Smart Bidding Strategies: Which One When, and Why

Picking a smart bidding strategy is one of the most consequential decisions in a Google Ads account, and most people get it wrong by treating it as a set-and-forget toggle. The short answer: use Maximize Conversions or Target CPA when you sell leads or non-revenue actions, use Maximize Conversion Value or Target ROAS when every conversion has a different value, and only move to a target-based strategy once you have enough conversion data to feed it. The rest is timing, tracking quality, and patience.

This guide explains each strategy, the conversion volume it realistically needs, the situations where it shines, and how to switch between them without torching your account. It is written for people already running campaigns who want fewer surprises, not a glossary of definitions.

What Smart Bidding Actually Does

Smart bidding is Google’s set of automated, auction-time bidding strategies. Instead of setting one bid per keyword, you hand the bid decision to the system, which adjusts for each auction using signals like device, location, time of day, audience, and query context. Your job shifts from setting bids to setting the right goal and feeding the system clean conversion data.

That reframe matters. With smart bidding you are not managing bids anymore. You are managing inputs (conversion tracking, conversion values, targets) and outputs (results you measure independently). If the inputs are wrong, the automation will be wrong with great confidence.

The Core Strategies, and When Each One Fits

There are two families. Conversion-count strategies optimize for the number of conversions. Value-based strategies optimize for the total value of those conversions. Pick the family first, then the specific strategy.

Maximize Conversions (with or without Target CPA)

Maximize Conversions spends your full budget to get as many conversions as possible. Add a Target CPA and it tries to get those conversions at or below your chosen cost per acquisition. Use plain Maximize Conversions when you launch, gather data, or deliberately spend a fixed budget. Add the tCPA target once you know what a profitable acquisition cost looks like and want to defend it.

This is the default for lead generation, sign-ups, calls, and any action where one conversion is roughly as valuable as the next.

Maximize Conversion Value (with or without Target ROAS)

Maximize Conversion Value spends your budget to generate the most total conversion value. Add a Target ROAS and it chases a specific return on ad spend. This is the right call for e-commerce, or any account where a 20 euro order and a 2,000 euro order should not be treated the same. To use it, you must pass real conversion values into Google Ads, ideally dynamic revenue, not a static average.

Target Impression Share, Manual CPC, and the edge cases

Target Impression Share bids to appear in a chosen position, useful for brand defense or pure visibility, not for efficiency. Manual CPC still has a place in tiny accounts, brand-new campaigns with zero data, or situations where you need absolute control while you debug. These are exceptions, not the main road.

How to Choose: A Decision Path

Work through this in order.

1. Do your conversions have meaningfully different values? If yes (e-commerce, mixed baskets, tiered services), go value-based. If no (one lead is one lead), go conversion-count.
2. Do you have reliable conversion tracking right now? If not, stop. Fix tracking first. No strategy survives bad data.
3. Do you know your profitable target? If you have a proven CPA or ROAS, you can run a target-based strategy. If you are still finding it, run the untargeted version (Maximize Conversions or Maximize Conversion Value) first and let the data show where you actually land.
4. Do you have enough volume to support a target? Thin accounts should stay on the untargeted versions longer, because target-based strategies need data to hit targets without throttling delivery.

For lead-focused B2B accounts, this usually means starting on Maximize Conversions, then graduating to Target CPA once volume and tracking are solid. The mechanics of that journey are in our B2B lead generation with Google Ads guide.

The strategy is rarely the bottleneck. The conversion you are optimizing toward is. Get the right conversion, with the right value, tracked accurately, and almost any sensible strategy will perform. Get that wrong and no strategy can save you.

Switching Strategies Without Wrecking Performance

Every time you change bidding strategy or move a target significantly, the campaign enters a learning period while the system recalibrates. During this window performance is unstable and not representative. The single biggest mistake is panicking mid-learning and reverting, which restarts the clock and teaches you nothing.

A few rules keep switches sane:

• Change one thing at a time. Do not switch strategy, change budget, and rewrite ads in the same week. You will never know what caused the result.
• Set targets near reality. Pull your actual CPA or ROAS from the last 30-60 days and set the target close to it. A target far tighter than current performance chokes delivery; one far looser wastes budget.
• Give it time before judging. A learning period commonly runs around one to two weeks, sometimes longer for low-volume campaigns. Read results after it settles, not during.
• Move targets in steps. When tightening a tCPA or raising a tROAS, shift in increments of roughly 10-20 percent and let each move stabilize before the next.

The Mistakes That Quietly Kill Results

Optimizing toward the wrong conversion. Count newsletter sign-ups and sales as one undifferentiated conversion and smart bidding will happily buy cheap sign-ups and starve your sales. Separate primary and secondary conversions deliberately.

Setting impossible targets. A Target ROAS twice your historical best will not magically deliver it. The system simply restricts spend trying to reach an unreachable goal, and your volume collapses.

Ignoring conversion lag. If your sales close days or weeks after the click, recent data looks artificially weak because conversions have not landed yet. Account for this lag before declaring a strategy a failure.

Changing too often. Constant tinkering keeps campaigns in perpetual learning. Smart bidding rewards stability and enough time to gather signal.

Where This Leaves You

Start by matching the strategy family to your business (count vs value). Confirm tracking is clean and conversions are defined correctly. Begin untargeted to discover your real numbers, then graduate to a target-based strategy with the target set near your actual performance. Change one variable at a time, respect the learning period, and move targets in small steps.

That discipline, more than any specific strategy choice, separates accounts that compound from accounts that thrash. Want a second set of eyes on your setup? Our team handles strategy selection and migration as part of ongoing Google Ads management.