How to Prepare for Google Ads’ August 17 Smart Bidding Update: Advice From PPC Experts

Anu Adegbola
14 hours ago
12 min read

Updated: 56 minutes ago

Google’s August 17 Smart Bidding update could change how advertisers think about Target CPA (tCPA) and Target ROAS (tROAS), particularly for campaigns that are limited by budget.

Under the change, Google is expected to optimise budget-constrained campaigns more closely toward the targets advertisers have actually set. That creates an important consideration for accounts where campaigns have historically performed significantly better than their stated targets.

If a campaign has a Target CPA of £30 but regularly generates conversions for £20, for example, advertisers may need to decide whether £30 really represents what they are willing to pay — or whether the target should be brought closer to actual performance.

PPC experts have offered a range of recommendations for preparing for the change. While their approaches differ, several clear themes emerge.

The update has already sparked a heated debate within the PPC community, with practitioners questioning whether Google’s move towards more predictable target-based bidding could reduce advertisers’ ability to use Smart Bidding for exploration and incremental growth, as explored in PPC Live’s recent coverage of the backlash.

TLDR

Google’s August 17 Smart Bidding update means budget-limited tCPA and tROAS campaigns will optimise more closely to their set targets.

PPC experts recommend auditing targets against actual performance, avoiding blanket changes based solely on the last 28–30 days, making gradual adjustments, and monitoring results closely after rollout.

The bigger strategic question is whether stricter targets could limit Smart Bidding’s ability to explore and make it harder to identify opportunities for profitable growth.

Audit Budget-Limited tCPA and tROAS Campaigns First

One of the most consistent recommendations is to identify which campaigns have the greatest exposure to the update before making changes.

Veronika Holler recommends starting with campaigns using Target CPA or Target ROAS that are already limited by budget. Rather than making account-wide changes unnecessarily, advertisers should concentrate first on campaigns where the new bidding behaviour could have the greatest impact.

Ana Kostic offers similar advice, recommending that advertisers review every campaign that is both limited by budget and using tCPA or tROAS. She suggests exporting 30–90 days of data and identifying gaps between the configured target and actual CPA or ROAS.

Boris Beceric also recommends beginning with budget-limited tCPA and tROAS campaigns, while Inderpail Rai advises checking bidding strategies campaign by campaign and comparing targets against both 30-day actual performance and the more recent 14-day trend.

Heidi Sturrock similarly recommends a quick audit of budget-capped tCPA and tROAS campaigns rather than panicking about the change.

However, some experts suggest looking more broadly. Susie Marino recommends checking CPA and ROAS targets across all campaigns, including those that are not currently limited by budget. A campaign that is unaffected today could become budget constrained later, so the update provides an opportunity to make sure targets throughout the account remain sensible.

Heather Robinson takes a similarly broad approach, recommending a review of all active campaigns — as well as campaigns likely to run soon — to identify those using Smart Bidding targets and determine which are consistently outperforming them.

Compare Your Targets With Actual Performance

Once potentially affected campaigns have been identified, the next step is understanding whether the targets entered into Google Ads still reflect reality.

This is perhaps the strongest point of agreement among the experts.

Melissa Mackey says advertisers should compare targets with actual performance and begin adjusting campaigns that are significantly outperforming their targets.

Veronika Holler recommends basing those decisions on meaningful historical performance rather than assumptions. If a campaign has a $20 Target CPA but has consistently generated conversions for $10, the configured target may no longer accurately describe the desired outcome.

Gianpaolo Lorusso similarly recommends progressively moving tCPA and tROAS targets toward actual performance, while Kat Sale advises ensuring bids have been reviewed and are close to where advertisers actually want them before the change.

Dez Calton says advertisers who already monitor targets and budgets regularly may not need to do anything drastic. For campaigns frequently limited by budget, however, the target should make sense both commercially and in relation to the CPA or ROAS the campaign is currently achieving.

Susan Yen frames the issue as one of intent. A Target CPA should represent what the advertiser is genuinely willing to pay. If it does not, the algorithm is effectively being given the wrong objective.

