The New Google Ads Bidding Strategy After 17 August 2026

Google's tCPA And tROAS Change: Switch To Maximise Conversions

Short answer: if your campaign is limited by budget, move off Target CPA and Target ROAS and run Maximise Conversions or Maximise Conversion Value. Control spend with your budget, not with a target.

What changed

From 17 August 2026, Google changed how target-based bid strategies behave in budget-limited campaigns. Previously, a campaign could outperform its target. Now the system optimises consistently toward the target you set.

It affects Target CPA, Target ROAS, and Target CPC in Demand Gen, across Search, Shopping, Performance Max, Demand Gen and Travel.

Google will not adjust your targets for you.

Why a target now works against you

Your target used to be a direction. It is now a destination.

If your campaign was delivering leads at $80 against a $120 target, that gap was free performance. It is gone. The system will drift you toward $120 because that is what you asked for.

So you are left with two bad outcomes.

Set the target tight, and you throttle. Delivery drops, impression share falls, and the campaign sits on its hands rather than bidding into demand it could have won.

Set the target loose, and Google walks your costs up to meet it. You pay more for the same volume.

Neither one captures a good market. When demand spikes, a seasonal window opens, or a competitor drops out of the auction, a target-based campaign does not lean in. It holds its number.

Why Maximise Conversions wins for most businesses

Maximise Conversions has one instruction: get as many conversions as possible for the budget available.

That is usually what the business actually wants.

Most businesses make more money at scale, not at efficiency. A lower cost per lead is worthless if it comes with half the leads. Twenty customers at $180 acquisition beats eleven at $120 in almost every P&L I have looked at, because fixed costs do not care about your CPA.

Your budget is already a hard limit. Google respects daily and monthly caps regardless of bid strategy. So you already have a spend control. Adding a target on top of it is a second brake on a car that was not speeding.

Google's own guidance names this. In its FAQ on the change, it lists switching to Maximise Conversions or Maximise Conversion Value as the way to maximise return within a set budget.

Stop throttling. Scale.

What you need in place first

This only works if the machine is optimising toward the right thing.

Check your conversion tracking is clean. One primary conversion action, deduplicated, no soft actions inflating volume. If Maximise Conversions is chasing newsletter signups, it will get you plenty of them.

For lead generation, import offline conversions. Send closed deals and deal values back into Google. Then Maximise Conversion Value optimises toward revenue instead of form fills, and lead quality holds as volume rises.

Then set the budget at the number you are genuinely willing to spend, and let it run.

When a target still makes sense

Two cases.

You have a hard commercial ceiling, where a customer above a certain acquisition cost is unprofitable rather than merely expensive. Then set the target at that real ceiling, not at your best-ever month.

Or you are in a thin account with low conversion volume, where you need stability more than scale.

Everyone else: your target was a comfort blanket. It is now a cost.