Smart Bidding Progression: How to Move from Maximize Conversions to tROAS

Smart Bidding progression is one of the most misunderstood parts of B2B SaaS Google Ads.

Most accounts either sit stuck on Maximize Conversions for a year with no plan to graduate, or jump straight to Target ROAS in month two and wonder why the whole account collapses.

Both are expensive mistakes, and both stem from the same root cause: treating Smart Bidding as a setting you pick once rather than a progression you move through.

I've moved 50+ B2B SaaS accounts through the full bidding progression at Scalix AI, using the same sequence every single time. 

What changes is timing, and getting the timing right is what separates accounts that compound from accounts that stall. 

Continue reading to find out how exactly I do it.

Quick overview

What you'll learn in this blog:

• Why Smart Bidding is a progression, not a setting
• When to move from Maximize Conversions to Target CPA
• What has to be true before you can safely move to Target ROAS
• The signals that tell you to hold, tighten, or reset
• Why skipping stages is the single most common Smart Bidding mistake in B2B SaaS

What is Smart Bidding progression?

Smart Bidding progression is the deliberate sequence of moving a Google Ads account through three bidding strategies as the account matures: 

• Maximize Conversions
• Target CPA
• Target ROAS

Each stage has a specific job, a specific set of prerequisites, and specific exit criteria that tell you when to move to the next one.

The reason it matters is that Smart Bidding is machine learning. Google's algorithm learns from conversion data, and the more data it has, the better it performs. Different bidding strategies require different amounts and types of data. 

If your account doesn't have conversion value data yet, Target ROAS has nothing meaningful to optimize for.

If you're still setting up the underlying account, my Google Ads for SaaS breakdown covers the fundamentals before you get to bidding decisions.

Why is Smart Bidding a progression instead of a permanent choice?

Because your account changes over time, and the algorithm needs to learn from data that reflects who actually becomes a customer.

In week one, your account has zero conversion data. Zero. Smart Bidding has nothing to optimize against. Maximize Conversions makes sense here because it tells Google to find any conversion, without constraint, so the algorithm can learn what conversions look like in your account.

By month two, you should have 30+ conversions and a stable conversion rate. Now the algorithm has data. Constraining it with a Target CPA gives it a benchmark to optimize toward, which is exactly what you want at this stage.

By month four or five, ideally your account has offline conversion tracking flowing revenue data back from your CRM. Now the algorithm knows not just what a conversion looks like, but what a valuable conversion looks like. Target ROAS becomes possible because there are actual revenue values to optimize against.

The progression matches the maturity of the account. By skipping stages, you are essentially trying to run before the account knows how to walk. 

When should you use Maximize Conversions?

Use Maximize Conversions at the start of every account, on every campaign, without exception. Don’t treat it as a permanent bidding strategy. It is where you collect as much raw data as possible. 

Maximize Conversions tells Google to spend your budget to get as many conversions as possible, with no constraint on CPA. That freedom is exactly what you want in month one, because the algorithm needs to learn who converts in your account. 

• What devices they use.
• What times of day they convert. 
• What audience signals correlate with conversions. 
• What search queries produce them. 

All of that data has to accumulate before any more constrained bidding strategy can work.

The rules I follow during Maximize Conversions:

• No target set. No tCPA, no bid caps, nothing that constrains the algorithm.
• Judge on learning, not efficiency. CPA will be higher than steady state. That's the price of data.
• Minimum 30 conversions before moving to the next stage. Some campaigns need more, but 30 is the floor.
• Budget generous enough to hit the conversion volume in a reasonable window. If your monthly budget is going to produce 10 conversions, you're not ready for Target CPA yet.

Most B2B SaaS accounts stay on Maximize Conversions for 30 to 60 days per campaign. It can be longer if conversion volume is low, or shorter if you're spending aggressively and hitting volume quickly.

When should you move to Target CPA?

Once you have 30+ conversions in the campaign and a stable conversion rate over the last two weeks. You will be tempted, but don’t do it. 

Target CPA is the long-term workhorse for most B2B SaaS accounts. It tells Google to optimize toward a specific CPA target, which gives the algorithm a clear benchmark and gives you predictable unit economics. This is where most of your accounts should live for months at a time.