Tim Searle notes that Google can recommend a tROAS based on recent performance directly within campaign settings, and he has seen these recommendations work in cases where previously higher tROAS targets had restricted delivery to the point that campaigns stopped serving.

However, his bigger question is what happens when a campaign’s potential improves beyond its existing target. If conversion rate increases over time, for example, a tROAS that was previously appropriate could become lower than what the campaign is capable of delivering. If Smart Bidding no longer seeks to substantially outperform that target, advertisers need to watch for situations where tROAS could effectively constrain potential performance.

Is this update - Smart bidding a good thing in general

Make Target Changes Gradually

Bringing targets closer to actual performance does not necessarily mean making one large adjustment immediately.

Several experts recommend incremental changes.

Heidi Sturrock suggests using roughly 30–45 days of historical performance and, where further adjustments are required to meet business goals, changing targets by around 10–15% per conversion cycle.

Selina Patel has been taking a similar approach, adjusting targets by approximately 10–20% each week rather than making significant changes all at once.

Ana Kostic recommends controlled adjustments of roughly 10–15%, avoiding jumps above 20%, and then waiting one or two conversion cycles before assessing the result.

Reva Minkoff also advocates gradual optimisation where efficiency is the priority. Once Target CPA reflects actual CPA, advertisers can test reducing it by around 10–20% at a time in an effort to improve efficiency.

For B2B advertisers, Dii Pooler recommends even greater caution. Accounts using Offline Conversion Imports and dealing with multi-week sales cycles may have incomplete recent performance data because of conversion lag. Rather than making a sudden tCPA cut, Pooler recommends staging smaller 5–10% adjustments over six to eight weeks so that CRM and offline conversion data have time to mature.

Don’t Blindly Use the Last 28 or 30 Days

Recent performance is an obvious reference point, but several experts warn against treating it as universally representative.

Samantha Noble points out that the appropriate performance window depends heavily on the client and industry. Some businesses have conversion cycles longer than 30 days, while seasonality can make one period substantially different from another.

Susan Richards-Benson recommends looking over a sufficiently long period — potentially including prior-year data — to understand actual CPA and ROAS and identify what is genuinely driving performance.

Tiffany Shears also stresses seasonality. Advertisers may need higher targets during quieter months to maintain volume and could potentially reduce targets during stronger periods when conversions naturally become cheaper.

Dii Pooler’s warning is particularly relevant for B2B accounts: a recent 30-day CPA may appear artificially strong or weak simply because later-stage offline conversions have not yet been imported. Advertisers should therefore account for conversion lag before deciding that recent actuals represent the right future target.

Decide Whether Volume or Efficiency Matters More

The August 17 update also creates a more fundamental strategic question: what is the campaign actually supposed to optimise for?

Reva Minkoff recommends explicitly deciding whether volume or efficiency is the priority. If volume matters most, Maximize Conversions may make more sense. If efficiency is the priority, advertisers should ensure Target CPA reflects actual CPA and then continue optimising that target over time.

Boris Beceric makes a similar distinction based on business constraints. If the budget is fixed, he recommends considering the removal of the target and using Maximize Conversions or Maximize Conversion Value. If efficiency is fixed, advertisers can retain tCPA or tROAS, but the target should reflect the genuine business objective and the campaign should have sufficient budget to pursue it.

His key warning is not to change targets simply because Google recommends doing so. The first question should be which constraint actually matters to the business.

Benjamin Wenner also argues for grounding targets in business economics rather than simply copying performance produced by a budget cap. He recommends resetting the target around the real business margin and then determining whether additional conversion volume actually exists at that level before committing more budget.

Consider Raising Budgets or Moving to Maximize Strategies

Adjusting the target is not the only possible response.

Aaron Levy sees several options. Advertisers could retain the existing target, move it closer to recent actual performance, or — where continued growth is the objective — raise the budget or move to a Maximize strategy. For campaigns that should continue scaling, Levy argues that lifting the budget constraint may be a better lever than changing the efficiency target.

