Performance Max Guide 2026: The Complete PMax Campaign Handbook

Performance Max (PMax) is a goal-based campaign type in Google Ads that replaced Smart Shopping and Local campaigns in 2022. Rather than choosing individual channels, you provide creative assets, audience signals, and a conversion goal — and Google's machine learning decides where, when, and how to show your ads.

PMax campaigns can appear across all of Google's advertising inventory:

• Google Search — text ads alongside standard Search campaigns
• Google Shopping — product listings from your Merchant Centre feed
• Display Network — banner and responsive ads across millions of websites
• YouTube — video ads (in-stream and in-feed)
• Discover — native ads in the Google Discover feed
• Gmail — promotional tab ads
• Google Maps — local ads for businesses with physical locations

The appeal is simplicity and reach. One campaign, all channels, AI-powered optimisation. The trade-off is transparency — you get far less control and visibility than with traditional campaign types.

At its core, PMax is an AI-powered campaign engine. You feed it three things:

• Creative assets — headlines, descriptions, images, videos, and logos that Google mixes and matches into ad combinations
• Audience signals — hints about who your ideal customer is (first-party data, custom segments, demographics)
• Conversion goals — what you want to optimise for (purchases, leads, store visits)

Google's algorithm then tests thousands of creative combinations across all available channels, learning which placements, audiences, and asset combinations drive the most conversions for your budget. It uses real-time auction signals — device, location, time of day, search intent — to decide bidding and placement on a per-impression basis.

This is fundamentally different from traditional campaigns where you choose the channel, write the ads, pick the keywords, and set the bids. With PMax, you set the destination and Google drives.

PMax is not the right choice for every situation. Here's when it works well and when it doesn't:

Use PMax when:

• Ecommerce with a product feed — PMax excels at Shopping placements and can outperform Standard Shopping on total conversion volume. It's particularly strong for ecommerce PPC with large catalogues
• Lead generation with strong tracking — if you have offline conversion imports or lead scoring feeding back into Google Ads, PMax can optimise for quality leads
• Local businesses — PMax replaced Local campaigns and is effective at driving store visits and local actions
• Scaling beyond Search — when your Search campaigns are maxed out and you want incremental reach across Display, YouTube, and Discover

Think twice about PMax when:

• You need full control over search terms and keyword targeting
• Your conversion tracking is incomplete or unreliable
• You have a very small budget (under £1,500/month) — the algorithm needs sufficient data to learn
• You're in a highly regulated industry where ad copy must be precisely controlled
• You're running brand-only campaigns that need tight cost control

A well-structured PMax campaign starts with methodical setup. Here's the process:

1. Choose your campaign goal

Select the conversion action(s) that align with your business objectives. For ecommerce, this is typically purchases with revenue values. For lead gen, it should be qualified leads — not just form submissions. The goal you choose directly determines how the algorithm optimises.

2. Set your budget

Google recommends a daily budget of at least 3x your target CPA. If your target cost per acquisition is £50, start with at least £150/day. Insufficient budget starves the algorithm of data and extends the learning phase.

3. Select your bidding strategy

PMax supports Maximise Conversions (with optional target CPA) and Maximise Conversion Value (with optional target ROAS). Start with Maximise Conversions to gather data, then layer on a target CPA or ROAS once you have 30+ conversions per month.

4. Build your asset groups

Asset groups are the building blocks of PMax (more on structure below). Each group should have a clear theme with dedicated creative assets and audience signals.

5. Configure audience signals

Audience signals are not hard targets — they're suggestions that help the algorithm find your ideal customers faster. Include your first-party data, custom segments, and interest categories.

6. Upload creative assets

Provide as many high-quality assets as possible — headlines, descriptions, images, videos, and logos. Google will test combinations and surface what performs best.

7. Configure URL expansion settings

URL expansion allows Google to send traffic to pages beyond your specified final URL. This can be powerful for ecommerce (sending users to relevant product pages) but risky for lead gen (sending users to blog posts instead of landing pages). Consider disabling URL expansion for lead gen campaigns or excluding irrelevant URLs.

Asset groups are the structural foundation of PMax campaigns. Getting them right is critical:

• Theme each asset group tightly — each group should represent a distinct product category, service line, or audience segment. Don't lump everything into one group
• Match assets to themes — the headlines, descriptions, and images in each asset group should be relevant to that group's theme. Generic assets dilute relevance
• Use 3-7 asset groups per campaign — too few limits the algorithm's ability to segment; too many spreads budget too thin
• Assign unique audience signals per group — each asset group can have its own audience signals, allowing you to guide the algorithm differently for each theme
• Don't duplicate themes across campaigns — PMax campaigns will compete against each other if they target the same products or audiences

For ecommerce, structure asset groups around product categories or collections. For lead gen, structure around service lines or customer personas.

Audience signals are arguably the most important lever you have in PMax. They tell Google's AI where to start looking for conversions, dramatically accelerating the learning phase.

