Video Ad Impact: 2026 Google Ads Incrementality

Understanding the true impact of your advertising spend is no longer a luxury; it’s a necessity. Incrementality testing offers a direct path to measuring the genuine, additional uplift your campaigns deliver, moving beyond correlation to establish clear causation. This is particularly vital for assessing video ad impact, where traditional last-touch attribution often falls short. Are your video ads actually driving new conversions, or are they simply reaching users who would have converted anyway?

Setting Up Your Incrementality Test in Google Ads (2026 Interface)

I’ve seen too many marketers pour money into video campaigns based on vanity metrics. The real question is always, “What would have happened if I hadn’t run that ad?” Incrementality testing answers that. In 2026, Google Ads has refined its testing suite, making it more accessible than ever. This is how we typically set it up for clients focusing on YouTube and Display video campaigns.

1. Accessing the Experiments Section

1. From your Google Ads dashboard, navigate to the left-hand menu.
2. Click on Experiments, which is located under the ‘Tools and Settings’ icon (the wrench symbol).
3. Select New Experiment. You’ll be presented with several experiment types. For measuring video ad impact, we almost always choose ‘Conversion Lift’ or ‘Brand Lift,’ depending on the campaign objective. If you’re focused purely on sales or leads, go with Conversion Lift. For awareness or consideration, Brand Lift is your friend.
4. Give your experiment a clear, descriptive name. Something like “Q3_Video_Ad_Incrementality_Test_CampaignX” works well.

Pro Tip: Don’t just jump in. Before you even touch the interface, define your hypothesis. What do you expect to see? “Video ad X will drive a 5% incremental lift in purchases compared to no video ad exposure.” This clarity will guide your setup and analysis.

2. Configuring Your Experiment Parameters

This is where the rubber meets the road. Incorrect setup here renders your results meaningless. Pay close attention.

1. Select Experiment Type: As discussed, choose Conversion Lift. (For brand awareness, you’d pick Brand Lift, which involves surveys).
2. Choose Campaign(s): Select the specific video campaigns you want to test. You can choose one or multiple, but for cleaner analysis, I recommend starting with a single, well-defined campaign or a small group of highly similar campaigns.
3. Define Your Control Group: This is the critical step for incrementality testing. Google Ads now offers advanced options here.
• Under ‘Experiment Split,’ select Randomized Holdout Group.
• For video campaigns, I strongly recommend a 10% to 20% holdout group. This means 10% to 20% of your target audience will NOT see your video ads, serving as your true control. Going too small (e.g., 5%) might not give you statistically significant results, especially with lower conversion volumes. Going too large can impact overall campaign performance if your ads are indeed incremental.
• Ensure the ‘Control Group’ is set to ‘No Ad Exposure’ or similar. Some platforms allow for ‘exposure to a different ad,’ but for true incrementality, you want no exposure to the tested ad.
4. Set Measurement Period: This is crucial. For most video campaigns, especially those driving lower-funnel conversions, I advise a minimum of 4 weeks. If your conversion cycle is longer (e.g., high-consideration purchases), extend this to 6-8 weeks. Short tests yield noisy data.
5. Specify Conversion Actions: Select the primary conversion actions you want to measure lift for (e.g., ‘Purchases,’ ‘Leads,’ ‘Sign-ups’). Make sure these are accurately tracked in your Google Analytics 4 property and imported into Google Ads.

Common Mistake: Not setting up a true randomized holdout group. Some marketers try to approximate this by pausing ads in certain geos or during specific times, but that introduces too many confounding variables. You need a truly random segment of your target audience that is deliberately excluded from ad exposure.

3. Launching and Monitoring Your Experiment

Once parameters are set, review everything carefully. This is your last chance to catch errors before data collection begins.

1. Review and Create: Google Ads will provide a summary of your experiment. Double-check the budget allocation, target audience, and split percentages.
2. Click Create Experiment. Your experiment will now be live.
3. Monitor Performance: During the experiment, resist the urge to make significant changes to the tested campaigns. Any alterations can invalidate your results.
4. Regularly check the ‘Experiments’ section for initial data. While you shouldn’t make decisions early, it’s good to ensure data is flowing correctly.

Editorial Aside: I once had a client panic two weeks into a four-week test because the control group seemed to be outperforming the test group slightly. They wanted to shut it down. We held firm, finished the test, and the final results showed a clear, statistically significant 8% incremental lift in purchases from the video campaign. Patience is a virtue in causal analysis.

Analyzing Incrementality Results in Google Ads

After your experiment concludes, the real work begins: interpreting the data to understand your video ad impact.

1. Accessing Experiment Results

1. Return to the Experiments section in Google Ads.
2. Click on your completed experiment. Google Ads will automatically generate a report.
3. Focus on the ‘Lift’ metrics. You’ll see figures like ‘Incremental Conversions,’ ‘Incremental Conversion Value,’ and ‘Return on Ad Spend (ROAS) Lift.’

