Google Performance Max Campaigns: The Complete Guide for 2026
Google Performance Max campaigns have fundamentally changed how advertisers reach customers across the Google ecosystem. Launched as a unified campaign type that serves ads across Search, Shopping, Display, YouTube, Gmail, Discover, and Maps, google performance max represents Google’s most ambitious step towards AI-driven advertising. For marketers in Singapore and globally, understanding how to leverage PMax effectively is no longer optional; it is essential for competitive performance.
Yet Performance Max remains one of the most misunderstood campaign types in the Google Ads platform. Its black-box nature, limited reporting transparency, and automated optimisation can frustrate advertisers accustomed to granular control. Many businesses launch PMax campaigns with incorrect expectations, inadequate assets, or flawed configurations, leading to disappointing results and wasted budgets.
This comprehensive guide cuts through the confusion to deliver a clear, practical understanding of Performance Max in 2026. We cover how the campaign type works, how to structure your asset groups and audience signals, which bidding strategies to use, how to avoid common mistakes, and critically, when PMax is the right choice and when it is not. Whether you are managing your own campaigns or working with a Google Ads agency, this knowledge will help you make better decisions and drive stronger results.
How Performance Max Works: All Google Channels
Performance Max is a goal-based campaign type that uses Google’s AI to serve ads across all Google-owned channels from a single campaign. Unlike traditional campaign types that target one channel (Search campaigns for search results, Display campaigns for the Display Network, YouTube campaigns for video), PMax dynamically allocates budget and serves ads across every Google property where your target audience is present.
The channels covered by PMax include Google Search (including Shopping placements), Google Display Network (millions of websites and apps), YouTube (pre-roll, mid-roll, and discovery ads), Gmail (native ads in the Promotions tab), Google Discover (the feed on Android devices and the Google app), and Google Maps (local ads for businesses with Google Business Profiles). This comprehensive coverage ensures that your ads can reach users at every stage of their journey, from initial awareness through to purchase intent.
At its core, PMax relies on Google’s machine learning to make real-time decisions about which users to target, which channels to use, which ad formats to serve, and how much to bid. The algorithm processes vast amounts of signal data, including user behaviour, search history, demographics, device, location, time of day, and contextual relevance, to optimise for your specified conversion goal.
Advertisers provide the campaign with conversion goals, budget, creative assets, audience signals, and optionally a product feed. Google’s AI then combines these inputs to create and serve ads automatically. The system tests different combinations of your assets (headlines, descriptions, images, videos) to determine which combinations perform best for different audience segments and placements.
This automation offers significant efficiency gains, particularly for businesses with limited in-house PPC management expertise. However, it also means relinquishing much of the granular control that experienced advertisers are accustomed to. Understanding this trade-off is fundamental to setting appropriate expectations and extracting maximum value from Performance Max.
Asset Groups: Structure and Best Practices
Asset groups are the building blocks of Performance Max campaigns, functioning similarly to ad groups in traditional campaign types. Each asset group contains a set of creative assets (text, images, videos), audience signals, and optionally a set of products from your Merchant Centre feed. Google’s AI combines these assets to create ads tailored to each placement and audience segment.
A single PMax campaign can contain multiple asset groups, and each should be organised around a distinct product category, service offering, or audience theme. For example, an e-commerce retailer might create separate asset groups for “Men’s Footwear,” “Women’s Footwear,” and “Accessories.” A service business might have asset groups for “SEO Services,” “PPC Services,” and “Web Design Services.”
Each asset group should include the maximum number of assets to give Google’s AI the widest possible range of combinations to test. The recommended asset inventory includes up to 15 images (various aspect ratios including landscape 1.91:1, square 1:1, and portrait 4:5), up to 5 videos (or Google will auto-generate them, which is rarely ideal), up to 5 headlines (30 characters each), up to 5 long headlines (90 characters each), up to 5 descriptions (90 characters each), a business name, and final URLs.
Image quality matters enormously. Blurry, text-heavy, or generic stock images will underperform against crisp, original, product-focused photography. Ensure your images clearly showcase your product or service, use consistent brand styling, and are sized correctly for each required aspect ratio. Avoid images with excessive text overlays, as these perform poorly on most placements.
