Skip to content
Academy

Bidding Strategies

Manual CPC, tCPA, tROAS, Max Conversions, when each makes sense.

INTERMEDIATE·9 MIN READ·PAID ADS·UPDATED JUN 2026
Share:

Bidding Strategies

Every time someone searches a keyword on Google, an auction runs in milliseconds. A bidding strategy is the rule you give the platform: how much to bid, toward what goal, and how much to let the algorithm decide.

Get this decision right and your budget compounds. Get it wrong and you either overpay for low-value clicks or barely show up at all.

Quick Summary

  • There are two types: manual (you set the bid) and automated/smart (the algorithm sets bids using machine learning).
  • Smart bidding only works well after you have 30 to 50 conversions per month, before that, it has too little data.
  • Target CPA (cost per acquisition) and Target ROAS (return on ad spend) are the two most powerful automated strategies for mature campaigns.
  • A study of 14,584 accounts found that Max Conversion Value outperforms Max Conversions by nearly 300% on ROAS for lead generation campaigns.
  • Never set an aggressive target right away, start near your historical average and tighten it over 2-3 weeks.

Why Bidding Strategy Is the Highest-Leverage Decision in Paid Media

Most marketers obsess over ad copy and landing pages. Those matter. But the bidding strategy determines how the platform spends every dollar, before a single user even sees your ad.

A mismatched bidding strategy can waste half your budget even when everything else is right. Conversely, switching from manual bidding to the correct smart strategy at the right time can cut your cost per acquisition by 30% or more with zero changes to creative.

Note

The auction context you need to know: Google runs a real-time auction for every single ad impression. Your bid is one input. The other key input is your Quality Score (a rating of your ad relevance and landing page experience). A higher Quality Score lets you win auctions with a lower bid. This is why bidding strategy and Quality Score are always discussed together.

The Full Bidding Strategy Map

Here is every major strategy and where it fits:

The Four Strategies You Will Use Most

1. Manual CPC (Cost Per Click)

You set the maximum you will pay for each click. The platform bids up to that ceiling in every auction.

When to use it: New campaigns with little or no conversion data. You need to gather signal before handing control to automation.

Tradeoff: You miss real-time signals the algorithm can use, such as the user's device, time of day, location, and past browsing behavior. A human cannot adjust bids that fast.

A simple rule: Use Manual CPC for the first 30 to 60 days of a new campaign, or any time conversion volume drops below 25 per month.


2. Maximize Conversions

You give the platform a daily budget. It spends all of it chasing as many conversions as possible, with no cost-per-conversion constraint.

When to use it: When you want to generate conversion data fast and have a flexible budget. Think of it as "feed the machine" mode.

Important nuance from real data: A 2024 study by Optmyzr across 14,584 accounts found that using Maximize Conversions without a target actually outperformed using it with a target on both ROAS and CPA. The constraint can strangle the algorithm before it has learned enough.


3. Target CPA (tCPA)

You tell the platform your acceptable cost per acquisition, say, $30 per lead. The algorithm adjusts bids auction by auction to hit that average over time.

When to use it: Once you have at least 30 conversions in the past 30 days. Ideally 50 or more.

How it works: Some individual conversions will cost $50 and some will cost $10, but the average converges on your target. The algorithm is making thousands of micro-decisions per day that no human bidder could replicate.

The data minimum is non-negotiable: The same Optmyzr study confirmed that accounts with fewer than 25 monthly conversions underperformed with every bidding strategy, including manual. Below 25 conversions, there is simply not enough signal.


4. Target ROAS (tROAS)

Instead of capping cost per conversion, you set a revenue return target. A tROAS of 400% means "return $4 for every $1 I spend."

When to use it: E-commerce campaigns where conversions have different values. A $500 sale and a $20 sale should not be treated as identical conversion events.

Minimum requirement: Google recommends 15 to 30 conversions with revenue values tracked in the past 30 days before enabling tROAS. Without value data, the algorithm optimizes for conversions equally regardless of value.

Key insight from research: For lead generation specifically, Max Conversion Value (a cousin of tROAS) outperforms Max Conversions by approximately 300% on ROAS. Most advertisers still use Max Conversions by default, which means switching to Max Conversion Value is a fast win hiding in plain sight.


5. Target Impression Share

You tell the platform what percentage of relevant auctions you want to appear in, or that you want to appear at the top of page one. The algorithm adjusts bids to hit that target.

When to use it: Brand awareness campaigns, or defensive campaigns to protect your own brand keywords from competitors.