Expected Outcomes: A positive ‘Incremental Conversions’ number means your ads drove more conversions than would have occurred naturally. The ‘ROAS Lift‘ tells you how much additional return you got for the ad spend on those incremental conversions. We aim for a positive ROAS lift, obviously!

2. Interpreting Key Metrics for Causal Analysis

The report will show you a comparison between your test group (exposed to ads) and your control group (not exposed). Here’s what to look for:

• Conversion Rate (Test vs. Control): This is fundamental. If your test group has a significantly higher conversion rate, you’re on the right track.
• Conversions per Impression (Test vs. Control): This metric helps normalize for potential differences in reach or frequency.
• Statistical Significance: Google Ads will often indicate if the observed lift is statistically significant. Look for a confidence level of 90% or higher. If it’s not statistically significant, it means the observed difference could be due to random chance, and you can’t confidently say your ads drove the lift. This is a common challenge, especially with smaller budgets or shorter test durations.
• Cost per Incremental Conversion: This is a powerful metric. It tells you the actual cost to acquire a conversion that would not have happened without your ad. Compare this to your overall CPA.

Case Study: Last year, I worked with a direct-to-consumer apparel brand running YouTube Shorts ads. Their initial Google Ads reports showed a decent ROAS, but we suspected some cannibalization. We set up a 6-week Conversion Lift experiment with a 15% holdout group. The results were revealing: while the overall campaign ROAS was 3.5x, the incremental ROAS was a more modest 1.8x. This meant nearly half of the reported conversions would have happened anyway. We used this data to reallocate budget to other channels, improving overall marketing efficiency by 12% over the next quarter. The shift saved them thousands monthly. According to eMarketer, understanding incremental lift is critical for optimizing ad spend, especially as privacy changes make traditional attribution harder.

3. Actioning Your Insights

The whole point of this exercise is to make better decisions. Based on your incrementality results:

• Scale Up: If your video campaign shows strong, statistically significant incremental lift and a healthy incremental ROAS, you have a strong case for increasing budget.
• Optimize or Pause: If the incremental lift is low or non-existent, it’s a red flag. Consider testing different creatives, targeting, or even pausing the campaign and reallocating budget to more incremental channels.
• Refine Strategy: Use the insights to understand which types of video ads, which audiences, or which placements are truly incremental. This feeds into your broader creative and media strategy.

Remember, incrementality isn’t a one-and-done test. It’s an ongoing process to continually refine your understanding of true ad impact. The market changes, consumer behavior shifts, and your campaigns need to adapt.

Advanced Incrementality Concepts and Tools (Beyond Google Ads)

While Google Ads provides a robust framework, the world of incrementality testing extends further. For a truly holistic view, especially when dealing with multiple channels, you might need more sophisticated approaches.

1. Meta’s Brand Lift Studies and Geo-Lift Experiments

Meta (Meta Business Help Center) offers powerful tools for incrementality, particularly for brand-focused video campaigns. Their Brand Lift studies, often conducted with partners like Nielsen, survey exposed and control groups to measure changes in brand awareness, ad recall, and purchase intent. For conversion-focused tests, their Geo-Lift experiments allow you to run campaigns in specific geographic areas while holding out comparable areas as a control. This requires careful selection of geos to ensure comparability, but it can be incredibly effective for larger advertisers.

2. Multi-Touch Attribution vs. Incrementality

This is a common point of confusion. Multi-touch attribution (MTA) models distribute credit across various touchpoints leading to a conversion. Incrementality, however, asks a different question: “Did the ad cause the conversion at all?” MTA tells you how different channels contributed if a conversion happened. Incrementality tells you if the conversion would have happened without the ad. Both are valuable, but they serve different purposes. I often tell clients to use MTA for tactical optimization within a channel and incrementality for strategic budget allocation across channels.

3. The Role of Marketing Mix Modeling (MMM)

For large organizations with significant spend across numerous offline and online channels, Marketing Mix Modeling (MMM) provides a macro-level view of incrementality. MMM uses historical data to statistically model the impact of various marketing inputs on sales or other outcomes. While not as granular as a direct experiment, it offers insights into the incremental contribution of broad channels (e.g., TV, OOH, digital video) and can inform high-level budget decisions. A recent IAB report highlighted the resurgence of MMM as a privacy-safe measurement solution for strategic planning.

Ultimately, true understanding of your advertising’s effectiveness comes from a combination of these approaches. Start with in-platform tools for campaign-level insights, then layer on broader models as your measurement maturity grows. The goal is always to move from correlation to causation, ensuring every dollar spent works harder.

Embracing incrementality testing shifts your marketing from guesswork to genuine strategic investment. By consistently applying these methods, you’ll uncover the true value of your video ad campaigns and make data-backed decisions that drive measurable growth.