so you switch from Target ROAS to Maximize Conversions. A week later, you switch to Target CPA.

What happens: Every switch triggers a learning phase. The algorithm never has stable data to optimize from. Performance stays volatile.

The fix: Commit to a strategy for at least 2-4 weeks. Make changes based on trends, not daily fluctuations. Use experiments to test alternatives without disrupting the main campaign.

Mistake 3: Ignoring Conversion Lag

The problem: You run an eCommerce campaign with 7-day cookies. After 3 days of Target ROAS, you see poor performance and change strategy.

What happens: You didn't wait for conversions to fully attribute. Day 4-7 conversions that the algorithm drove are assigned to "after" the change, making the old strategy look worse than it was.

The fix: Wait for your full conversion window before evaluating. If purchases typically happen 5 days after click, wait 5 days + reporting lag before judging.

Mistake 4: Wrong Conversion Actions

The problem: You're tracking page views as conversions, then wondering why Target CPA bids so low.

What happens: The algorithm is optimizing for page views (which are easy and cheap), not actual purchases.

The fix: Only include meaningful conversion actions in your bidding optimization. Use primary/secondary conversion action settings to control what the algorithm optimizes for.

Mistake 5: Mismatched Goals and Strategies

The problem: You set Target ROAS at 400%, but your boss evaluates you on lead volume.

What happens: The algorithm optimizes for efficiency, volume drops, your boss is unhappy, you're confused because "ROAS is great!"

The fix: Align your bidding strategy with actual business goals. If volume matters, use a strategy that prioritizes volume. If efficiency matters, use Target CPA/ROAS.

Portfolio Bidding: The Advanced Move

For accounts with multiple campaigns, portfolio bid strategies share data across campaigns.

How it works: Instead of each campaign learning independently, a portfolio strategy pools conversion data. A campaign with 20 conversions benefits from learnings in another campaign with 30.

Best for:

Set up:

Watch out:

Testing Smart Bidding Changes

Never change bidding strategies based on theory. Test.

Using Google Ads Experiments

What to Measure

When to Conclude

Wait until:

Making the Switch

When you're ready to change bidding strategies:

From Manual to Smart Bidding

Between Smart Bidding Strategies

Conclusion: The Right Strategy Is Situational

There's no universally "best" Smart Bidding strategy. The right choice depends on:

The framework:

Master Smart Bidding, and you've mastered the core of modern Google Ads. The algorithm does the heavy lifting—your job is to point it in the right direction.