Video assets are critical but often neglected. If you do not provide video assets, Google will auto-generate videos from your images, and these automated creations are typically low quality and poorly branded. Invest in creating at least one high-quality video in landscape (16:9), square (1:1), and vertical (9:16) formats. Even simple product showcase videos or animated slideshows outperform auto-generated alternatives significantly.
Headline and description copy should cover different angles and value propositions. Include benefit-driven headlines, feature-focused headlines, urgency-driven headlines, and brand-name headlines. This variety allows the AI to match the most relevant message to each user context. Avoid repetition across headlines and descriptions, as this limits the AI’s ability to test distinct combinations.
Audience Signals: Guiding the Algorithm
Audience signals are suggestions you provide to Google’s AI about who your ideal customers are. Unlike audience targeting in traditional campaigns, audience signals are not hard restrictions. They serve as starting points that help the algorithm learn faster and find converting users more efficiently. The AI will ultimately expand beyond your signals to reach users it predicts will convert, regardless of whether they match your specified audiences.
The available audience signal types include custom segments (based on search terms people use, websites they visit, or apps they use), your data segments (remarketing lists, customer match lists, and similar segments), interests and demographics (in-market audiences, affinity audiences, life events, and detailed demographics), and demographic targeting (age, gender, parental status, household income).
First-party data signals are the most valuable input you can provide. Upload your customer email lists as Customer Match audiences and create website visitor remarketing lists segmented by behaviour (all visitors, product viewers, cart abandoners, past purchasers). These signals tell the AI exactly what your converting customers look like, enabling it to find similar users more quickly and accurately.
Custom segments based on search terms are particularly powerful for Performance Max. Input the keywords and search queries that your best customers use when looking for your products or services. This signals purchase intent to the AI and helps it prioritise users exhibiting similar search behaviour. Think of these as keyword-like inputs that influence the algorithm without restricting it.
Layering multiple signal types provides the richest starting point. Combine your customer match list with relevant in-market audiences and custom search term segments to create a comprehensive picture of your ideal customer. The more data points you provide, the faster the AI learns and the more efficiently it allocates your budget during the critical learning phase.
Avoid the temptation to leave audience signals blank or to add overly broad signals. While PMax will eventually optimise regardless, providing no signals extends the learning period and wastes budget on exploratory impressions. Conversely, overly narrow signals can limit the AI’s ability to discover valuable audiences you have not yet identified.
Bidding Strategies: tCPA, tROAS and Beyond
Performance Max supports two primary bidding strategies: Maximise Conversions (with optional target CPA) and Maximise Conversion Value (with optional target ROAS). Your choice of bidding strategy significantly impacts campaign performance and should align with your business objectives and available conversion data.
Maximise Conversions aims to generate the highest number of conversions within your budget. Adding a target CPA (cost per acquisition) constrains the algorithm to only bid when it predicts it can acquire a conversion at or below your specified cost. This strategy works well for lead generation businesses, service companies, and any business where conversions are relatively uniform in value. Set your target CPA based on your historical CPA plus a 10% to 20% buffer to allow the algorithm room to optimise.
Maximise Conversion Value aims to generate the highest total conversion value within your budget. Adding a target ROAS (return on ad spend) constrains the algorithm to achieve a specified return. This strategy is ideal for e-commerce businesses and any company where conversion values vary significantly (for example, if some products sell for $50 and others for $500). Your target ROAS should reflect your actual profitability requirements, accounting for cost of goods, margins, and acceptable customer acquisition costs.
Launching a new PMax campaign without a target CPA or target ROAS (using pure Maximise Conversions or Maximise Conversion Value) gives the algorithm maximum flexibility during the learning phase. This approach can accelerate learning but may result in volatile costs during the initial period. Once the campaign has accumulated sufficient conversion data (typically 30 to 50 conversions over a 30-day period), introduce a target to improve efficiency.
Budget setting for PMax campaigns should account for the algorithm’s need for sufficient daily budget to learn and optimise. Google recommends a daily budget at least three times your target CPA. Underfunded campaigns cannot gather enough data to optimise effectively, resulting in inconsistent performance and extended learning periods.
Conversion tracking accuracy is paramount for PMax bidding. If your conversion tracking overcounts, undercounts, or assigns incorrect values, the bidding algorithm will optimise towards flawed data. Audit your conversion tracking setup before launching PMax campaigns, ensuring that all valuable actions are tracked, attribution is correctly configured, and conversion values are accurate.