Not for lead gen or e-commerce: Impression share bidding ignores conversion efficiency entirely. You will pay for visibility, not results.

Real Company Case Studies

Case Study: Automated Bidding in Action

Case study 1, Smart Bidding adoption at scale (Google internal data, 2024): Google reported that over 80% of advertisers now use at least one form of Smart Bidding, up from approximately 50% in 2021. Among advertisers who properly implemented Smart Bidding after meeting the conversion threshold, one documented case showed cost per acquisition decreased by 29% and conversion volume increased by 41%, all at the same budget level within four months of switching.

Case study 2, Amazon seller dynamic bidding analysis (February 2024): A study of approximately 130,000 Amazon ad campaigns found that sellers using "down only" dynamic bidding (the algorithm can lower bids in real time but not raise them) achieved the lowest ACoS (advertising cost of sale) and the best average conversion rate. The key insight: for lower-margin products, giving the algorithm permission to pull back on poor-fit auctions mattered more than winning more impressions.

The Learning Period: The Concept Most Advertisers Ignore

Every time you switch to a smart bidding strategy, or make major changes to a campaign running smart bidding, the algorithm enters a "learning period" of 2 to 4 weeks.

During this period, performance looks worse than normal. The algorithm is testing bids across the range of auctions to build a model. Advertisers who panic and switch strategies after one week never allow the system to learn, and they restart the learning period every time they switch.

Practical rule: Give any smart bidding strategy a minimum of 4 weeks and 50 conversions before judging its performance.

Budget changes during smart bidding also require a stabilization period. Research suggests 2 to 3 weeks for performance to normalize after a significant budget increase or decrease.

Common Mistake

The three most expensive bidding mistakes:

  1. Switching to smart bidding too early. If you have fewer than 30 monthly conversions, the algorithm does not have enough signal. It will spend erratically.
  2. Setting an unrealistic target. If your historical CPA is $80 and you set a tCPA of $20, the algorithm will constrain bids so tightly that your ads barely show. Always start your target at or near your historical average and tighten it by 10-15% every two weeks.
  3. Counting the wrong conversions. If your conversion tracking fires on micro-events like "viewed a video" or "scrolled 50%", the algorithm optimizes for those. It has no idea that is not what you actually want. Only count conversions that directly represent business value before enabling smart bidding.

A Step-by-Step Switching Framework

Here is the safest way to move from manual to automated bidding:

  1. Confirm conversion tracking is working. Check that conversions fire only on high-value actions (purchases, form submissions, calls).
  2. Verify your conversion volume. You need 30 per month at minimum. 50 is better.
  3. Calculate your historical CPA or ROAS. This becomes your starting target.
  4. Switch strategy. Set your initial target at your historical average, not your goal.
  5. Do not touch anything for 4 weeks. Let the algorithm complete its learning period.
  6. After 4 weeks, evaluate. If performance is stable, tighten your target by 10-15%.
  7. Repeat step 6 every 2 weeks until you reach your goal CPA or ROAS.
Pro Tip

Pro tip, portfolio bid strategies: In Google Ads, you can apply a single Target CPA or Target ROAS strategy across multiple campaigns at once using "portfolio bid strategies." This pools conversion data across campaigns, which is useful when individual campaigns have low volume but the combined account hits the 50-conversion threshold. Found under Tools and Settings, Shared Library in Google Ads.

Manual vs. Smart: The Honest Comparison

FactorManual CPCSmart Bidding
Data neededNone30-50+ conversions/month
Real-time adjustmentsNoYes, device, time, location, audience
Learning curveLowMedium
Best forNew campaigns, testingMature campaigns with clear goals
RiskLeaves money on tableCan waste budget without sufficient data

Research from Optmyzr's 2024 study found that smart bidding's sweet spot is accounts spending $10,000 to $50,000 per month. Below $10,000 with sufficient conversions, smart bidding can still match or beat manual performance. The conversion volume threshold matters more than budget size.

The One-Line Takeaway

Match your bidding strategy to your data volume, not your ambition, and you will consistently outspend competitors who optimized copy while ignoring how the auction actually works.

  • Quality Score, Your Quality Score directly affects how much you pay per click even with the same bid. A higher score means your money goes further under any bidding strategy.
  • Google Search Ads, Search campaigns are where most bidding strategy decisions play out first. Understanding the auction mechanics makes bidding choices clearer.
  • Google Shopping and PMax, Performance Max campaigns use Target ROAS as their native bidding mode, making tROAS mastery essential before running PMax.
Test Your Knowledge
Loading questions…

Related Concepts

You Might Also Like