The setup rules for Target CPA:

• Start at 20 to 30% above your observed CPA. If your Maximize Conversions CPA was $150, set tCPA at $180 to $195. Never tighter than that on the first move.
• Give it two weeks to stabilize at the new target before making any adjustments.
• Tighten gradually. Never more than 15 to 20% at a time. Aggressive tightening triggers extended learning phases and volatile performance.
• Watch the conversion volume. If volume drops more than 20% after a tCPA change, you’ve tightened too far.

The target cpa vs maximize conversions question comes up in almost every audit I run. And the answer is simple. It’s a progression, not a comparison. Maximize Conversions comes first. Target CPA comes after enough data exists.

Be mindful of the timing when using target CPA. Use it the moment 30+ conversions accumulate and the conversion rate stabilizes.

When should you move to Target ROAS?

Consider the following three things.

1. Offline conversion tracking is live. Without OCT, Target ROAS is optimizing toward whatever revenue values you assigned to conversion actions when you set them up, which is usually a guess. Don’t confuse it as a real ROAS target. You will be optimizing against fiction.
2. Conversion values are accurate. OCT alone isn't enough. The revenue values coming back from your CRM have to reflect actual pipeline or closed-won revenue. If your Google Ads account thinks every closed deal is worth $10,000 when actual ACVs range from $8,000 to $80,000, Target ROAS is set to optimize badly.
3. 60+ valued conversions exist in the campaign. Target ROAS needs volume to work. 60 is the practical minimum. Below that, the algorithm can't distinguish signal from noise and performance becomes erratic.

When all three are true, Target ROAS becomes the most powerful bidding strategy for B2B SaaS. 

It tells the algorithm to optimize for revenue, not conversions. This is what you really want if you're running a business that measures success in dollars closed.

If you are missing either of these three things, you are set up for a disaster. 

So, don't jump into it because a Google rep suggested it, or if your budget increased. Only do it if your prerequisites are in place. 

What signals tell you when to hold, tighten, or reset your bidding?

Three signals I watch on every account, every week.

Hold when the campaign is inside the learning phase. 

After any structural change to a campaign (new bid strategy, new conversion action, significant budget change), Google enters a learning phase that lasts 7 to 14 days. Performance during this window is not representative. Don't touch anything until the learning phase completes. Trust the process.

Tighten when performance is stable and beating target. 

If tCPA is set at $180 and actual CPA is running $145 for three weeks, you have room to tighten. Move tCPA down 10 to 15%, wait for two weeks, and then evaluate. Repeat until performance holds at target or volume starts to drop.

Reset when the account has been struggling for 30+ days despite adjustments.

Sometimes an account gets stuck in a bad learning cycle, usually because too many changes happened too fast or because the underlying conversion data got corrupted. When adjustments stop producing results, sometimes the right move is to pause the campaign, wait 48 hours, and relaunch with cleaner settings. It's an aggressive move, but it works.

I never change bids and budgets at the same time as a rule. I want all moves to be isolated to see exactly which move led to those numbers.

What does Smart Bidding progression look like when it works?

Oneleet is a good example. When they approached Scalix, they were literally starting from ground zero, in an extremely competitive space. And when I say ground zero, I mean they had never run Google Ads before. This is a niche where Vanta and Drata are running on massive budgets!

I built their account from scratch. Four campaigns (Brand, Non-Brand, Competitor, and a fourth structured for their specific ICP), full account architecture, conversion tracking connected before launch. 

Every campaign started on Maximize Conversions. As each one crossed 30 conversions, I moved it to Target CPA, set 25% above observed baseline. Then tightened gradually over the following weeks.

Within three months, we drove 301 conversions across four campaigns with conversion volume increasing every week. That kind of compounding only happens when the bidding progression is handled correctly, in sequence, and without any shortcuts.

Oneleet went on to rank #1 in compliance and raised a $33M Series A. Google Ads is now a core growth channel for them. That result was possible because we didn't skip stages, didn't rush Target CPA, and didn't jump to Target ROAS before the account was ready.

What are the biggest Smart Bidding mistakes in B2B SaaS?

Four mistakes cause almost every bidding failure I audit.

1. Setting Target CPA at launch. 

The most common mistake. Someone reads about tCPA, sees that it's more sophisticated, and sets it on day one with zero conversion data. 