Feed-Only vs Full Asset Campaigns
Performance Max campaigns can run in two modes: feed-only (with a Merchant Centre product feed but no additional creative assets) and full asset (with both a product feed and creative assets including text, images, and videos). Understanding the distinction and choosing the right approach for your business is a critical strategic decision.
Feed-only PMax campaigns serve ads primarily through Shopping placements, using product images, titles, prices, and descriptions from your Merchant Centre feed. These campaigns function similarly to the legacy Smart Shopping campaigns and are the minimum viable PMax setup for e-commerce businesses. Without additional creative assets, the campaign has limited ability to serve ads on Display, YouTube, Gmail, and Discover.
Full asset PMax campaigns combine your product feed with additional creative assets, enabling the campaign to serve across all Google channels. This broader reach allows the campaign to build awareness among upper-funnel audiences on YouTube and Display while also capturing demand from high-intent shoppers on Search and Shopping. Full asset campaigns typically deliver higher overall reach and can acquire new customers who are not yet actively searching for your products.
For most e-commerce businesses, full asset campaigns outperform feed-only campaigns over time, as the additional channel coverage creates demand that subsequently converts through Shopping and Search. However, the path to conversion is longer and more complex, which can make attribution and ROAS measurement more challenging.
Service businesses without a product feed run PMax as full asset campaigns by default. These campaigns drive traffic to landing pages rather than product listings and rely heavily on creative assets and audience signals to identify and convert potential customers. For service businesses, the quality of your landing page experience and the clarity of your conversion actions are critical success factors.
A hybrid approach works well for businesses with both products and services. Create separate asset groups within a single PMax campaign, some linked to your product feed and others focused on service offerings. This structure allows the AI to allocate budget dynamically across both product and service promotions based on performance.
Search Themes and Keyword Intent
Search themes were introduced to give advertisers more influence over where their Performance Max ads appear in Search results. This feature partially addresses one of the most common criticisms of PMax: the lack of keyword-level control that search advertisers are accustomed to in standard Search campaigns.
Search themes are keyword-like inputs that signal to Google’s AI the types of search queries you want your PMax ads to appear for. You can add up to 25 search themes per asset group. Unlike keywords in Search campaigns, search themes are signals rather than targets. The AI will use them as guidance but may also serve your ads for queries outside your specified themes if it predicts they will convert.
Effective search theme selection requires understanding user intent at different stages of the buying journey. Include a mix of high-intent themes (such as “buy running shoes singapore” or “seo agency pricing”) and broader discovery themes (such as “best running shoes 2026” or “how to improve website ranking”). This range allows PMax to capture users at various funnel stages.
Search themes are particularly valuable for businesses in niche or specialised categories where Google’s AI might struggle to identify relevant search queries from audience signals alone. If your products or services have technical or industry-specific terminology, adding these terms as search themes helps the algorithm understand your market and find relevant searchers.
Monitor the search terms report (available under the Insights tab) to understand which queries are triggering your PMax ads. If you see irrelevant queries consuming budget, add them as account-level negative keywords (PMax supports account-level negatives through your Google Ads representative or the account-level negative keyword list feature). Refine your search themes based on the queries that are driving conversions.
The relationship between search themes and your standard SEO strategy deserves attention. Queries that you rank well for organically may not need PMax coverage, while queries where your organic visibility is weak represent opportunities for PMax to fill the gap. Coordinating your paid and organic search strategies avoids unnecessary spend and maximises total search visibility.
Cannibalisation with Search Campaigns
One of the most debated aspects of Performance Max is its potential to cannibalise traffic from standard Search campaigns. Understanding how PMax interacts with your existing Search campaigns is essential for maintaining efficient account structure and accurate performance attribution.
Google’s stated priority hierarchy gives exact match keywords in standard Search campaigns priority over PMax for identical search queries. This means that if a user searches for “digital marketing agency singapore” and you have that exact match keyword in a Search campaign, your Search ad should serve rather than your PMax ad. However, for queries that do not have an exact match keyword equivalent, PMax will compete with and often win the auction against your broad match and phrase match keywords.
In practice, cannibalisation is a genuine concern. Advertisers frequently observe their Search campaign traffic declining after launching PMax, with PMax claiming credit for conversions that Search campaigns were previously driving. This can create a misleading impression that PMax is generating incremental value when it may be partially redistributing existing performance.
