Bid modifiers in traditional Google Ads let you adjust bids by +/- percentage for specific conditions: devices, locations, audiences, time of day, day of week. In AI Max campaigns, most bid modifiers are either unavailable or have reduced effect. This is one of the most significant control changes in AI Max.
Why Bid Modifiers Changed in AI Max
Traditional bid modifiers were a workaround for the limitations of manual and rule-based bidding. The advertiser had to manually specify: "mobile users convert at lower rates, so bid 20% less for mobile." The advertiser was doing what the AI now does automatically.
AI Max's bidding model evaluates each impression-level signal in real time: device, location, time, audience, query, landing page, and more. The AI calculates a predicted conversion probability for each combination of these signals and bids accordingly. This is strictly more powerful than manual bid modifiers — the AI can account for hundreds of signal combinations simultaneously, not just the dozen or so manual modifiers an advertiser might set.
The tradeoff: you can no longer directly specify "bid X% more for this condition."
What Happened to Each Modifier Type
Device bid modifiers: not applicable in AI Max for Search or AI Max for Shopping. The AI determines device-level bidding based on its conversion probability model. You can still see device performance segmented in reports, but you cannot apply a manual device bid adjustment.
Location bid modifiers: not applicable in AI Max campaigns. The AI handles location-based bid optimization. You can include or exclude locations (hard constraint), but not apply percentage bid adjustments by location.
Audience bid modifiers: replaced by audience signals. Instead of applying +20% for a remarketing list, you add the remarketing list as an audience signal. The AI uses the signal to bid more for users it predicts will convert — which implicitly gives the effect of a positive bid adjustment without you specifying the percentage.
Ad schedule bid modifiers: mostly not applicable in AI Max. The AI determines time-of-day bidding based on predicted conversion rates by time. You can still use ad scheduling to pause ads entirely during certain hours (e.g., "no ads when our business is closed") — but not to apply percentage adjustments.
Device exclusions (not modifiers): you can exclude specific device types entirely. "-100% for mobile" (effectively excluding mobile) is still supported. Full device exclusions are respected as constraints, not signals.
The Replacement: Conversion Value Rules and Audience Signals
The two AI-compatible replacements for bid modifiers:
Conversion value rules communicate value differences to the AI without specifying bid percentages. "Users from this audience are worth 1.5x" tells the AI that winning auctions for this audience generates higher value — implicitly causing the AI to bid more for them.
Context from the July 2026 ToS (https://yositeup.com/blog/google-ads-tos-july-2026-ai-automation-what-changed): value rules are categorized as advertiser-provided data inputs, not bid override instructions. The AI uses them to calibrate its value prediction, then determines the bid.
Audience signals tell the AI which users are high-value. Adding a Customer Match list of your best customers as a signal shapes the AI's prediction of which other users might also be high-value. The AI extrapolates from the signal, effectively bidding more for similar users.
The Data Deletion Impact on AI Bid Decisions
The June 2026 data deletion (https://yositeup.com/blog/google-ads-reporting-data-deleted-june-2026) removed signal data that informed the AI's device, location, and time-of-day bid differentiation. The AI's model for optimizing across these dimensions needed to rebuild from current data.
If you notice the AI not differentiating between device types, times of day, or locations as efficiently as traditional campaigns did (lower overall ROAS), this may reflect the learning phase post-deletion. Adding fresh audience signals and conversion value rules can accelerate the AI's re-calibration.
DSA Migration: Losing Audience Bid Adjustments
DSA campaigns supported audience bid adjustments in Observation mode. These adjustments (+15% for remarketing list A, -10% for low-value segment B) don't carry over to AI Max for Search (https://yositeup.com/blog/google-ads-dsa-ai-max-migration-february-2027).
Post-migration replacement strategy:
- Previous +% audience adjustment → add the audience as a signal in AI Max for Search
- Previous -% audience adjustment → consider whether to add as a negative audience (exclusion) or simply rely on the AI's conversion probability model to de-prioritize these users
For most cases, letting the AI optimize without manual negative audience adjustments is correct. The AI will naturally bid less for audiences with lower predicted conversion rates.
AI Max for Shopping and Bid Modifiers
For AI Max for Shopping (https://yositeup.com/blog/google-ai-max-shopping-replacing-performance-max-2026): the same principle applies. No device, location, or audience bid modifiers. Shopping campaign bid optimization is fully AI-controlled.
The exception: target ROAS is set at the campaign level and can be adjusted. Tightening ROAS target (increasing it) means the AI must achieve higher efficiency before bidding — which implicitly reduces reach. This is the macro-level control that replaces campaign-level bid adjustments.
Top comments (0)