Run an Impact Estimate (Beta) in Google Ad Manager: Everything You Need to Know

Point of this article..

• Impact Estimate (Beta) lets publishers test identifiers before rolling them out widely.
• Results help measure revenue, eCPM, and impression changes with statistical significance.
• PPID and First-Party IDs are key identifiers publishers can test to optimize performance.

---

Discover how to run an impact estimate (Beta) in Google Ad Manager, review the results, and learn the differences between PPID and First-Party IDs to prepare for a cookie less future.

As the advertising industry transitions away from third-party cookies, publishers must find new ways to sustain performance while respecting user privacy. Google Ad Manager’s Impact Estimate (Beta) provides a safe testing environment to simulate the effect of identity solutions such as Publisher Provided Identifiers (PPIDs) and First-Party IDs. Instead of making risky changes to live setups, publishers can experiment first, review accurate results, and make informed decisions.

This guide explains what the tool is, how to set it up, how to review results, the practical value it brings, and the difference between PPIDs and First-Party IDs.

What Is “Run an Impact Estimate (Beta)”?

The Impact Estimate (Beta) is a feature in Google Ad Manager designed to predict how identifiers influence monetization. It allows publishers to run controlled experiments that compare baseline traffic (with identifiers disabled) to variant traffic (with identifiers enabled).

The main goals are to:

• Forecast performance metrics such as revenue, eCPM, and impression lift.
• Split traffic into groups (baseline vs. variant) to ensure a fair comparison.
• Provide insights across browsers and devices, particularly where third-party cookies are unavailable.
• Support future readiness by helping publishers adopt privacy-safe identifiers like PPID and First-Party IDs.

In short, it’s a tool to reduce guesswork and strengthen strategies with real data.

Set Up an Impact Estimate

Getting started with an impact estimate requires choosing a test type and configuring specific settings. Once set, publishers can run the estimate without changing their existing ad serving setup.

Steps to set up an impact estimate:

1. Sign in to Google Ad Manager.
2. Go to Optimization → Impact estimates.
3. Choose the type of estimate available:
   • PPID and First-Party IDs → Estimates the effect of PPIDs (for programmatic) and First-Party IDs on traffic.
     
     For the Beta, only PPID and First-Party IDs are supported.
     
   • Secure signals → Estimates revenue impact of active secure signals.
     
4. For your selected type, click New impact estimate.
5. Configure settings:
   • Estimate options → Select PPID for programmatic, First-Party IDs, or both.
     Important: To run a PPID estimate, PPID must already be included in ad requests to Ad Manager.
   • Demand channel settings → Choose to override all demand settings and share IDs where not enabled, or only adjust for allowed bidders.
   • Estimate variants → Click Show variants breakdown to see baseline (feature off) vs. variants (feature enabled).
   • Delivery settings → Define timeframe and traffic allocation.
   • Estimate procedure → Review how the estimate will be executed.
     
6. Click Run → Confirm.

Once confirmed, Ad Manager automatically applies the necessary controls. Importantly, this process does not change your existing live settings, ensuring experiments remain safe.

Review the Results

Impact estimates can be reviewed at any stage. For PPID and First-Party ID estimates, results exclude the first week of data because buyers need time to incorporate new identifiers into their targeting and bidding. This ensures results reflect actual impact rather than early noise.

Steps to review results:

1. Sign in to Google Ad Manager.
2. Go to Optimization → Impact estimates.
3. In the table, check:
   • Impact estimate → The type of estimate (e.g., First-Party IDs).
   • Estimate period → The timeframe selected.
   • Status → Scheduled, running, or complete.
4. Click the name of an estimate to dive deeper:
   • Change timeframe → Adjust “View metrics by” (e.g., Weekly).
   • Compare variants → Review baseline vs. variants for revenue, eCPM, and impression lift.
   • Apply breakdowns → Explore results by dimensions like browser, device, or demand channel.
     Tip: Add filters for more specific data views.
   • Expand rows → See detailed charts for each dimension.
5. (Optional) End a running test by selecting End impact estimate → Confirm.
   
   Note: In Beta, settings cannot be modified mid-test. End and restart if changes are needed.

