Google Ads audience targeting in 2026 is no longer about picking a demographic and hoping for the best. It's a layered system where first-party data, in-market signals, custom intent, and Performance Max audience signals work together to find buyers at the moment they're ready. Get the architecture right and your cost per acquisition drops without touching bids. Get it wrong and Smart Bidding wastes spend across audiences that look broad on paper but never convert.
This guide covers all seven Google Ads audience types, how each one works in 2026, where to use them, and how the privacy changes rolling through Chrome, the EU, and Australia are reshaping the playbook. It sits inside the wider performance marketing programme we run for clients, and pairs with our top agency tips for Google Ads for account-level execution.
The 7 Google Ads audience types at a glance
Google Ads offers seven core audience types, each serving a distinct role in your marketing funnel:
- Affinity: reaches users based on long-term interests and habits. Top of funnel (awareness). Best on Display, Video, Demand Gen.
- In-Market: reaches users actively researching products in your category. Mid-to-lower funnel. Best on Search, Display, Video, Shopping, Performance Max signals.
- Custom: targets users by keywords, URLs, and apps you define. Mid funnel. Best on Display, Video, Demand Gen.
- Remarketing: re-engages people who interacted with your site, app, or YouTube. Lower funnel (conversion). Works across all campaign types.
- Customer Match: targets users on Google's network using your customer data. Lower funnel plus exclusions. Best on Search, Shopping, Display, Video, Gmail.
- Similar Segments: finds new users who resemble your existing customers. Mid-to-lower funnel. Best on Search, Display, Video, Shopping.
- Demographics: layers age, gender, household income, parental status. Works across any campaign type as a refinement on the above.
Performance Max sits outside this list because it doesn't use audiences in the traditional sense. It uses audience signals, which are directional inputs rather than targeting restrictions. We cover that in detail below.
1. Affinity audiences
Affinity audiences reach users based on their long-term interests and habits. Think 'outdoor enthusiasts', 'luxury shoppers', or 'business professionals'. Google builds these segments from browsing patterns, app usage, and YouTube viewing over extended periods.
These are broad segments, often spanning millions of users in a single market. That breadth makes them strong for top-of-funnel brand awareness and weak for direct response. Use affinity audiences when you want to introduce your brand to a category-relevant audience at scale. Pair them with qualified keywords on Search or with strong creative on Display and Video.
Custom affinity audiences let you build your own affinity segment using interests, URLs, places, and apps. They're useful when the standard segments are too broad or don't match your customer profile.
2. In-market audiences
In-market audiences target users who are actively researching or comparing products and services in a specific category. Google identifies these users through search behaviour, content consumption, and browsing patterns. If someone has spent the past week comparing accounting software, they appear in the 'Business Software' in-market segment until that signal decays.
In-market audiences are the workhorse of audience targeting for direct response campaigns. They sit closer to the conversion event than affinity but stay broad enough to deliver volume. Use them on Search to bid higher on commercial-intent keywords when the user is also in-market. Use them on Display, Video, and Demand Gen to find users in the consideration window. Use them as audience signals in Performance Max to point the algorithm at the right consideration pool.
Google maintains 18 main in-market categories, each with subcategories. The categories cover Apparel and Accessories, Autos and Vehicles, Baby and Children's Products, Beauty Products and Services, Business Services, Computers and Peripherals, Consumer Electronics, Consumer Software, Dating Services, Education, Employment, Financial Services, Gifts and Occasions, Home and Garden, Real Estate, Sports and Fitness, Telecom, and Travel.
The subcategories are where the real targeting precision lives. 'Financial Services' contains banking, credit cards, investment services, loans, insurance, and tax preparation, each with their own deeper subcategories. For the full breakdown, including which subcategories work best for which campaign types, see our complete Google Ads in-market audiences list.
3. Custom audiences
Custom audiences let you build your own segments using keywords, URLs, and apps your ideal customers engage with. You can target users who have searched for specific terms on Google or who regularly visit competitor websites.
In 2026, custom audiences are powerful in two specific scenarios. First, when you have a clear competitive set and want to reach users who visit those competitor sites. Adding three to five competitor URLs to a custom audience and pairing it with Display or Video campaigns puts your ad in front of people already shopping in your category. Second, when standard in-market segments are too broad or don't exist for your niche. A custom audience built from high-intent search keywords gives you in-market-style reach for product categories Google doesn't natively segment.