To manage cannibalisation, maintain strong exact match keyword coverage in your Search campaigns for your highest-value queries. This ensures Search campaigns retain priority for the terms you care most about. Use account-level negative keywords to exclude specific queries from PMax where you have well-optimised Search campaigns running.
Incrementality testing is the gold standard for measuring PMax’s true contribution. Google offers conversion lift studies for larger advertisers, and you can conduct your own tests by running PMax in specific geographic regions while maintaining Search-only coverage in control regions. Compare total performance (PMax plus Search versus Search alone) to determine whether PMax is genuinely driving incremental conversions or merely redistributing them.
Monitor your Search campaign metrics closely after launching PMax. Watch for declining impressions, impression share, clicks, and conversions in Search campaigns. If you see significant decline without corresponding incremental performance from PMax, the campaign may be cannibalising rather than complementing your Search activity. Adjust your PMax configuration, audience signals, and search themes to reduce overlap.
Reporting, Insights Tab and Transparency
Reporting transparency remains the most significant limitation of google performance max campaigns. Compared to standard campaign types, PMax offers substantially less visibility into where your budget is being spent, which placements are performing, and which audiences are converting. This opacity is a deliberate trade-off for automation but creates challenges for advertisers who need to justify spend and optimise performance.
The Insights tab is your primary window into PMax performance drivers. It provides information on top-performing audience segments, search themes driving conversions, asset performance ratings, and auction insights. While not as granular as standard campaign reporting, the Insights tab offers actionable data that should be reviewed weekly to inform optimisation decisions.
Asset performance ratings (Best, Good, Low, or Learning) indicate how each individual asset performs relative to others in the same asset group. Assets rated “Low” should be replaced with new alternatives, while “Best” assets should be analysed to understand what makes them effective. Regularly refreshing underperforming assets keeps the creative mix fresh and maintains campaign performance over time.
Search terms reporting in PMax has improved but remains incomplete. Google now shows the top search queries driving impressions and conversions, though this does not cover all queries and may not include long-tail terms. Use this data to identify new keyword opportunities for your Search campaigns and to spot irrelevant queries that should be added to your negative keyword list.
Placement reporting is extremely limited. PMax does not provide detailed placement-level data for Display and YouTube, making it difficult to identify and exclude low-quality placements. You can view broad placement categories and top placements through the placements report, but the granularity is insufficient for the level of control most advertisers desire.
Third-party tracking and attribution tools can supplement PMax’s native reporting. Implementing UTM parameters on your PMax final URLs, using server-side conversion tracking, and integrating your CRM data with Google Ads all provide additional data points for understanding PMax’s contribution to your marketing performance. For businesses serious about measurement, these supplementary data sources are essential.
Best Practices and Common Mistakes
Success with Performance Max requires disciplined setup, ongoing management, and realistic expectations. These best practices, drawn from managing PMax campaigns for businesses across Singapore, will help you avoid the most common pitfalls and extract maximum value from the campaign type.
Start with strong conversion tracking. PMax is only as good as the data it optimises towards. Ensure all valuable conversion actions are tracked accurately, conversion values are correct, and your attribution model is appropriate. Enhanced conversions, which use first-party data to improve conversion measurement, should be implemented for all PMax campaigns.
Provide comprehensive, high-quality assets from day one. Launching with minimal assets forces Google to auto-generate creative, which invariably underperforms. Invest in professional photography, custom video production, and well-crafted copywriting before launching your campaign. Your assets are the raw material from which all your ads are created.
Allow adequate learning time. PMax campaigns typically require two to four weeks to exit the learning phase and achieve stable performance. Resist the urge to make significant changes during this period, as each major change resets the learning process. Monitor performance but avoid knee-jerk reactions to early-stage volatility.
Common mistakes include setting target CPA or target ROAS too aggressively at launch (which starves the campaign of data), using too few asset groups (which limits the AI’s ability to serve relevant ads), neglecting audience signals (which extends the learning period), ignoring the Insights tab (which means missing optimisation opportunities), and failing to coordinate with existing Search campaigns (which leads to cannibalisation).