First-party data (most valuable)

• Customer match lists — upload your existing customer email lists so Google can find similar users
• Website visitors — remarketing audiences of people who have already engaged with your site
• Converters — audiences of past purchasers or lead submitters

Custom segments

• Search term-based — people who have searched for specific terms on Google
• URL-based — people who browse websites similar to URLs you specify (e.g., competitor websites)
• App-based — people who use specific apps related to your product

Demographics and interests

• In-market audiences — people actively researching products or services like yours
• Affinity audiences — people with long-term interests aligned with your offering
• Demographic targeting — age, gender, parental status, household income

Remember: audience signals are suggestions, not restrictions. Google will expand beyond your signals if it finds converting audiences elsewhere. The stronger your signals, the faster the algorithm learns.

PMax creates ads dynamically by combining your assets. The more high-quality assets you provide, the more combinations Google can test. Here's what to supply per asset group:

• Headlines — up to 15. Include a mix of benefit-driven, feature-driven, and brand headlines. Varying lengths (15-30 characters) ensures good fit across placements
• Long headlines — up to 5. These appear in Display and Discover placements where there's more space (up to 90 characters)
• Descriptions — up to 4. One short (60 characters) and the rest longer (up to 90 characters). Focus on your unique value proposition and clear calls to action
• Images — up to 20. Include landscape (1200x628), square (1200x1200), and portrait (960x1200) formats. High-quality product or lifestyle images outperform stock photos
• Videos — up to 5. If you don't upload videos, Google will auto-generate them from your images and text — and they're typically poor quality. Upload at least one 15-second and one 30-second video
• Logos — square (1200x1200) and landscape (1200x300) versions. These appear on Display and YouTube placements

Check the asset performance labels in Google Ads regularly. Replace assets rated "Low" with fresh alternatives and keep assets rated "Best" running.

Performance Max behaves very differently depending on whether you're running ecommerce or lead generation campaigns:

PMax for ecommerce

• Connect your Google Merchant Centre feed — this unlocks Shopping placements, which typically drive the majority of PMax conversions for ecommerce
• Use Maximise Conversion Value with a target ROAS — this ensures the algorithm optimises for revenue, not just transaction count
• Structure asset groups around product categories for relevance
• Enable URL expansion — letting Google route users to the most relevant product page increases conversion rates
• Supplement with dedicated ecommerce PPC strategies

PMax for lead generation

• Implement offline conversion tracking or lead scoring — without this, PMax optimises for form fills regardless of lead quality
• Use Maximise Conversions with a target CPA — set the CPA at your maximum acceptable cost per qualified lead
• Disable URL expansion or carefully exclude irrelevant pages — you don't want traffic landing on your blog or careers page
• Add negative keywords at the account level to prevent brand cannibalisation
• Monitor lead quality weekly, not just volume — PMax can deliver high volumes of low-quality leads if not constrained properly

One of PMax's biggest frustrations is limited reporting. Here's how to extract the most insight from what's available:

• Insights tab — shows top-performing audience segments, search term themes, and trending topics driving your performance
• Asset performance — each asset receives a performance label (Best, Good, Low, or Learning). Use these to iterate on creative
• Placement reports — see which websites, YouTube channels, and apps your ads appeared on. Exclude low-quality placements regularly
• Search terms insights — grouped by theme rather than individual queries. Less granular than Search campaigns but still useful for identifying patterns
• Conversion reports — break down by asset group to understand which themes drive the most value
• Auction insights — compare your share of voice against competitors in Shopping and Search placements

Key metrics to watch: cost per conversion, conversion value per cost (ROAS), impression share, and the split between Search/Shopping vs Display/YouTube placements. If Display and YouTube are consuming most of your budget without converting, your audience signals may need tightening.

These are the pitfalls we see most often when auditing PMax campaigns:

• Not uploading videos — if you don't provide video assets, Google auto-generates them from your images. These auto-generated videos are almost always low quality and can waste budget on YouTube placements that don't convert
• Skipping audience signals — running PMax without audience signals forces the algorithm to learn from scratch. First-party data and custom segments dramatically reduce the learning period and wasted spend
• Leaving URL expansion on for lead gen — URL expansion can send traffic to irrelevant pages like blog posts, about pages, or even job listings. Disable it or use URL exclusions for lead gen campaigns
• Only one asset group — lumping all products or services into a single asset group reduces relevance and limits the algorithm's ability to match the right creative to the right audience
• Insufficient budget — PMax needs sufficient daily budget (at least 3x your target CPA) to generate enough data for effective learning. Underfunded campaigns stay in the learning phase indefinitely
• Making changes during the learning phase — significant edits to budget, bidding, or asset groups reset the learning period. Resist the urge to tinker during the first 4-6 weeks
• Ignoring placement reports — PMax can place your ads on low-quality apps and websites. Review the placement report monthly and exclude underperforming placements

A Google Ads audit can identify these issues and prioritise fixes that will have the biggest impact on your PMax performance.

This is the most common comparison advertisers make. Here is how they stack up:

Many successful advertisers run both: PMax for broad reach and new customer acquisition, and Standard Shopping for high-margin or hero products where precise control matters. The key is ensuring they don't cannibalise each other — use campaign priority settings and shared negative keywords to manage overlap.