Smart Bidding

Paid Media

Also: Automated Bidding · Google Smart Bidding

What it isGoogle's machine-learning bid automation

Optimises forConversions or conversion value

Needs timeWeeks of stable data to learn

Watch forGarbage conversion data in, garbage bids out

Quick definition

Smart Bidding is Google Ads' set of automated bidding strategies that use machine learning to set bids for every auction in real time. Instead of a human adjusting bids manually, the system predicts the likelihood and value of a conversion and bids accordingly, aiming for goals like target CPA or target ROAS.

How it varies across Australia

Adoption of Smart Bidding across Australian search accounts has climbed steadily as manual bidding options get harder to find in the Google Ads interface. Businesses with clean conversion tracking and enough conversion volume tend to see it outperform manual bidding within a couple of months. Businesses without that foundation often see CPA drift before it settles.

See acquisition performance benchmarks across Australian industries →

What it actually means

Smart Bidding is Google deciding what to bid on your behalf, auction by auction, using signals a human could never process fast enough. Device, location, time of day, browsing history, competing bids. All of it gets weighed in milliseconds toward whatever goal you set, usually target CPA (cost per acquisition) or target ROAS (return on ad spend).

The trade you're making is control for scale. You give up the ability to hand-tune a bid for a specific keyword. In exchange, you get a system optimising across thousands of signals per second, something no account manager can match.

The catch is that Smart Bidding is only as good as the conversion data it learns from. If your conversion tracking counts junk leads as real ones, or if your attribution setup double-counts sales, the algorithm optimises straight toward the wrong outcome with total confidence. It doesn't know the data is bad. It just does what it's told.

This is where Smart Bidding gets confused with a general concept of automation. It's a specific Google Ads feature, sitting inside a broader shift toward automated bidding across ad platforms, tied directly to your conversion rate and conversion event definitions.

Smart Bidding isn't a strategy. It's an engine. Feed it bad fuel and it will optimise beautifully toward the wrong outcome.

How it shows up

Smart Bidding shows up as one of several bid strategy options inside a Google Ads campaign: Target CPA, Target ROAS, Maximise Conversions, or Maximise Conversion Value. It also shows up in the learning phase, a period after launch or a major edit where performance is unstable while the algorithm gathers data. It shows up again whenever someone in a campaign review asks why CPA spiked for no obvious reason, usually the answer is a conversion tracking change or a budget shift that reset the learning phase.

The Australian context

Smaller Australian accounts often struggle to hit the conversion volume Smart Bidding needs to learn properly. Google recommends a minimum number of conversions per month before a strategy stabilises, and many local businesses with modest budgets sit below that threshold. In those cases, consolidating campaigns or switching to a value-based goal like Maximise Conversion Value can help the algorithm find a stable pattern faster than splitting spend across many small campaigns.

Where people get this wrong

Switching bid strategies too often.Every change resets or disrupts the learning phase. An account bouncing between target CPA and Maximise Conversions every fortnight never gives the algorithm enough stable data to actually optimise.

Setting a target CPA or ROAS based on wishful thinking.If the target is unrealistic given historical performance, Smart Bidding will either fail to spend the budget or chase the goal by narrowing reach so far that volume collapses.

Turning it on without fixing conversion tracking first.The algorithm optimises toward whatever you tell it is a conversion. Junk leads, bot traffic or duplicate conversion events all get treated as real signal, and bids drift toward attracting more of the same.

Smart Bidding vs Manual Bidding

	Smart Bidding	Manual Bidding
Who sets the bid	Google's algorithm, per auction	A human, per keyword or ad group
Speed of adjustment	Real time, thousands of signals	As fast as a person can review and edit
Data dependency	High, needs clean conversion volume	Low, works with limited data
Best fit	Accounts with reliable conversion tracking and volume	New accounts or highly niche, low-volume campaigns

Common questions

How long does Smart Bidding take to learn?

Google's own guidance suggests one to two weeks of stable settings before performance normalises, though accounts with low conversion volume can take longer. Avoid major budget or targeting changes during this window or the learning phase effectively restarts.

Is Smart Bidding always better than manual bidding?

No. It tends to outperform manual bidding once an account has enough conversion volume and clean tracking. Small accounts with thin conversion data sometimes get more predictable results from manual or semi-automated strategies until volume builds.

What's the difference between Target CPA and Target ROAS?

Target CPA optimises for a set cost per conversion, useful when every conversion is worth roughly the same. Target ROAS optimises for revenue relative to spend, better when order values vary significantly, like in ecommerce with mixed basket sizes.

Why did my CPA spike right after switching to Smart Bidding?

This is usually the learning phase. Performance is often volatile in the first one to two weeks as the algorithm tests bid levels. If it doesn't stabilise after that, the target is likely unrealistic or conversion tracking has a problem.

Debrief

Get the next one

No spam. No fluff. Just the next article, straight to your inbox.

Keep exploring

About New Rebellion

New Rebellion is a marketing intelligence consultancy. We build tools, score Australian businesses on how their marketing actually performs, and publish Debrief every day. This dictionary is part of how we work in the open.

How we think →