Your paid search team just lost a lever they've relied on for years. As of 2026, Google Ads no longer allows manual language targeting in Search campaigns. The setting is gone. Google's AI now decides which users understand which languages, using signals like browser settings, historical search behavior, and landing page content. If you run multilingual campaigns in EMEA, APAC, or North America's bilingual corridors, this change directly affects your CAC payback math.
The announcement first surfaced in Google's help documentation last August, with full removal scheduled by year-end 2025. The rollout completed earlier this year. Display, YouTube, and Performance Max campaigns retain their manual language controls. Search does not.
What Google's AI Actually Does Now
Google's documentation describes a "state-of-the-art system" that incorporates multiple signals: historic searches, search term language, user language settings, the language of ads previously shown, and landing page language. The system examines behavior patterns across weeks or months rather than relying on single-session indicators.
In practice, this means a user with a Spanish browser setting who frequently searches in English may see your English-language ads. A bilingual professional in Montreal might receive French or English ads depending on Google's inference of their comprehension. The algorithm prioritizes what it believes the user understands over what the advertiser explicitly selected.
Industry reaction has been mixed. Marjorie Vizethann, CEO of Alpine Analytix, noted that large brands have been setting their language targeting to "all" for over 15 years to capture English speakers in other countries. For those advertisers, the change is minimal. For teams that relied on precise language segmentation to control message relevance and budget allocation, the loss of control is significant.
The Real Risk: Diluted Relevance, Inflated Impressions
The concern isn't that AI can't detect language. It's that AI detection optimizes for reach, not for the specific audience your campaign was designed to convert.
Consider a B2B software company targeting German-speaking IT decision-makers in DACH markets. Previously, you could ensure your German-language ads appeared only to users whose settings indicated German comprehension. Now, Google's AI might serve those ads to a bilingual user in Switzerland who primarily searches in English, or to a German expat in the UK whose browser is set to German but who prefers English-language vendor communications.
The result: impressions increase, but message relevance may decline. Your German ad copy reaches users who would have converted better with English. Your English ads reach users who would have preferred German. The aggregate effect is harder to measure than a clean A/B test, but it shows up in conversion rate compression and rising cost-per-qualified-lead.
Research from Growleads found that 62% of advertisers believe Performance Max campaigns have made their overall ad performance worse. The language targeting removal extends similar automation logic to Search, the campaign type that typically drives the most direct-response revenue.
What Stays the Same
This change applies only to Search campaigns. Display, YouTube, and Shopping campaigns retain manual language targeting. If you run a multilingual strategy across campaign types, you now have inconsistent controls: full language selection in Display, none in Search.
Location targeting remains intact. You can still target Germany, France, or Quebec. But location and language are not the same thing. A user in Brussels might speak French, Dutch, or English. A user in Singapore might search in English, Mandarin, or Malay. Without language targeting, you're relying on Google's inference to match the right ad to the right user.
Adapting Without Losing the Forecast
The CFO question is straightforward: does this change our CAC payback assumptions? The honest answer is that it introduces variance you didn't have before. Here's how to manage it.
First, segment your Search campaigns by language at the creative level. If you previously ran a single campaign targeting English and German, split into separate campaigns with language-specific ad copy and landing pages. Google's AI uses landing page language as a signal. A German landing page increases the probability of serving to German-comprehending users.
Second, monitor search terms reports more aggressively. As Digital Ninjas noted, you may see more non-primary language queries entering your search terms report. Negative keyword management becomes critical. If you're seeing English queries on your German campaign, add them as negatives.
Third, use audience segmentation as a proxy. Custom audiences based on in-market signals, job titles, or company size can help narrow reach even when language targeting is unavailable. This doesn't replace language controls, but it reduces the probability of serving to irrelevant users.
Fourth, run a pre/post analysis. Compare conversion rates, cost-per-conversion, and qualified lead rates for the 90 days before and after the change took effect. If you see degradation, quantify it. That number belongs in your next pipeline review.
The Broader Pattern
This change fits Google's broader automation trajectory. AI Max for Search, Smart Bidding Exploration, and Performance Max all reflect the same philosophy: give the algorithm more latitude, reduce manual controls, and trust the system to optimize.
The Interactive Advertising Bureau found that more than 70% of marketers had experienced an AI-related incident in their advertising: off-brand content, unexpected placements, or copy that didn't reflect their messaging. The platforms aren't going away. The challenge has shifted from "Should I run ads?" to "Do I actually understand what my ads are doing?"
For B2B marketers, the stakes are higher. Longer sales cycles mean a misaligned impression today might not show up as a lost deal for six months. By then, the attribution trail is cold.
The Pilot Plan
If you haven't already, run this diagnostic over the next two weeks:
- Pull search terms reports for all multilingual Search campaigns. Flag queries in unexpected languages.
- Compare conversion rates by inferred language (use landing page language as a proxy) for the past 90 days versus the prior period.
- Identify campaigns where language-specific creative is not matched to language-specific landing pages. Fix those first.
- Document any CAC payback variance and bring it to your next pipeline review with a hypothesis on cause.
The language targeting removal is not a crisis. It's a control you no longer have. The question is whether your measurement and creative infrastructure can compensate. If it can, the change is noise. If it can't, you're flying blind in markets where precision used to be possible.
