Google Ads audience signals are inputs you give Google AI to help it understand which kinds of users are most likely to convert. In Performance Max, they are suggestions rather than hard targeting rules. Google can still serve ads beyond the audiences you provide if its models predict other users are likely to help achieve the campaign goal. For the latest experimentation controls, see our Performance Max asset-group A/B testing update.
That distinction matters. In 2026, advertisers need to separate three related concepts: Performance Max audience signals, audience segments used in campaigns such as Search and Display, and Lookalike segments used in Demand Gen and selected Video campaigns. Audience signals are only one input to review: our Performance Max sourced-video audit explains why creative source approvals need separate attention.
This guide explains Google Ads audience targeting, the main Google Ads targeting options, and the audience targeting strategies advertisers should use across Performance Max, Search and Demand Gen in 2026. It also covers how Observation and Targeting differ, how Smart Bidding uses first-party audiences, and how to evaluate the audiences Google discovers after launch.
What Are Google Ads Audience Signals?
Audience signals let you share information about the demographics, interests and behaviours of customers who are relevant to your business. In Performance Max, signals are typically applied at the asset-group level and help Google’s machine-learning systems find likely converters faster.
They are not a guarantee that Performance Max will show ads only to people inside those audiences. Google can expand beyond your signals when it predicts other users are likely to convert.
Google notes that new audience signals can take up to two weeks to be fully integrated and optimised by its machine-learning models.
Google Ads Audience Targeting: Signals vs Segments vs Lookalikes
| Concept | Where it is used | How it behaves |
|---|---|---|
| Audience signals | Performance Max, and selected AI-led campaign workflows | Suggestions that help Google AI identify likely converters; not strict targeting constraints |
| Audience segments | Search, Display, Video, Shopping and other campaign types | Groups based on interests, intent, demographics or first-party data; can be observed or targeted depending on campaign type |
| Lookalike segments | Demand Gen and supported Video brand campaigns | Built from first-party seed lists to find users with similar characteristics; Demand Gen is transitioning toward suggestion-style behaviour during 2026 |
This is why it is misleading to treat every Google Ads audience feature as the same form of targeting.
Performance Max Audience Signals Are Not Hard Targeting
Performance Max uses audience signals as a starting point for Google AI. If you know that high-value customers tend to come from a particular CRM list, website behaviour or custom segment, you can provide that signal. Google then combines it with other real-time auction signals to optimise toward your conversion goal.
The campaign may still reach users outside that audience. This is a fundamental difference between Performance Max and traditional audience targeting.
For a deeper look at how this affects channel control and Search overlap, see our guide to Performance Max vs Search in 2026 and our Performance Max for SaaS guide.
Which Audience Signals Should You Use in Performance Max?
The strongest audience signals usually come from data that reflects real customer behaviour rather than generic assumptions.
| Signal type | Best use |
|---|---|
| Customer Match | High-value customers, qualified leads, past purchasers or other CRM-based segments |
| Website visitors | Users who visited pricing, product, demo or other high-intent pages |
| App users | Known users or customers with meaningful in-app behaviour |
| YouTube viewers | Users who engaged with your videos or channel |
| Custom segments | Users associated with relevant search terms, URLs or apps |
| In-market segments | Users showing recent purchase intent in a category |
| Affinity segments | Broader interests and long-term habits, usually more useful for upper-funnel context |
| Detailed demographics | Long-term demographic traits where relevant and available |
For B2B SaaS, a strong starting signal is often a clean Customer Match list of qualified opportunities or customers, combined with custom segments tied to the category. Our Google Ads for SaaS guide covers how to connect targeting, campaign structure and budget to pipeline.
Customer Match: The Most Valuable First-Party Signal
Customer Match lets you use your first-party customer data, such as email addresses, phone numbers or mailing addresses, to create audience segments in Google Ads. It can support re-engagement, exclusions, retention and AI-led prospecting workflows.
Customer Match data also plays a role in Google’s evolving customer lifecycle goals for acquisition and retention. The October 2026 update adds more options for prioritizing new, returning and high-value customers, but an audience signal is not the same as a customer-value bidding goal.
- Existing customers: exclude them from acquisition campaigns where appropriate.
