Measurement firm Measured published an analysis on August 11, 2026 arguing that the coming Google Ads bidding change converts a setting many advertisers treated as a ceiling into a direct instruction, six days before the rollout begins.

The document, written by Jarah Burke, VP, Client Partner at Measured, takes the mechanics of the August 17 change as settled and moves the argument somewhere else: to whether the numbers currently sitting in Target CPA and Target ROAS fields were ever meant to describe the economics of the business paying for the media.

According to the analysis, the change alters what a target means rather than only how it performs. Burke writes that it changes what the target itself means, how it needs to be measured and managed, and potentially what placements and audiences see an advertiser's ads. That framing separates the piece from most of the commentary circulating since June, which has concentrated on arithmetic and deadlines.

The setting that was never an instruction

The mechanics are not in dispute. On August 17, 2026, Google changes how budget-constrained campaigns using Target CPA and Target ROAS behave. Campaigns carrying a Limited by budget status while running those strategies have historically been permitted to deliver well inside their stated targets, sometimes for months.

Measured uses the same worked example that appears in Google's own documentation. A campaign with a $10 Target CPA might consistently deliver a $5 actual CPA. According to the analysis, that gap will no longer simply remain as upside for the advertiser, because the system will begin optimizing more consistently toward the stated target. The campaign could begin moving closer to a $10 CPA unless the advertiser lowers the number.

Google states that affected campaigns converge toward their targets within one to two conversion cycles.

The sentence Measured builds its case around is short. "After August 17, 2026, Google will treat that number as an instruction," Burke writes.

That distinction carries weight because of how the field has been used in practice. For many advertisers, according to the analysis, Target CPA or Target ROAS has historically functioned more like a guardrail: a ceiling, an inherited setting, a deliberately loose constraint, or simply a number that has worked well enough operationally. None of those uses describe an economic threshold. All of them are about to be read as one.

Measured is explicit that inaction is itself a position. "Remember, the decision to adjust or the decision to do nothing are both decisions for change," the analysis states.

Why a $5 CPA is not automatically a $5 target

The most contested passage in the document concerns what advertisers enter when they do act.

According to Measured, the obvious move is to reset the target to recent delivery so that performance stays roughly where it is. The analysis treats that as an incomplete answer. A campaign achieving a $5 CPA does not automatically mean $5 becomes the new target, Burke writes, and a business that can profitably acquire customers at $8 may decide it is willing to trade some efficiency for more volume.

The reasoning runs through the limits of platform reporting. The CPA and ROAS figures Google reports are described in the analysis as useful operating metrics that cannot on their own establish whether an investment is creating incremental business value. What counts as unacceptable diminishing returns, according to the document, is answered by business economics rather than by campaign settings.

"The point is to make that decision deliberately," the analysis states.

That is a narrower claim than it first appears. Measured is not arguing that advertisers will pay more. It is arguing that the August 17 change forces a decision that many accounts have deferred, and that the Bid Target Adjustment Tool, available inside Google Ads accounts since July 6, 2026, presents that decision in the shape of an administrative task.

Conversions counted against conversions caused

The second argument concerns what the additional volume represents.

Google's stated rationale is that bidding more consistently toward the target allows campaigns to capture additional conversions or conversion value. Measured does not contest that outcome. It contests the inference.

"Platform-attributed conversions and incremental business impact are not the same thing," the analysis states.

If Google reports more conversions as bidding shifts, according to the document, advertisers still need to determine how much of that added volume represents additional business, whether incremental revenue grows with spend, whether customer quality changes, and whether the marginal return remains worthwhile. The analysis notes that algorithms execute instructions literally: a system told to acquire conversions will acquire them, regardless of whether those conversions are incremental.

Then comes the structural point. "That distinction becomes especially important when the platform selling the media is also the system deciding which additional opportunities are worth buying," Burke writes.

That sentence describes a conflict that predates the August 17 change by many years, and it is the reason a measurement vendor has a commercial interest in this particular update. Measured sells incrementality testing, media mix modeling and cross-channel attribution to enterprise brands, and the analysis closes by inviting readers to schedule time with the company. The argument and the sales motion point in the same direction, which is worth stating plainly even where the argument holds.

The company is not new to the trade record on this point. In June 2026, Measured launched a Model Context Protocol server allowing brands to query incrementality data through AI interfaces, drawing on more than 30,000 incrementality tests across over 200 clients, with CEO Trevor Testwuide framing the design around where strategic media decisions are beginning to happen.

The part that does not show up in a ROAS column

The most technically specific section of the analysis concerns inventory rather than price.

For Performance Max and Demand Gen campaigns, Google has said advertisers may see shifts in how traffic is distributed across placements as bidding behavior changes. Measured reads that as a reporting problem. A relatively stable campaign-level ROAS could conceal meaningful movement underneath it, according to the analysis, with additional spend flowing into different inventory, audiences or sales channels, including the split between retail and ecommerce, carrying very different incremental performance or customer quality.

That reading aligns with what Google itself set out on August 5, 2026, when Ads Product Liaison Ginny Marvin published a question-and-answer summary stating that inaction may mean entering different auctions than a campaign previously competed in in order to reach the stated target. A campaign converting well inside its target has been buying a narrower and cheaper slice of the auction. Pulling delivery toward a looser figure means bidding into queries and placements the campaign previously skipped.

