Back to Blog
GuidesMar 25, 2026|11 min read

Google Performance Max Campaigns: Complete Setup and Optimization Guide 2026

EA
Eduard Andrei

Founder at Adship

Google Performance Max Campaigns: Complete Setup and Optimization Guide 2026

Performance Max (PMax) is now Google's default campaign type for most advertisers. In 2025, Google sunset Smart Shopping and Local campaigns and merged them into PMax. In 2026, it's the primary vehicle for reaching customers across Google Search, Shopping, Display, YouTube, Discover, and Gmail — all from a single campaign.

For advertisers used to having control over keywords, audiences, and placements, PMax can feel like a black box. This guide demystifies it: how it works, how to set it up correctly, and how to optimize for maximum return.


What Is Performance Max and Why Google Made It the Default

Performance Max is an automated campaign type that uses Google's AI to serve ads across all Google channels simultaneously. You provide asset groups (creative inputs), audience signals (guidance), and a conversion goal — Google's algorithm handles the rest.

Why Google pushed it: PMax allows Google's AI to optimize across its entire inventory instead of siloed campaign types. For Google, this means higher revenue per advertiser. For advertisers, it means less manual management, but also less transparency into exactly where and how ads appear.

The core promise: Find more converting customers that keyword targeting alone would miss — people on YouTube who haven't searched yet, Gmail readers who match your customer profile, Display users who resemble your buyers.

For ecommerce brands, PMax replaced Smart Shopping campaigns and now includes organic Shopping inventory alongside paid. If you were running Smart Shopping before 2025, you're already in PMax.


How PMax Works: Asset Groups, Audience Signals, and AI Bidding

Asset Groups

The core unit of a PMax campaign is the asset group — a collection of creative inputs grouped around a theme, product category, or audience.

Each asset group contains:

  • Images: Up to 20 (various sizes: square, landscape, portrait)
  • Videos: Up to 5 (YouTube links or uploaded — 15 seconds minimum)
  • Headlines: Up to 5 (30 chars max each)
  • Long headlines: Up to 5 (90 chars max)
  • Descriptions: Up to 5 (60 or 90 chars)
  • Business name and logo
  • Call to action (optional to specify; Google auto-tests)

Google dynamically combines these assets into ads optimized for each channel and user. Strong asset input = better algorithm performance.

Audience Signals

Audience signals are suggestions, not restrictions. You tell Google who your ideal customers are; Google uses this as a starting point but may expand beyond it.

You can provide:

  • Customer lists (email upload — your past buyers)
  • Custom segments (people who searched specific keywords, visited competitor sites)
  • Website visitors (remarketing lists)
  • Interests and demographics

The most powerful signal: upload your purchaser list. Google will find people like your actual buyers, not just people who look demographically similar.

AI Bidding

PMax runs exclusively on Smart Bidding:

  • Maximize conversions — spend budget to get as many conversions as possible
  • Maximize conversion value — optimize for total revenue, not conversion count
  • Target CPA (tCPA) — hit a cost-per-conversion target
  • Target ROAS (tROAS) — hit a return on ad spend target

For ecommerce, Maximize conversion value with a tROAS is usually the right choice once you have sufficient conversion history (50+ purchases/month).


Setting Up a Performance Max Campaign: Step by Step

Step 1: Choose Your Conversion Goal

Before creating the campaign, ensure your conversion tracking is set up and firing correctly. PMax is only as good as the signal you feed it.

  1. Go to Google Ads → Goals → Conversions
  2. Verify your primary conversion action (usually "Purchase") is tagged and receiving data
  3. Set conversion value rules if your products have varying margins

Step 2: Create the Campaign

  1. Click + New Campaign
  2. Select your goal: Sales (for ecommerce), Leads, or Local store visits
  3. Choose Performance Max campaign type
  4. Connect your Merchant Center account (for Shopping integration — critical for ecommerce)

Step 3: Set Budget and Bidding

  • Budget: Start at 3–5x your target CPA per day. If target CPA is $30, start at $90–$150/day. Underfunding PMax kills the learning phase.
  • Bidding strategy: Start with Maximize conversion value without a ROAS target during the first 30 days. Add a tROAS target after you have 50+ conversions.

Step 4: Build Your Asset Groups

Create at least 2 asset groups — one generic (brand/catalog) and one specific (top product or category).

Asset group best practices:

  • Minimum quality score is "Good" — aim for "Excellent"
  • Include at least 1 video (Google auto-generates one if you don't, and the auto-generated versions are low quality)
  • Use all available headline and description slots
  • Upload images in all three aspect ratios (1:1, 1.91:1, 4:5)

For ecommerce: connect your Product Feed from Merchant Center to the campaign. This enables Dynamic Product Ads within PMax — showing specific products to people likely to buy them.

Step 5: Add Audience Signals

Add at minimum:

  1. Customer match list (existing purchasers) — highest priority
  2. Website visitors (past 30–90 days)
  3. Custom intent segment (people searching your core keywords)

Do not add too many audience signals — quality beats quantity. Your past purchasers list is worth more than 10 generic interest audiences.

Step 6: Geographic and Language Settings

Set your target geographies. If you ship to multiple countries, create separate PMax campaigns per country rather than combining them — conversion rates, CPA, and ROAS vary significantly by market.

Step 7: URL Expansion

PMax will send traffic to your most relevant landing pages by default ("URL expansion enabled"). For most ecommerce brands, leave this on. If you have weak pages you don't want Google to send traffic to, use URL exclusions to block them.


Performance Max for Ecommerce: Merchant Center Integration

For Shopify or WooCommerce stores, connecting your product feed is the difference between good and great PMax performance.