- Qualified leads: help Google identify patterns associated with stronger prospects.
- High-value customers: create more useful seed data than a broad list of every contact in your CRM.
- Lapsed users: build re-engagement strategies around known historical customers.
Data quality matters more than list size alone. A smaller list of real high-value customers can be more strategically useful than a large list mixing customers, low-intent leads and irrelevant contacts.
Custom Segments in Google Ads
Custom segments let you give Google additional context using relevant search terms, websites or apps. They are useful when the built-in in-market or affinity categories are too broad for your business.
For example, a B2B SaaS company could create a custom segment around category-defining search queries, competitor websites and relevant software-review destinations. That does not mean every user who visits those sites will be targeted exactly. The behaviour depends on campaign type and the way Google applies those audience inputs.
For competitor-led audience research, see our competitor audience targeting guide and the broader competitor ads analysis framework.
Demand Gen Lookalike Segments: What Changed in 2026?
Lookalike segments are available in Demand Gen and supported Video brand campaigns. They use a first-party seed list, such as Customer Match, website visitors, app users or YouTube engagement, to identify new users with similar characteristics.
During 2026, Google is transitioning Demand Gen Lookalike segments toward a suggestion model. Instead of being limited strictly by the original similarity threshold, the seed audience can act as a signal that allows Google AI to reach additional qualified users when doing so supports the campaign goal.
Google says the update began rolling out in March 2026, with opt-out controls planned later in the year. This makes Lookalikes increasingly similar in philosophy to other Google AI audience systems: the advertiser supplies useful first-party context, while Google expands when it predicts additional performance.
We cover this development separately in our Google Lookalike Segments update.
Performance Max vs Search: How Audiences Work Differently
| Feature | Performance Max | Search |
|---|---|---|
| Primary audience input | Audience signals | Audience segments |
| Hard targeting? | No, signals can be expanded beyond | Can restrict reach when Targeting is selected |
| Observation mode | Not the same concept | Yes |
| Keywords required | No | Generally yes, except specific formats such as Dynamic Search Ads |
| AI expansion beyond supplied audience | Yes | Depends on campaign settings and bidding |
| Typical role | Guide Google AI toward likely converters | Understand or restrict performance for defined audience segments |
Audience signals do not replace keywords. Search remains valuable for capturing explicit query intent, while audience data adds customer context. Our guides to Google Ads bidding strategies and Google Ads keyword match types explain the other side of that equation.
Observation vs Targeting in Search Campaigns
Search campaigns can use two audience settings:
| Setting | What it does | When to use it |
|---|---|---|
| Observation | Adds audience reporting without restricting campaign reach | Usually the safer default for Search when you want audience insights without narrowing keyword reach |
| Targeting | Restricts delivery to users who meet the audience criteria | Use when you intentionally want to narrow reach, such as a dedicated remarketing or tightly defined audience campaign |
Do not assume that a high-performing audience in Observation should automatically be moved to Targeting. That can materially reduce reach. Targeting should be a deliberate campaign-design decision.
How Smart Bidding Uses Audience Data
In Search campaigns using Smart Bidding, first-party audience segments added under Observation can already be used as signals by Google’s bidding system. That means advertisers do not always need to manually change bids for every audience segment.
Google also notes that bid adjustments behave differently with conversion-based automated bidding when multiple audience segments overlap. The practical takeaway is to evaluate the bidding strategy, conversion tracking and audience data together rather than treating audience bid modifiers as an isolated optimisation tactic.
For more on when to rely on automation and when to intervene, see our guide to Google Ads automation and AI in PPC.
How to Evaluate Audience Signals After Launch
Do not evaluate audience signals only by asking whether the exact audience you supplied converted. In AI-led campaigns, Google may find additional segments that perform well.
Use Audience Insights to understand the characteristics, interests and behaviours associated with users who view your ads and convert. Google currently provides audience insights across campaign types including Search, Performance Max, Demand Gen, Video and App.
- Review which audience segments make up conversions.
- Compare the segments you supplied with audiences Google discovered.
- Track CAC, conversion value and downstream quality, not just clicks.
- Use insights to improve creative, landing pages and future campaign structure.
- Do not assume correlation automatically proves that an audience caused the conversion.