Two documents, one from the platform and one from a measurement vendor, arrive at the same mechanical description and different conclusions about what advertisers can learn from their own dashboards.

What Measured says it will track

The analysis sets out a review sequence for the days before the rollout and a monitoring plan for after it.

Before August 17, according to the document, the work involves identifying campaigns on target-based strategies that are currently budget constrained and outperforming their given targets, then interrogating why each target holds its current value: whether it represents a true business objective, an inherited setting, a deliberately loose ceiling, or a platform efficiency metric never reconciled with incremental return. Measured also recommends documenting current performance before any change, including spend, target, platform CPA or ROAS, conversion volume, inventory mix and incremental performance.

On timing, the analysis states that adjustments made intentionally rather than reflexively need time to settle, and that Google's one to two conversion cycle guidance may translate into several weeks for advertisers with longer conversion delays.

After August 17, according to Measured, the company will monitor not only Google-reported performance but the economics underneath it: spend, bids, inventory mix, platform ROAS and incremental metrics. That monitoring window overlaps with the one set out by Mike Ryan, Head of Ecommerce Insights at Smarter Ecommerce, who put the evaluation period at 30 to 60 days and said he expects to publish data between mid-September and mid-October 2026.

Why this matters for marketers

The August 17 change has been public since June 15, 2026, when Google disclosed it as part of a three-part package alongside Smart Bidding Exploration and promotion mode. Notification emails and updated Help Center documentation followed on July 2, 2026, and a Display and Video 360 announcement on July 14 extended the same date to Demand Gen line items using Target CPA, Target ROAS or Target CPC. The affected formats are Search, Shopping, Performance Max, Demand Gen and Travel campaigns carrying the Limited by budget status. Hotel and Display campaigns already operate under the new logic.

What has been missing from most of that record is a measurement argument. Practitioner criticism has run hot since July, when a post by freelance Google Ads consultant Joey Bidner gathered 71 reactions and 27 comments after he described the update as the most frustrating he had encountered, and Google rejected the claim that the change reaches unconstrained campaigns days later. Those exchanges concerned intent and scope. The Measured document concerns evidence.

The gap it identifies is measurable and already documented elsewhere in the sector. Meta's own research, published in a framework covered in 2026, found that at the median advertisers undervalue the platform by 31 percent when using rules-based attribution compared with incrementality measurement, a finding that runs in the opposite direction to Measured's caution but rests on the same premise: attributed conversions and caused conversions are different quantities. The IAB and IAB Europe formalised that separation in September 2025, defining incrementality as the causal impact of marketing compared with what would have occurred without it and distinguishing it explicitly from attribution and return-on-ad-spend arithmetic.

Access to the method has widened in parallel. Google reduced the minimum budget for its own incrementality experiments to $5,000 in November 2025, down from thresholds approaching $100,000, which places causal testing within reach of accounts that would previously have relied entirely on platform reporting. Reliability remains contested: a LiveRamp and MMA study published in 2026 modelled how identity errors can cut a measured return of $1.50 to $0.43, turning a profitable campaign into an apparent loss on paper.

Cost conditions add a second variable. Channable data published in July 2026 recorded a 46 percent year-on-year drop in ROAS as Google click costs rose, meaning the August 17 recalibration lands on auctions that were already more expensive than a year earlier. Separating platform-driven cost movement from auction-driven cost movement will be difficult in the weeks after the rollout, because both will be visible in the same reports at the same time.

Calendar pressure completes the picture. August 17 falls in the northern hemisphere holiday period and roughly ten weeks before Q4 peak trading, a collision Smarter Ecommerce flagged in feed adoption data published on July 27, 2026. If convergence takes one to two conversion cycles, and if evaluation takes 30 to 60 days, the first defensible read on what changed arrives close to the point at which seasonal budgets are already committed.

Measured closes with a line that concedes the uncertainty rather than resolving it: "In media buying, change is the only constant."

Timeline

Summary

Who: Measured, a marketing measurement firm serving enterprise brands, in an analysis written by Jarah Burke, VP, Client Partner. The subject is a Google Ads change affecting advertisers running budget-limited Target CPA and Target ROAS campaigns.

What: The analysis argues that the August 17 change converts the stated bid target from a guardrail into an instruction, that additional conversions reported by the platform are not automatically incremental business, and that placement and audience mix can move underneath a stable campaign-level ROAS figure.

When: Published August 11, 2026, six days before the change takes effect on August 17, 2026. The change was first disclosed on June 15, 2026, with notification emails on July 2 and the Bid Target Adjustment Tool on July 6.

Where: Google Ads, Search Ads 360, Display and Video 360, Google Ads Editor and the Google Ads API, covering Search, Shopping, Performance Max, Demand Gen and Travel campaigns carrying a Limited by budget status.

Why: Budget-limited campaigns delivering below their stated targets have accumulated efficiency the platform describes as a by-product of suppressed bids. Aligning delivery with stated targets removes that gap, which leaves advertisers to establish whether the conversions gained afterwards represent new business or a more expensive route to demand they were already capturing.