Veronika Holler also recommends questioning budget caps. If a campaign is substantially outperforming its target, there may be a greater opportunity in increasing investment rather than simply tightening the target. Where the budget itself is a strict limit, Maximize Conversions or Maximize Conversion Value could provide Smart Bidding with more flexibility.

Boris Beceric likewise recommends Maximize strategies when the business has a genuinely fixed budget.

Not everyone is convinced that increasing budgets is the answer, however.

Jeremy Courty argues that many businesses — particularly SMEs — have fixed budgets for legitimate reasons. In those cases, raising the budget simply because Google recommends it may not be commercially realistic.

Will your client benefit from this update

Be Careful About Removing Smart Bidding’s Exploration Headroom

One of the more nuanced concerns surrounding the update is whether tighter alignment with targets could restrict campaigns from discovering additional opportunities.

Jeremy Courty highlights this potential trade-off. Leaving a low tROAS target unchanged could cause actual ROAS to fall as Google optimises more closely toward it. But raising the tROAS target to match previous overperformance could potentially make the campaign more conservative, concentrating spend on safer queries or remarketing rather than exploring a wider range of auctions.

Dii Pooler raises a related concern for B2B advertisers. Cutting Target CPA too aggressively before delayed offline conversion data arrives could unintentionally reduce the algorithm’s exploration headroom and suppress mid-funnel lead volume.

These perspectives suggest advertisers should consider not only whether a target reflects past performance, but also what changing it might do to the campaign’s ability to discover additional demand.

Monitor Performance Closely After August 17

Preparation does not end when the update goes live.

Melissa Mackey recommends establishing regular reporting now so performance can be reviewed weekly, monthly or at another cadence appropriate to the account.

Kat Sale similarly recommends setting aside time to monitor bids after the change.

Heather Robinson suggests that campaigns whose targets are already relatively close to actual performance may not require immediate intervention. In those cases, monitoring first and acting only where necessary may be the better approach.

Adekojede Muyiwa also advises advertisers to benchmark results before the update, observe what actually changes after August 17 and make decisions based on evidence rather than assumptions surrounding the announcement.

Ana Kostic reinforces this test-and-learn approach: plan changes, make controlled adjustments, monitor the outcome and adjust again based on the data.

Richard Russell and his team have been reviewing the accounts they manage and bringing bidding targets as close to actual performance as possible. While he expects there may be some initial “hiccups” after the change, he does not expect them to be problems that cannot be resolved through ongoing account management.

Russell also highlights an important step that goes beyond campaign optimisation: communicating the change to clients before it happens. His team has been meeting with clients to explain the update in straightforward terms and, importantly, tell each client whether their campaigns are actually likely to be affected.

That communication can be particularly valuable when performance fluctuates after the rollout. Clients who already understand what is changing, why it matters and what their PPC team is monitoring are less likely to be caught off guard by short-term movements.

Look Beyond Campaign-Level Performance

Some experts expect the update to require more granular monitoring.

Samantha Noble recommends tracking performance not only at campaign level but at ad group level, particularly because different parts of an account can respond differently depending on conversion cycles and seasonality.

Tiffany Shears also recommends reviewing individual ad groups where there is enough conversion volume. Some ad groups may naturally convert much more cheaply than others, so applying the same assumptions across an entire campaign could hide important differences.

After the update, Shears recommends comparing actual CPA with the stated target each week. If CPA moves toward the target while conversion rate remains unchanged, it may indicate that previously available efficiency headroom is simply being spent. Conversely, if conversions and conversion rate improve, the additional spend may be producing genuine incremental value.

She also recommends monitoring conversion quality, since a reduction in raw conversion volume is not necessarily negative if the resulting leads or sales are more valuable.

Watch Impression Share, Ad Rank and Other Supporting Metrics

CPA and ROAS should not be viewed in isolation.

Susan Richards-Benson recommends experimenting with bid CPA or ROAS closer to actual performance and monitoring the effect not only on core efficiency metrics, but also impression share, ad rank and CTR.