For reporting workflows, see our guide to AI PPC reporting. A broader Google Ads audit should also verify audience setup, first-party data quality and campaign-level targeting settings.
Google Ads Audience Signals for B2B SaaS
B2B SaaS advertisers should prioritise first-party signals that reflect actual pipeline quality rather than broad website traffic.
- High-value customers: useful Customer Match seed data.
- SQLs and opportunities: often more informative than every form fill.
- Pricing and demo visitors: stronger first-party intent than generic blog traffic.
- Competitor and category custom segments: useful context for relevant prospecting tests.
- Existing customers as exclusions: valuable when the campaign objective is net-new acquisition.
This approach aligns audience strategy with pipeline rather than surface-level lead volume. Our AI audience targeting for B2B guide, SaaS PPC strategy guide and Google Ads agency for B2B SaaS page go deeper into campaign execution.
Common Google Ads Audience Signal Mistakes
- Treating Performance Max signals as hard targeting. PMax can serve beyond the signals you provide.
- Using low-quality first-party lists. A list full of weak leads gives Google weak context.
- Switching Search audiences from Observation to Targeting without a reason. This can unnecessarily restrict reach.
- Assuming Lookalikes work the same way they did historically. Demand Gen Lookalikes are moving toward suggestion-style expansion in 2026.
- Optimising only to CPL. Audience quality should be evaluated against downstream revenue or pipeline where possible.
- Ignoring exclusions. Existing customers or converted users may need to be excluded depending on the acquisition goal.
- Expecting immediate results. Google states that new audience signals can take up to two weeks to fully integrate.
Best Practices for Google Ads Audience Signals in 2026
- Start with first-party data. Customer and qualified-lead lists usually provide more useful business context than broad affinity segments.
- Segment by quality. Separate customers, opportunities, leads and generic visitors rather than mixing them into one list.
- Use custom segments to add category context. Relevant search terms, websites and apps can help Google understand the market you operate in.
- Keep Search Observation broad unless restriction is intentional. Gather data before narrowing reach.
- Judge success on business outcomes. Use pipeline, CAC and conversion value alongside platform metrics.
- Review Audience Insights. The segments Google discovers can inform creative and future targeting hypotheses.
- Give new signals time to integrate. Avoid judging changes after only a few days.
Frequently Asked Questions
What are audience signals in Google Ads?
Audience signals are inputs that tell Google AI about the demographics, interests, behaviours or first-party characteristics of users who may be valuable to your business. In Performance Max, they help guide optimisation but do not restrict delivery only to those users.
Are Performance Max audience signals targeting?
No. Performance Max audience signals are suggestions rather than hard targeting criteria. Google may serve ads to users outside the audiences you provide if its models predict they are likely to help achieve your campaign goal.
What audience signals should I use in Performance Max?
Start with high-quality first-party data such as Customer Match, qualified website visitors, app users and YouTube viewers. Add custom segments, in-market audiences and relevant demographic signals where they provide useful context.
What is the difference between audience signals and audience segments?
Audience signals guide Google AI in campaigns such as Performance Max, while audience segments are defined groups of users that can be observed or targeted in other campaign types. The exact behaviour depends on the campaign type and audience setting.
Should I use Observation or Targeting in Search campaigns?
Observation is usually the safer choice when you want audience reporting without limiting keyword reach. Targeting restricts delivery to users inside the selected audience and should be used when narrowing reach is intentional.
Can Performance Max target people outside my audience signals?
Yes. Google can expand beyond the audience signals you provide when its AI predicts other users are likely to help achieve the campaign’s conversion or value goal.
What is Customer Match in Google Ads?
Customer Match uses first-party customer information such as email addresses or phone numbers to create audience segments in Google Ads. It can support re-engagement, exclusions, retention and AI-led prospecting strategies.
How long do Google Ads audience signals take to work?
Google says new audience signals can take up to two weeks for its machine-learning models to fully integrate and optimise. Audience list sizes may also take additional time to populate after a new or updated list is added.
The best way to use audience signals in 2026 is to treat them as structured inputs into Google’s AI, not as a replacement for campaign strategy. Feed the system cleaner first-party data, understand when Google can expand beyond it, and use audience insights to improve the next round of campaigns.