Reva Minkoff, Founder and President of Digital4Startups Inc., led our latest SMX Now webinar, examining Google’s changes to target bidding — and why paid search marketers shouldn’t panic.
Drawing on nearly two decades of PPC experience, Minkoff argued that while the update represents a meaningful shift in how advertisers should manage Target CPA and Target ROAS, the industry has effectively been here before. Google’s target bidding strategies operated in a remarkably similar way around a decade ago.
What changed with target bidding?
Previously, target bid strategies could function more like efficiency safeguards. If a campaign could outperform its Target CPA or Target ROAS, Google could continue delivering that better performance.
That meant a campaign with a $10 Target CPA might consistently generate conversions for $5.
Under the new approach, the target is more literally a performance target. If Target CPA is set at $10, Google will seek conversions around that CPA rather than necessarily trying to beat it.
The upside is greater predictability. Advertisers may find it easier to forecast what happens when budgets increase because Google is attempting to maintain performance around a defined efficiency level.
The downside is that campaigns which historically outperformed their targets may see that advantage shrink.
We’ve seen this before
The important context is that Target CPA originally behaved much like this.
Around 2015 and 2016, Google described Target CPA as a strategy designed to set bids so that the average cost per conversion would equal the advertiser’s chosen target. Some conversions could cost more and others less, but the system aimed for the target on average.
A decade later, much of that logic has returned.
The advertising ecosystem surrounding it may look dramatically different — with Performance Max, AI Max, Demand Gen and other technologies emerging since then — but advertisers have experience managing this kind of bidding environment.
Start by deciding whether you want volume or efficiency
Minkoff’s first recommendation is to establish what the campaign is actually supposed to accomplish.
If the priority is generating as much conversion volume as possible from a fixed budget, Maximize Conversions or Maximize Conversion Value may make more sense.
Target CPA and Target ROAS become more appropriate when efficiency is the constraint.
For example, a business might be prepared to spend aggressively as long as leads remain below a $50 CPA. An ecommerce advertiser might be willing to scale provided ROAS remains above an acceptable threshold.
The distinction matters because applying a target to a campaign whose real objective is maximum volume can unnecessarily restrict delivery.
Make your target reflect reality
Once efficiency is established as the objective, advertisers need to choose an appropriate starting target.
Rather than picking an arbitrary number, Minkoff recommends starting reasonably close to actual performance. If a campaign is currently generating conversions at a $30 CPA, for example, that provides a logical reference point for the initial Target CPA.
From there, the target becomes a lever for improving efficiency.
For completely new campaigns without enough historical information, advertisers don’t necessarily need to invent a target. Starting with a maximize strategy can provide the data needed to establish an appropriate target later.
Gradually push targets toward better performance
One of the most useful lessons from the earlier Target CPA era is that advertisers can progressively test the limits of Google’s bidding system.
If actual CPA is consistently meeting or beating the target, particularly when a campaign is limited by budget, Minkoff recommends considering a gradual reduction in the Target CPA.
That could mean reducing it by around 10% to 20%, allowing the campaign to run for one or two conversion cycles and then evaluating the results. If performance remains healthy, advertisers can repeat the process.
Minkoff has seen the approach produce significant results before. One transportation industry client achieved a 75% reduction in CPA over two weeks as its target was progressively lowered from $10 to $7.50 and then $5.
A B2B financial services client followed a similar pattern. As Google continued delivering actual CPAs around the target, the team lowered that target further and continued improving efficiency.
Don’t change targets too quickly
Progressive adjustment does not mean constantly tinkering with campaigns.
Advertisers need enough data to establish whether the bidding system is genuinely meeting its target. Depending on campaign volume and conversion-cycle length, that evaluation might happen weekly, biweekly or monthly.
Changing targets before conversions have had enough time to mature risks making decisions using incomplete performance data.
The better approach is to allow the campaign to settle, compare actual CPA or ROAS with the target and then decide whether another adjustment is justified.
When target bidding stops working, move down the bidding ladder
Advertisers also aren’t locked into a single bidding strategy.
If Target CPA or Target ROAS stops producing conversions, Minkoff recommends first checking the fundamentals, including conversion tracking, landing pages and search queries.
If those appear healthy, removing the target and moving to Maximize Conversions can help determine whether the target itself is restricting the algorithm.
If Maximize Conversions still cannot generate sufficient activity, Maximize Clicks can be used to build traffic and data before working back toward conversion-focused bidding.
The broader principle is that advertisers can move up and down the bidding hierarchy as campaign conditions change.
Your bidding is only as good as your conversion data
None of these strategies will work particularly well if Google is optimizing toward the wrong signals.
That makes conversion tracking — and especially conversion quality — fundamental.
A store visit, for example, isn’t necessarily equivalent to a purchase. Likewise, a cheap lead isn’t valuable if it turns out to be spam or has virtually no chance of becoming a customer.
If poor-quality leads are reported to Google as successful conversions, the bidding system can rationally pursue more of them. From the algorithm’s perspective, it is doing what it was instructed to do.
Advertisers therefore need to ensure primary conversions represent genuine business outcomes and, particularly for lead generation, feed meaningful quality signals back into the system.
Separate campaigns when their economics are different
Target bidding also makes campaign structure more important.
Brand and non-brand traffic, for example, frequently have different economics. Brand conversions are typically cheaper, while non-brand activity can be more competitive.
Combining them can make it harder to establish an appropriate efficiency target for each type of traffic.
Minkoff applies similar thinking to new customer acquisition campaigns. When new customers have different lifetime values or justify a different acquisition cost, separating those campaigns makes it easier to assign targets and report on performance according to their specific purpose.
Watch more than CPA and ROAS
CPA and ROAS remain central, but advertisers should monitor the surrounding signals as targets change.
Search impression share and search impression share lost to budget can reveal whether campaigns are being constrained. Impression volume is also worth monitoring because aggressive efficiency targets may cause Google to reduce delivery when it doesn’t believe the target can be achieved.
CPCs could rise as well if Google enters more expensive auctions in pursuit of conversions around the specified target.
Whether those changes are problematic depends on the campaign’s objective. A decline in volume may be acceptable when maintaining a strict CPA is the priority. It is far less desirable when maximum conversion volume is the real goal.
The target bidding “apocalypse” is really a strategy reset
Google’s target bidding change requires advertisers to become more deliberate about the relationship between campaign goals, bidding strategies and performance targets.
The playbook Minkoff outlined is relatively simple: determine whether volume or efficiency matters most, choose the bidding strategy that reflects that goal, establish a realistic target and adjust it progressively as performance data comes in.
Most importantly, this isn’t an entirely new challenge for PPC marketers.
The technology surrounding paid search may have changed dramatically over the past decade, but the fundamental job remains familiar: give the bidding system the right objective, provide it with high-quality conversion data and keep testing how far performance can improve.
Search Engine Land is owned by Semrush. We remain committed to providing high-quality coverage of marketing topics. Unless otherwise noted, this page’s content was written by either an employee or a paid contractor of Semrush Inc.