The trade-off: custom audiences require active maintenance. URLs change, competitors come and go, and keyword sets need refreshing every quarter to stay relevant.
4. Remarketing audiences
Remarketing remains one of the highest-return tactics in Google Ads, but the approach has matured well beyond 'show the same ad to everyone who visited your site'.
Standard remarketing
Standard remarketing targets past website visitors with Display and Video ads across the Google Display Network and YouTube. Segment your lists by recency (1 to 7 days, 8 to 30 days, 31 to 90 days) and by page depth. Someone who viewed a pricing page is far more valuable than someone who bounced from the homepage.
Dynamic remarketing
Dynamic remarketing automatically generates personalised ads showing the specific products or services a user viewed. For eCommerce, this means showing the exact shoes someone left in their cart. For lead-gen businesses, it means highlighting the specific service page they explored. It requires a product or service feed connected to your Google Ads account.
RLSA (remarketing lists for search ads)
RLSA applies your remarketing audiences to Search campaigns. You can either adjust bids for past visitors searching relevant terms or restrict your Search campaigns to only show ads to people already in your remarketing lists. RLSA is particularly powerful for broad keywords that would otherwise be too expensive. Targeting 'accounting software' only to users who have previously visited your pricing page dramatically improves conversion rates and reduces wasted spend.
5. Customer Match
Customer Match lets you upload your own customer data (email addresses, phone numbers, or mailing addresses) and target those users across Google's network: Search, Shopping, YouTube, Gmail, and Display. The data is hashed before upload, so Google never sees the raw values.
In 2026, Customer Match has become the single most important audience type for sophisticated advertisers. Three uses matter most:
- Upsell and cross-sell: target existing customers with relevant new products or upgrades.
- Suppression: exclude existing customers from acquisition campaigns so you stop paying to re-acquire them.
- Seeding for similar segments: use a high-value customer list as the seed for a Similar Segment to find more buyers who resemble your best ones.
Customer Match requires consent from the users whose data you upload, and the list needs to refresh regularly as customers churn or change contact details. Most accounts we audit either don't use Customer Match at all or have stale lists that haven't been refreshed in six months. Both are missed opportunities.
6. Similar segments
Similar segments use your existing audience data to find new users who share characteristics with your best customers. Google deprecated traditional lookalike audiences in favour of similar segments, which are now generated automatically from your remarketing lists and Customer Match lists.
The quality of a similar segment depends entirely on the quality of the seed list. A similar segment built from a list of 'newsletter signups' will find more newsletter signups. A similar segment built from a list of 'customers who spent over $500 in the last 90 days' will find more high-value buyers. Always seed similar segments from your strongest first-party data, never from generic site visitors.
7. Demographic targeting
Demographic targeting layers on age, gender, household income, and parental status to refine any of the above. In 2026, detailed demographics also include education, marital status, homeownership, and employment industry.
Demographics rarely work well on their own. They work well as exclusions (cut out demographics that don't fit your product) or as layered refinements on top of in-market or remarketing audiences. The risk of using demographics as your primary targeting is volume: you narrow reach quickly, and Smart Bidding loses the signal it needs to optimise.
Google Ads video targeting options in 2026
Video campaigns on YouTube use the same seven audience types described above, plus a separate set of content-based targeting options that don't apply to Search or Display in the same way.
Content targeting for video includes Topics (broad subject categories), Placements (specific YouTube channels, videos, or websites), Keywords (contextual keywords tied to video content), and Devices (computer, mobile, TV screen). Topic and placement targeting is what makes YouTube unique. You can place your ad on specific competitor channels, on videos about specific products, or alongside topics like 'home renovation' or 'small business accounting'.
Google's official guidance for conversion-focused Video campaigns is to keep targeting broad and let the algorithm find converters. Adding content targeting (keywords, topics, or placements) on a conversion campaign typically limits reach and hurts performance. The opposite is true for awareness or consideration campaigns, where content alignment matters more than algorithmic efficiency.
