PPC Live #16: Performance Max, Leveraging Google Ads simulations, and the New Rules of Ecommerce Advertising

At PPC Live #16, a clear message emerged: PPC professionals must evolve from campaign managers to strategic partners. Sophie Fell emphasized the importance of testing and curiosity with tools like Performance Max. António demonstrated how using marginal ROAS and Google Ads bid simulations can lead to smarter, data-driven budget allocation. Sem Tielemans reinforced this shift, urging a focus on data ownership, attribution, and multichannel strategy as automation reduces the value of hands-on campaign execution. Read below to get more of the key takeaways from our expert speakers.

Watch the live stream of all talks.

AI Anxiety? 6 Guardrails to Tame Performance Max – Sophie Fell

Sophie explored the evolution, power, and complexity of Google’s Performance Max (PMax) campaigns, highlighting the anxieties many advertisers have due to automation, lack of control, and opaque results.

To address these concerns, she proposes six guardrails—practical strategies to harness Performance Max effectively and responsibly. These strategies are backed by data from Optimyzr and Sophie’s agency experience.

In her talk, she emphasized critical thinking, testing assumptions, and using tools intelligently rather than blindly trusting automation.

The Six Guardrails :

1. Use First-Party Data
2. Check Default Settings (Don’t Skip the Basics)
3. Consider Brand Exclusions & Negative Keywords
4. Leverage Audience Signals & Search Themes
5. Exclude Poor Placements
6. Review & Control Final URLs

Bonus Insights

Reminder: “Better” performance doesn’t always mean more incremental performance.

The Optimyzr study showed counterintuitive results: accounts with fewer controls (no negatives, no signals) sometimes perform better—but this might be due to brand traffic inflating results.

Multiple attendees questioned the validity of the findings without isolation of variables in the data, suggesting the need for more robust testing.

Marginal ROAS and budget redistribution” leveraging Google Ads simulations – António Migel Lima

António delivered a data-driven session exploring how to optimize budget allocation in Google Ads using marginal ROAS (Return on Ad Spend) and bid simulations. He opened with analogies to farming and resource allocation (e.g., tractors or laptops) to illustrate the principle of diminishing returns—the idea that initial investments are typically more productive than later ones.

He contrasted average ROAS, commonly used in PPC analysis, with marginal ROAS, which reflects the return on the next pound spent. Lima emphasized that campaigns with similar average ROAS might have drastically different marginal efficiencies, especially depending on metrics like impression share.

To estimate marginal ROAS, he outlined three methods:

1. Experimentation (e.g., A/B testing with different ROAS targets)
2. Statistical modeling (e.g., regression analysis)
3. Google Ads bid simulations (his preferred method)

He detailed how Google Ads provides simulated performance curves that estimate outcomes at different bid targets. By extracting this data at scale via the Google Ads API (GAQL) and storing it in a data warehouse, teams can calculate marginal ROAS across all campaigns and rank them by efficiency.

This data enables:

• Strategic conversations with senior stakeholders using a single, intuitive curve.
• Tactical decisions, like adjusting individual campaign targets to hit a desired budget with the highest possible efficiency.

The key message: Not all budget is equally productive. By leveraging bid simulations and marginal ROAS, marketers can make smarter, data-backed budget allocation decisions that go beyond simplistic averages.

What’s next for PPC? Data, AI, and the New Rules of Ecommerce Advertising – Sem Tielemans

Sem Tielemans, from EO, shared insights gathered from working with over 100 agencies across Europe. EO is a CSS partner focused on helping advertisers improve their Google Shopping performance through tools like product segmentation (e.g., “Heroes and Zombies”) and feed optimization.

Key Themes:

1. Shift in Agency Value:
   • Traditional PPC work (like campaign management) is becoming less valuable due to automation.
   • Agencies must now focus more on strategic thinking, data management, and attribution modeling.
   • Clients want growth partners, not just campaign managers.
2. Collaboration & Strategy:
   • PPC specialists need to collaborate more—internally and with clients—to understand business objectives and guide strategy across multiple channels.
3. Importance of First-Party Data:
   • With cookies fading out and GA4 being complex, first-party data is crucial.
   • Agencies must help clients track and interpret performance effectively.
4. AI and Automation:
   • Tools like Performance Max can be helpful but must be fed with the right data.
   • Relying too heavily on automated platforms like Google can lead to wasted budget on easy wins (like branded search).
   • Agencies should control AI, not be controlled by it.
5. Multichannel and Influencer Strategy:
   • PPC professionals should look beyond paid search—incorporate paid social, YouTube, TikTok, etc.
   • More touchpoints mean a need for better attribution modeling and insight into where influence happens, not just conversions.
6. Opportunities for PPC Specialists:
   • Specialists who understand data, multiple platforms, and business strategy will have a competitive advantage.
   • Agencies struggle to find talent that can do both strategy and execution—this is a growth area.

Takeaways:

• Move from execution to strategy.
• Embrace first-party data and attribution modeling.
• Be a business partner, not just an ad manager.
• Adapt to a multichannel world and learn to connect the dots.

Watch the live stream of all talks.