Google has nothing to optimize against, so the campaign either serves almost no impressions or serves badly. 

Fix it by resetting to Maximize Conversions and waiting for 30 conversions.

2. Jumping to Target ROAS without OCT. 

The second most common mistake. Someone reads that Target ROAS is the most powerful strategy and switches without offline conversion tracking live. 

The algorithm optimizes toward fake revenue values and the account destabilizes. 

Fix it by moving back to Target CPA and setting up OCT properly first. 

I covered the full OCT setup in a separate guide, and it's the prerequisite for most of the sophistication in any B2B SaaS account.

3. Tightening Target CPA too aggressively. 

Trying to force the algorithm below what the auction supports. 

If your observed CPA is $150 and you set tCPA at $80 hoping to force efficiency, all that happens is impressions dry up, and Google can't find enough qualified traffic to fill your budget. 

Make sure to tighten gradually. 15% at a time, maximum.

4. Changing everything at once. 

New bidding strategy, new budget, new keywords, new ad copy, all in the same week. When the account moves, you have no idea what caused it. Isolate every change. 

Test one thing at a time so you know what helped and what didn’t. 

The smart bidding strategy b2b SaaS accounts need is usually simpler than teams assume. Maximize Conversions to build data. Target CPA to run the account. Target ROAS only when the prerequisites are real. That's it.

How does Smart Bidding progression interact with the rest of your account?

Bidding progression touches every other decision in the account. The structure of Google Ads accounts in B2B SaaS matters here because different campaign types need different bidding strategies at different times. For example:

• Brand campaigns can move to Target CPA fastest because they have the highest conversion rates and accumulate data quickly. 
• Non-Brand campaigns take longer because they need more conversion volume to stabilize.
• Bidding on competitor keywords for B2B SaaS is even more complex because the auctions are volatile and conversion rates fluctuate more.

Demand Gen campaigns often stay on Maximize Conversions longer than Search campaigns because upper-funnel conversion data is thinner. Performance Max works best when Target ROAS is fed by clean OCT data, which is why I never launch PMax in accounts without OCT already running.

The RSA framework for B2B SaaS I use across every account also feeds into bidding decisions. Cleaner ad copy produces higher conversion rates, which shortens the timeline to move from Maximize Conversions to Target CPA. Everything connects.

For teams thinking about broader SaaS growth strategies, bidding progression is one lever inside the larger paid media system. It matters, but it's not the only thing that matters. Getting it right is necessary but not sufficient.

Should you use Smart Bidding at all, or manual bidding?

Smart Bidding, every time. Manual bidding in 2026 is malpractice.

Manual CPC bidding gives you control over individual bids at the keyword level. That felt like an advantage a decade ago when Google's algorithm was less sophisticated. Today, Google's Smart Bidding has access to real-time signals that a human manager can't process: device, location, time of day, audience, browser, search context, and hundreds of other variables all evaluated at auction time.

The only situations where manual bidding still has a place are very new campaigns with no conversion data at all (though even here, Maximize Conversions usually works better), or specific tests where you're deliberately isolating variables for research purposes.

For running a compounding B2B SaaS account, Smart Bidding wins. The top Google Ads agencies in the US all use Smart Bidding as the default. The question is which Smart Bidding strategy, not whether to use it.

The $1M Google Ads Playbook (B2B SaaS Edition)

The full framework I use at ScalixAI. Campaign architecture. Smart Bidding progression in detail. Conversion tracking. The 30/60/90 rollout. Every framework in one place, built from nine years inside Google and $1B+ in managed ad spend.

Download the Playbook →

The Bottom Line

Smart Bidding progression is a sequence. Start with Maximize Conversions to collect data. Target CPA to run the account for the long haul. And target ROAS only when offline conversion tracking is live and revenue values are accurate. 

Trying to force the algorithm to do things it doesn't have the data to do breaks accounts.

Smart Bidding progression is what turns a Google Ads account into a compounding growth channel only if you crack this. 

Every stage builds on the last. Every constraint you introduce is grounded in the data the account has actually generated. The algorithm gets smarter every month because you're feeding it cleaner signals.

If your account is stuck on the wrong bidding strategy, or if you're not sure where you are in the progression, that's the highest-leverage question to answer this quarter.