There are also several 2026-specific changes worth knowing: Google Ads is auto-linking YouTube channels from 10 June 2026, Shorts inventory is now available through Demand Gen and Performance Max, and Connected TV targeting has expanded across smart TV apps. For the full breakdown, see our guide to Google Ads video targeting options.
How Smart Bidding uses audience signals
Smart Bidding has fundamentally changed how audience targeting works. In earlier years, advertisers manually set bid adjustments for each audience segment, raising bids 20% for in-market users, lowering them for broad demographics. Smart Bidding strategies like Target CPA, Target ROAS, and Maximise Conversions now handle this automatically.
Google's AI evaluates hundreds of signals in real time for every auction: device, location, time of day, browser, operating system, and crucially, the user's audience membership. When you add audience segments to a Smart Bidding campaign, you are not telling Google how much to bid for those users. You are giving the algorithm a signal that helps it identify high-value patterns faster. The machine learning model uses your audience data as one input among many to predict conversion probability and set the optimal bid.
This is why audience signals matter even when you are not manually adjusting bids. The audiences you add act as training data for the model.
Audience signals in Performance Max
Performance Max represents the most significant shift in how audiences work on Google Ads. Unlike traditional campaigns where you target specific audiences, Performance Max uses audience signals as directional inputs. You provide customer segments, custom audiences, and demographic hints, and Google's AI uses these as starting points to find converting users across Search, Shopping, Display, YouTube, Discover, Gmail, and Maps.
The critical distinction: audience signals in Performance Max are suggestions, not restrictions. Google will expand well beyond your provided audiences if it identifies converting users elsewhere. This means your signals need to be high quality. They set the initial direction for the AI.
For each asset group, we typically provide three to five audience signals: a Customer Match list of past purchasers, an in-market segment matching the core product category, a custom audience built from competitor URLs and high-intent keywords, a remarketing list of recent website visitors, and a demographic filter if the product has a clear demographic skew. Read our Performance Max guide for the full asset group setup.
First-party data: Customer Match, Enhanced Conversions, Consent Mode v2
First-party data has become the single most valuable asset in digital advertising. With third-party cookies being phased out across browsers and privacy regulations tightening globally, the data you collect directly from your customers is what separates high-performing accounts from the rest.
Customer Match (covered above) sits at the centre of any modern first-party data strategy. If you're not actively using and refreshing Customer Match lists, you're operating with one hand tied behind your back.
Enhanced conversions improve measurement accuracy by securely sending hashed first-party conversion data (email, phone, address) back to Google. This recovers conversions that would otherwise be lost to cross-device journeys and cookie restrictions. Google reports that enhanced conversions typically recover 5% to 15% of previously untracked conversions.
Consent Mode v2 is now mandatory in the EU and EEA, and we recommend implementing it for all Australian accounts as well. It adjusts Google tag behaviour based on user consent choices, allowing Google to model conversions from users who decline cookies. Without Consent Mode, you lose visibility into a growing share of your conversion data.
Observation vs Targeting: getting it right
Two settings that confuse many advertisers are 'Observation' and 'Targeting', and choosing the wrong one can quietly destroy campaign performance.
Observation adds an audience to your campaign for reporting only. Ads still show to everyone, so use it when you want to test audience hypotheses without restricting reach. Targeting restricts ads to users in the selected audience, so use it when you have proven, high-performing audiences and want to concentrate spend.
The best practice in 2026 is to start with Observation, gather 30 to 60 days of data, and then shift high-performing audiences to Targeting with increased bids.
Audience layering combines multiple targeting methods. For example, you might layer an in-market audience with a demographic filter and a geographic restriction: 'Users in-market for home loans, aged 30 to 45, in Sydney.' Each additional layer narrows reach but increases intent alignment. The key is finding the balance between precision and volume. Too many layers and you won't generate enough impressions to let Smart Bidding optimise effectively.
Privacy changes reshaping audience targeting in 2026
Privacy changes are reshaping audience targeting permanently, and advertisers who adapt early gain a structural advantage.
Third-party cookie deprecation across Chrome (Google's revised timeline now extends through 2026) means remarketing lists built on cookie-based tracking will shrink. The countermeasure is investing in first-party data collection: email signups, loyalty programmes, CRM integrations, and server-side tracking.
