Optimising PMax campaigns – five key strategies for success
This article from the team at Upp.ai outlines five core strategies to get the best from your Google Performance Max (PMax) campaigns. It focuses on structuring asset groups, product feeds, budgets, audience signals and insight-driven adjustments so your campaigns don’t just run - they perform.
What is the first key strategy for PMax optimisation?
Structure asset groups for AI-driven efficiency.
To give Google’s automation the best chance, your asset groups must be well-segmented and populated with strong creative inputs. Examples include segmenting by product category (e.g., apparel: tops, trousers, dresses), using multiple headlines, descriptions, and image formats (square, horizontal, vertical) plus video assets.
How should you optimise your product listings to support PMax?
Optimise product listings in your Google Shopping feed.
Key actions: ensure accurate and updated pricing and stock, optimise product titles (brand + product type + colour/size/material), utilise custom labels (e.g., margin tier, demand status), and monitor the feed for errors or disapprovals.
What budget and target settings should you apply for sustainable growth?
Set campaign targets and budgets appropriately.
PMax campaigns tie budget and bidding together under automation (tROAS, maximise conversion value). Best practice: avoid making frequent or large changes to ROAS targets (Google recommends adjusting once per ~14 days and limiting change to ~20%) and steer clear of regularly hitting the “Limited by budget” flag since that constrains delivery.
How can audience signals improve PMax campaign performance?
Leverage audience signals for smarter targeting.
Although PMax automates targeting, supplying strong audience signals accelerates learning. Use first-party data (customer lists, website visitors), combine with in-market/affinity audiences, monitor which segments convert better using Audience Insights and refine accordingly.
What does the final strategy involve around insights and adjustment?
Analyse insights and make data-led adjustments.
You must review asset performance, product feed health, audience effectiveness, conversion value at SKU level, and allow the system time to learn rather than making knee-jerk changes. These adjustments help you move from simple automation to effective orchestration.