Tiffany Shears similarly suggests progressively adjusting targets until indicators such as clicks, impressions or rank lost impression share begin moving in the wrong direction, at which point advertisers can hold the target and assess performance.

This provides a broader way to judge whether a target adjustment is improving efficiency or beginning to restrict the campaign’s ability to compete.

Strengthen Search Terms, Ad Copy and Fundamental Optimisation

The update may also encourage advertisers to look beyond bidding as their primary optimisation lever.

Mery Hayles recommends reviewing performance against targets and budgets, but also going deeper into search term reports to identify wasted spend and missed high-quality opportunities.

She also recommends using ad copy optimisation to improve results more sustainably.

If advertisers have historically relied on campaign-level targets and budgets to control performance, a change that makes those targets more literal could increase the importance of bottom-up optimisation — including better search terms, stronger ads and more disciplined account fundamentals.

Build Scalable Auditing for Large Accounts

For advertisers and agencies managing large portfolios, manually reviewing every campaign may be impractical.

Jeremy Courty recommends using an LLM to help create an MCC-level script that pulls the relevant data at scale.

The resulting report should identify campaigns using tCPA or tROAS and include their configured target, actual CPA or ROAS, the percentage difference between target and actual performance, and whether the campaign is limited by budget.

That makes it easier to isolate the campaigns that combine all three risk factors: target-based bidding, budget limitations and significant overperformance against the configured target.

Treat Your Target as a Real Business Instruction

Perhaps the most important lesson from the update is that advertisers may need to become much more deliberate about the targets they enter into Google Ads.

As Yen argues, the biggest preparation may be a change in mindset rather than simply changing bids. If Smart Bidding is going to treat targets more literally, advertisers need to make sure those targets represent genuine business objectives.

That point is echoed by Boris Beceric, who recommends deciding whether budget or efficiency is the true constraint before changing anything, and by Benjamin Wenner, who argues that targets should ultimately reflect real business margins rather than numbers created by previous budget constraints.

Höller similarly suggests that the bigger challenge is identifying which campaigns still have room to grow and which are being restricted by outdated targets or artificial limits.

Expect Some Volatility — But Avoid Panic

Finally, several experts advise against overreacting.

Calton believes advertisers who already monitor their targets and budgets regularly should not need to make drastic changes.

Heather Robinson recommends being proactive but acknowledges that the full impact of the update is not yet known. Where targets already sit close to actual performance, observation may be preferable to unnecessary intervention.

Sturrock expects some short-term volatility as the algorithm adjusts, but argues that advertisers who align their targets with reality should stay the course rather than panic.

Levy similarly expects a potentially turbulent adjustment period but notes that advertisers have already experienced numerous changes to Smart Bidding over the years.

The common thread is that August 17 should not trigger indiscriminate account changes. It should trigger closer scrutiny of what advertisers are telling Google’s bidding algorithms to achieve.

Russell also expects some initial hiccups, but sees them as manageable rather than a reason for drastic intervention. His approach is to get bidding as close to reality as possible before the rollout and then deal with any issues that emerge based on actual performance.

The Bottom Line

The clearest advice from PPC practitioners is to know exactly what your targets say before Google starts taking them more literally.

Audit campaigns using tCPA and tROAS, particularly those limited by budget. Compare targets with actual performance, but account for seasonality, conversion delays and genuine business margins before making changes. Where adjustments are necessary, consider making them gradually rather than dramatically.

Just as importantly, decide what each campaign is actually supposed to achieve. If budget is fixed, a Maximize strategy may be more appropriate. If efficiency is the priority, the target needs to represent a genuine CPA or ROAS objective. And if the campaign has room to scale, increasing budget could be more valuable than simply tightening the target.

After August 17, close monitoring will be essential. Advertisers should watch not only CPA and ROAS but conversion volume, conversion quality, impression share, ad rank and the ability of campaigns to continue discovering new demand.

The update ultimately makes the number entered into the Target CPA or Target ROAS field more consequential. As a result, target setting may become less of a loose bidding signal and more of a strategic business decision.