Google's Privacy Sandbox APIs (Topics API, Protected Audience API, and Attribution Reporting API) are replacing cookie-based targeting with privacy-preserving alternatives. Topics API assigns users to broad interest categories based on browsing history, refreshed weekly, with no cross-site tracking. Protected Audience API enables on-device remarketing auctions without exposing user data to third parties. These APIs are less granular than cookies, which means your first-party data and contextual targeting strategies become even more important.
Australian Privacy Act reforms proposed in 2025 introduce stricter consent requirements and expanded definitions of personal information. Advertisers operating in Australia should implement Consent Mode v2 now, audit data collection practices, and ensure all Customer Match uploads comply with data handling requirements.
What consistently drives results
After managing over $12 million in Google Ads spend, here are the audience targeting practices that consistently drive results for our clients.
Build your first-party data engine before you need it. Every website visit, email signup, and purchase should flow into a Customer Match list. Segment ruthlessly: separate purchasers from leads, high-value from low-value, recent from lapsed.
Use exclusion audiences as actively as you use targeting audiences. Suppress existing customers from acquisition campaigns, exclude recent purchasers from remarketing, and remove converted users from lead-gen campaigns to avoid wasting budget.
Test audience signals in Performance Max by running controlled experiments. Swap one signal at a time and measure the impact on CPA and conversion volume over two to four weeks.
Align audiences to funnel stages: Affinity and broad Custom audiences for awareness, In-market and RLSA for consideration, Remarketing and Customer Match for conversion.
Refresh your remarketing lists regularly. A 540-day-old cookie is unlikely to convert. We cap most remarketing lists at 90 days and create separate asset groups for each recency window.
Monitor audience overlap in Audience Manager to ensure you are not bidding against yourself across campaigns.
Frequently asked questions
How many in-market audiences does Google Ads have?
Google Ads offers 18 main in-market audience categories, each containing multiple subcategories. The full list includes Apparel and Accessories, Autos and Vehicles, Baby and Children's Products, Beauty Products and Services, Business Services, Computers and Peripherals, Consumer Electronics, Consumer Software, Dating Services, Education, Employment, Financial Services, Gifts and Occasions, Home and Garden, Real Estate, Sports and Fitness, Telecom, and Travel. See the complete in-market audiences list for every subcategory.
What's the difference between affinity and in-market audiences in Google Ads?
Affinity audiences target users based on long-term interests and habits, useful for brand awareness. In-market audiences target users actively researching or comparing products in a specific category right now, useful for direct response. Affinity is for the top of the funnel, in-market is for the middle and bottom.
Do audience signals still matter in Performance Max?
Yes. Performance Max audience signals are not restrictions, but they direct Google's AI toward the most relevant pool of potential customers at campaign launch. High-quality signals (Customer Match lists, in-market segments, custom audiences from competitor URLs) accelerate learning and improve cost per acquisition during the ramp-up phase.
What replaced similar audiences in Google Ads?
Similar audiences were deprecated and replaced by Similar Segments, which are automatically generated from your remarketing and Customer Match lists. They work across Search, Display, Video, and Shopping campaigns.
How do you target YouTube viewers in Google Ads?
YouTube and Video campaigns support all seven core audience types plus content-based targeting (Topics, Placements, Keywords, Devices). For awareness campaigns, layer in content targeting alongside affinity or in-market audiences. For conversion campaigns, Google recommends keeping targeting broad and using audience signals rather than content restrictions. See our Google Ads video targeting options guide for the full breakdown.
Should I use Observation or Targeting for my audiences?
Start with Observation when testing a new audience. Let the data accumulate for 30 to 60 days. If a specific audience shows materially better performance than the campaign average, move it to Targeting with an increased bid. Use Targeting from day one only when you already have proven audience data from previous campaigns.
Get audit-ready audience strategy from a Google Premier Partner
As a Google Premier Partner for five consecutive years, ClickedOn manages audience strategy across the full funnel for brands in financial services, eCommerce, health, and technology. If your Google Ads account relies on broad targeting without a clear audience strategy, you are almost certainly leaving conversions on the table.
Get in touch for a free audit. We'll show you exactly where your audience gaps are and how to close them.