Statistical Significance Explained

Impact estimates rely on a 99% confidence interval to confirm whether results are meaningful.

• Statistically significant → The observed lift (e.g., higher revenue) is almost certainly due to the identifier, not random chance.
• Not statistically significant → The observed effect could be random noise in the data. This doesn’t rule out impact but shows results aren’t conclusive.

How Lift Is Indicated

Lift compares variant performance to baseline:

• Green lift % → Positive impact when enabling a feature.
• Red lift % → Positive impact when disabling a feature.
• Grey or “insufficient data” with asterisk → Lift was statistically insignificant.
• Green or red with no asterisk → Statistically significant and reliable.

Why Use the Impact Estimate?

Publishers need clarity as cookies deprecate. The Impact Estimate provides data-backed evidence to guide identity strategies.

Key advantages:

• Risk-free testing → Experiments don’t alter live ad setups.
• Optimization insights → See which identifiers improve revenue and eCPM.
• Cookieless readiness → Test strategies for Safari, Firefox, and future Chrome.
• Audience validation → Understand how identifiers affect segmentation and targeting.

Practical Value in Running Impact Estimates

The true strength of the Impact Estimate lies in its strategic value. It’s not just about numbers it helps publishers move forward with confidence.

• Evidence before rollout → No more guesswork; results guide decisions.
• Low-risk environment → Tests don’t disrupt active campaigns.
• Resource focus → Insights highlight where identifiers deliver the most value.
• Advertiser trust → Publishers can share data to prove inventory strength.

PPID vs. First-Party ID

Impact estimates often test identifiers, with PPID and First-Party IDs being the most important. While similar, they serve different purposes.

• PPID (Publisher Provided Identifier) → A Google-supported, encrypted identifier provided by publishers, mainly for Ad Manager. It helps with cross-device recognition, audience segmentation, and frequency capping.
• First-Party ID → A broader category of identifiers created from publisher-owned data such as CRM IDs, hashed emails, or login credentials, useful across systems beyond Ad Manager.

Key Differences Table PPID vs. First-Party ID

Feature	PPID (Publisher Provided ID)	First-Party ID
Origin	Assigned by publisher, encrypted by Google	Generated from publisher’s first-party data
Scope	Limited to one publisher’s ecosystem	Broader use across systems
Use Case	Frequency capping, segmentation, cross-device recognition	Personalization, targeting, measurement
Privacy	Partitioned per publisher, privacy-safe	Depends on publisher’s practices
Integration	Optimized for Ad Manager programmatic workflows	Broader ecosystem integration

Summary: PPID is a specialized identity solution optimized for Ad Manager, while First-Party IDs are a broader identity approach powered by publisher-owned data.

Best Practices for Running Impact Estimates

To maximize the tool’s effectiveness, publishers should:

• Run estimates long enough for meaningful data.
• Test multiple options (PPID, First-Party IDs, no ID).
• Segment results by devices, browsers, and demand sources.
• Re-run estimates as buyer behavior evolves.
• Use results alongside live reporting for validation.

Conclusion

The Impact Estimate (Beta) feature in Google Ad Manager helps publishers navigate the privacy-first future with confidence. By testing identifiers like PPIDs and First-Party IDs in a controlled, risk-free environment, publishers gain valuable insights into performance and prepare their strategies for the cookieless era. With clear data, publishers can optimize monetization, build trust with advertisers, and stay competitive in a changing ecosystem.

FAQ

1. What is Run an Impact Estimate (Beta)?
   It’s a tool in Google Ad Manager that simulates how identifiers affect revenue, eCPM, and impressions.
2. Why is the first week excluded from results?
   Buyers need time to ingest identifiers into their bidding models, so early data isn’t accurate.
3. Can I test multiple identifiers?
   Yes, you can test PPID, First-Party IDs, or both together.
4. Are results guaranteed in live campaigns?
   No, they are predictive but highly reliable based on controlled experiments.
5. What’s the difference between PPID and First-Party ID?
   PPID is a publisher-provided identifier optimized for Ad Manager, while First-Party IDs are broader publisher-owned identifiers usable across different systems.