Yesterday, Google's August 17 Smart Bidding update went live. Budget-limited campaigns using Target CPA or Target ROAS will now optimize toward the target you actually set, not the lower CPA or higher ROAS the algorithm quietly discovered was achievable. For accounts that have been coasting on overperformance, the math just changed.
The industry reaction has been predictably loud. Some practitioners are calling it a revenue grab; others are shrugging it off as a non-event. Ginny Marvin, Google's Ads Product Liaison, has been threading the needle between both camps, and her framing offers a useful lens for executives trying to separate signal from noise.
The Intention Gap
In a recent conversation with Cypress North, Marvin explained that the update addresses a real problem: when advertisers increased budget on campaigns that were quietly outperforming their targets, results would fluctuate unpredictably. The new behavior ties delivery directly to the number you entered, making budget scaling more consistent.
That sounds reasonable until you consider how many accounts got here. Plenty of advertisers set conservative targets years ago, watched performance improve, and never updated the number. Others deliberately used a high Target CPA as a ceiling, knowing the algorithm would beat it. Both groups now face the same question: was that target intentional, or did it simply drift out of date?
Google shipped a Bid Target Adjustment Tool on July 6 to help advertisers audit affected campaigns. The tool surfaces historical performance and offers three options: keep the target, match it to recent performance, or set a custom target. If your Target CPA is $10 but your actual CPA has been $5, you can update to $5 and maintain current performance. If you do nothing, the system will now work harder to hit that $10 target, which means fewer conversions at the same budget.
The update applies to Search, Shopping, Performance Max, Demand Gen, and Travel campaigns. App, Video Reach, and Video View campaigns are excluded. Display and Hotel campaigns already operated this way.
AI Max and the Shift to Intent Matching
The bidding change lands in the middle of a larger structural shift. AI Max for Search campaigns, which moved out of beta in April, represents Google's clearest signal yet that keyword syntax is giving way to intent matching. The feature uses Gemini to match ads based on inferred user intent rather than keyword permutations, bringing Performance Max-style AI directly into standard Search campaigns.
Marvin has been direct about what this means for eligibility in AI-powered search surfaces. In a PPC Chat community Q&A, she explained that ads appearing in AI Overviews and AI Mode require AI-powered targeting solutions: broad match on Search, or the keywordless targeting available through AI Max, Performance Max, Shopping, or Dynamic Search Ads. Smart Bidding is also required. The relevance bar is higher in AI Search, with ads matched to Google's understanding of user intent based on both the query and the content in the response.
Google's internal data shows that advertisers activating AI Max in Search campaigns typically see 14% more conversions or conversion value at a similar CPA/ROAS. For campaigns still mostly using exact and phrase keywords, the typical uplift is 27%. Those numbers are compelling, but they come with a caveat: AI Max won't be effective if campaigns are limited by budget. If a campaign is budget-constrained when AI Max is enabled, an alert will appear.
The Fundamentals That Still Matter
Here's where Marvin's guidance gets interesting for executives. Despite all the AI-powered automation, she keeps returning to the same fundamentals: conversion tracking accuracy, first-party data quality, and target-setting discipline.
In an interview with Optmyzr, Marvin emphasized that even with all the tech changes, core marketing principles still matter. Higher ad rank generally leads to lower CPCs. Investing in relevance and quality remains the best long-term strategy for cost efficiency. The system is still fundamentally designed to reward quality with better economics.
The August bidding update reinforces this point. If your conversion tracking is noisy, the algorithm will optimize toward a noisy signal. If your targets don't reflect your actual business goals, the system will now hit those misaligned targets more precisely. Automation amplifies whatever you feed it.
On the Ads Decoded podcast, Marvin and Brandon Ervin, Director of Product Management for Search Ads, discussed why hyper-granular campaign structures that were rational five years ago may now be a hidden cost. The shift is from manual bids with single keyword ad groups to Smart Bidding with high-quality conversion data and accurate targets. Control still exists; it just looks different. Brand controls, geo controls at the ad group level, and negative keyword lists replace the old levers.
Journey-Aware Bidding and the Lead Gen Problem
For B2B marketers with complex sales cycles, the most relevant announcement from Google Marketing Live 2026 may be journey-aware bidding, currently in beta for Target CPA Search campaigns. The feature improves prediction quality by learning from all of your conversion goals, not just those you're bidding to. If you track phone calls, form submissions, newsletter signups, and downstream sales, the system can see the whole picture and optimize accordingly.
This matters because lead gen has always been the awkward fit for automated bidding. A form fill is not a sale. A qualified lead is not a closed deal. The gap between what you can track in-platform and what actually drives revenue has forced many B2B advertisers to either accept noisy optimization or build elaborate offline conversion import workflows.
Journey-aware bidding doesn't eliminate that gap, but it narrows it. The system can now factor in leading indicators without requiring you to bid directly to them. Whether that translates to better pipeline quality depends entirely on the accuracy of your conversion data and the clarity of your lead-to-sale mapping.
What This Means for Your Next Pipeline Review
The August bidding update is not a crisis. It's a forcing function for hygiene you should have been doing anyway. If your targets reflect your actual business goals and your conversion tracking is accurate, the change is neutral to positive. If your targets drifted out of date or your tracking is noisy, you have work to do before the algorithm starts optimizing more precisely toward the wrong number.
The broader shift toward AI-powered intent matching is harder to ignore. Marvin has been clear that DSAs will migrate to AI Max for Search in . The direction is set. The question is whether you're building the data foundation and campaign structure to benefit from it, or whether you're still running 2019 playbooks on 2026 infrastructure.
For the next board prep, the conversation isn't about whether to adopt Smart Bidding or AI Max. It's about whether your conversion data is clean enough to trust the optimization, whether your targets reflect your actual CAC payback requirements, and whether your team has the discipline to audit and update those targets as performance changes. The AI is getting smarter. The fundamentals haven't moved.
