To create a video experiment in Google Ads, follow these steps:

1. In your Google Ads account, click the Campaigns icon.
2. Click the Campaigns drop-down in the section menu.
3. Click Experiments.
4. Click Video experiments.
5. Click the plus button to start a new experiment.
6. Enter a name and an objective for the experiment.
7. Enter a name for each experiment arm.
8. Click Select campaigns and assign a campaign to each arm. Ensure each campaign has different video ads.
9. (Optional) Add more experiment arms by clicking Add experiment arm. You can include up to four experiment arms.
10. Click the Select a metric drop-down and select one success metric (e.g., CTR, Conversion rate, CPC).
11. Address any insights or recommendations that appear.
12. Click Save.
13. (Optional) If your success metric is 'Brand Lift', click Create Product or Brand in the notification to create a Brand Lift study.

In a Google Ads video experiment, you can select from the following success metrics:

• Absolute brand lift (not available as part of Demand Gen beta)
• Click-through rate (CTR)
• Conversion rate (Conv. rate)
• Cost per conversion (Cost/conv.)
• Cost per click (CPC)
• Cost-per-thousand impressions (CPM)
• Cost per view (CPV)
• Video viewing rate

Directional results and conclusive results differ in their confidence intervals:

• Directional results: You can view the supposed winner of your experimental arms while the experiment is still running. Directional results can be accessed as early as when the experiment reaches a threshold of 70% confidence. You also have the option to wait until it reaches 80% confidence, which is still considered directional.
• Conclusive results: These results are available when the experiment reaches completion at a 95% confidence interval. This means that if you were to repeat the experiment, you would get the same result 95% of the time.

Starting with a hypothesis is crucial for your video experiment because it helps determine your testing strategy and ties directly to your overall campaign goal. For example, a hypothesis like 'Which of these video ads drives the higher conversion rate for my customer acquisition campaign: a 2-minute tutorial or a 15-second direct-offer video?' provides a clear focus and measurable objective for the experiment. This ensures that the experiment is aligned with your marketing goals and provides actionable insights.