How to connect:

  • Shopify: Google & YouTube app (free) syncs your catalog automatically
  • WooCommerce: Google Listings & Ads plugin
  • Standalone: Google Merchant Center → Products → Feeds → Add feed

Once connected, Google can run Shopping ads (product cards in the Shopping tab) alongside your text, display, and video ads — all from one PMax campaign.

Feed optimization matters: Products with complete titles, descriptions, high-quality images, and correct GTINs outperform incomplete listings significantly. Optimize your feed before scaling PMax.

Conversion value rules: If your products have different margins, create conversion value rules to weight high-margin products more heavily. This tells the tROAS algorithm to prioritize what's actually profitable, not just what has volume.


Budget and Bidding Strategy for PMax

Learning Phase (First 30 Days)

  • Do not change bids, budgets, or asset groups frequently — each change restarts learning
  • Minimum 6 weeks without major changes for optimal learning
  • Expect higher CPAs in weeks 1–2; this is normal
  • Judge performance after 50+ conversions, not after week one

Scaling Past Learning Phase

Once past learning, you have two levers:

  1. Increase daily budget by 15–20% every 2 weeks if ROAS targets are met
  2. Reduce tROAS target to allow more volume at slightly lower returns

Don't increase budget and decrease ROAS simultaneously — restart issues compound.

When to Add a tROAS Target

Add a tROAS target after:

  • 30+ days of campaign running
  • 50+ conversions at campaign level
  • Stable performance trend (not trending worse)

Start conservatively: if your blended ROAS is 4x, start tROAS at 3.5x. Don't start with an aspirational target you haven't hit — the algorithm will restrict spend trying to hit an impossible goal.


Reporting and Optimization: Working Within PMax's Limitations

PMax is famously opaque. The primary optimization levers are assets and audience signals, not keywords or placements. Here's how to work within constraints.

Asset Performance Report

Campaigns → Asset groups → View assets

Google labels each asset as Best, Good, Low, or Learning. Replace "Low" assets after 30+ days. Don't obsess over "Learning" labels — every new asset starts there.

Search Terms Insight

Available under Insights & reports → Search terms insight — shows categories of search terms triggering your PMax ads. Not exact keywords, but directional enough to understand intent.

Use this to:

  • Identify irrelevant search categories → add negative keywords at account level
  • Find high-performing categories → create Search campaigns to capture keyword-level control

Placement Reports

Download placement reports via Report Editor → Placements to see where your display/YouTube ads are showing. Exclude placements with high spend and zero conversions.

Audience Insights

Check Audience insights for signals about who's converting. Use this to refine your customer match lists and custom segments.


PMax vs Smart Shopping vs Standard Shopping: When to Use Each

ScenarioBest Campaign Type
Strong pixel data, 50+ purchases/monthPerformance Max (tROAS)
Launching new Shopify store, limited dataStandard Shopping + Search
Precise audience and keyword control neededStandard Search or Standard Shopping
Brand awareness for new marketDisplay or YouTube (not PMax)
Testing specific products or offersStandard Shopping (controllable)
Maximizing reach across all Google channelsPerformance Max

The general rule: PMax outperforms Standard Shopping for mature advertisers with pixel history. It underperforms for new advertisers, low-volume businesses, or campaigns needing tight control.


How PMax and Search Work Together

Keep dedicated Search campaigns for branded terms and the highest-value queries where you need explicit copy, landing pages, negatives, and bid control. Use PMax for broader discovery, Shopping inventory, and cross-channel reach. Review overlap through search insights, then move valuable query themes into Search when tighter control would improve the experience. The goal is a clear role for each campaign type, not forcing one automated campaign to replace every controlled search program.


Managing PMax Alongside Meta Ads with Adship

Most ecommerce businesses running PMax are also running Meta Ads. The challenge: fragmented reporting across platforms makes it hard to see which platform is actually driving incremental revenue.

Common issue: Both Meta and Google claim credit for the same purchase. Your Facebook last-click shows 200 conversions. Your PMax shows 180 conversions. Your actual orders: 250.

With Adship, you can manage your Meta and TikTok campaigns in one dashboard while you run PMax in Google Ads. Keeping the Meta and TikTok side organized makes it easier to compare what each channel reports and make better budget allocation decisions.

Budget allocation signals to watch:

  • If PMax ROAS (as reported) > Meta ROAS by 50%+, shift budget toward Google
  • If Meta is driving most new customer revenue (higher purchase rate on new buyers), Meta may deserve more budget
  • Run incrementality tests (Meta's built-in, Google's conversion lift) annually to validate platform ROAS claims

Key Takeaways

  1. PMax is input-dependent — the algorithm is only as good as your assets, audience signals, and conversion data
  2. Don't interrupt learning — 30–60 days without major changes is the minimum for real signal
  3. Optimize assets, not audiences — audiences are signals; creative is your primary lever
  4. Connect your product feed — non-negotiable for ecommerce
  5. Add tROAS only after 50+ conversions — don't target what you haven't achieved
  6. Negative keywords at account level — only way to restrict irrelevant search categories
  7. Separate campaigns by country — performance varies too much to combine

Performance Max is powerful when configured correctly. The brands struggling with it are usually underfunding the learning phase, providing too few creative assets, or setting overly aggressive ROAS targets before the algorithm has enough data to work with.

Meet the AI Ad Operating System

Scan, spy, create, and launch Meta and TikTok ads with guarded actions taken with your approval.

Start Free
Share this article

Research. Create. Launch. Learn.

Adship connects Agent, Ad Spy, Canvas, and Learning Loop across Meta and TikTok, with bulk launch and guarded actions built in.

Get Started: It's Free