Another frequent error is treating PMax as a “set and forget” campaign. While automation handles many optimisation tasks, human oversight remains essential. Review performance weekly, refresh creative assets monthly, update audience signals quarterly, and continuously refine your strategy based on Insights data and business performance.
For businesses working with a digital marketing agency, ensure your agency provides transparent reporting on PMax performance, including Insights data, asset performance, and incrementality analysis. The campaign type’s opacity makes agency accountability even more important.
When to Use PMax and When Not To
Performance Max is a powerful tool, but it is not the right solution for every advertising objective. Understanding when PMax excels and when alternative campaign types are more appropriate will help you build an effective, efficient Google Ads account structure.
PMax is ideal for e-commerce businesses seeking to maximise online sales across all Google channels, particularly when you have a well-structured product feed and sufficient conversion volume. It is also effective for lead generation businesses with clear conversion actions and adequate monthly lead volume (30 or more conversions per month). Businesses targeting broad audiences or looking to expand beyond their existing customer base benefit from PMax’s cross-channel reach and automated audience discovery.
PMax is less suitable for brand awareness campaigns where conversions are not the primary objective, for businesses with very low conversion volumes (fewer than 15 to 20 per month) where the algorithm lacks sufficient data to optimise, for accounts that require granular placement control or keyword-level reporting, and for businesses in sensitive or highly regulated industries where ad placement control is essential.
Many successful accounts use PMax alongside standard Search campaigns rather than replacing them entirely. This hybrid approach maintains keyword-level control for your most important search queries while using PMax to capture additional demand across the full Google ecosystem. The key is ensuring coordination between campaign types to minimise cannibalisation and maximise total performance.
For businesses just starting with Google Ads, beginning with standard Search campaigns is generally advisable. Search campaigns provide clearer data, more control, and faster learning about what works for your business. Once you have established conversion tracking, accumulated performance data, and built remarketing audiences, adding PMax to the mix amplifies your reach and efficiency.
Evaluate PMax performance holistically, considering its impact on total account performance rather than in isolation. A PMax campaign that appears to deliver a lower ROAS than your Search campaigns may still be driving incremental value by reaching new audiences who subsequently convert through other channels. Total business growth, not individual campaign metrics, should be the ultimate measure of success.
Frequently Asked Questions
Can I run Performance Max and standard Search campaigns simultaneously?
Yes, and in most cases you should. Google gives exact match keywords in standard Search campaigns priority over PMax for matching queries. Maintain strong Search campaigns for your highest-value keywords while using PMax to capture additional demand across other channels and queries. Monitor both campaign types closely for signs of cannibalisation and adjust your account structure as needed to optimise total performance.
How long does it take for a Performance Max campaign to start performing well?
Expect a learning period of two to four weeks during which performance may be volatile and costs may be higher than your targets. The algorithm needs this time to test different asset combinations, audience segments, and placements to identify what works for your business. Providing strong audience signals, high-quality assets, and adequate budget accelerates this learning process. Avoid making significant changes during the learning period, as each major change can reset the process.
Why is my Performance Max campaign spending mostly on Display instead of Search?
PMax allocates budget to the channels where it can most efficiently achieve your conversion goals. If Display is receiving a disproportionate share of spend, it may indicate that the algorithm is finding cheaper conversions on Display, that your Search campaigns are already covering high-intent queries, or that your audience signals are oriented towards upper-funnel audiences. Check your Insights tab to understand where conversions are coming from and adjust your audience signals and search themes to steer the algorithm towards your preferred channels.
How do I add negative keywords to Performance Max campaigns?
Performance Max supports account-level negative keywords, which can be added through the account-level negative keyword list in your Google Ads settings. Campaign-level negative keywords are not directly available in the PMax interface, but you can request them through your Google Ads representative or agency partner. Regularly reviewing the search terms in the Insights tab and adding irrelevant queries as account-level negatives is an important ongoing optimisation task for managing Google Ads costs effectively.
Is Performance Max suitable for small businesses with limited budgets?
PMax can work for small businesses, but there are minimum thresholds for success. Your daily budget should be at least three times your target CPA to give the algorithm sufficient data to optimise. If your target CPA is SGD 50, you need at least SGD 150 per day (approximately SGD 4,500 per month). Businesses with very small budgets may find standard Search campaigns more controllable and efficient. As your budget and conversion volume grow, adding PMax becomes increasingly viable and beneficial.
