YouTube Ads Creative Analysis

Posted by Brian Craft, Satish Shreenivasa, Huikun Zhang, Manisha Arora and Paul Cubre – gTech Data Science Team


Introduction


Why analyze YouTube ads?

YouTube has billions of monthly logged-in users and every day people watch billions of hours of video and generate billions of views. Businesses can connect with YouTube users using YouTube ads, which are promotional videos that appear on YouTube's website and app, with a variety of video ad formats and goals.

Image of a sample YouTube in-stream skippable video ad
A sample YouTube in-stream skippable video ad

The Challenge

An effective video ad focuses on the ABCDs.

  • Attention: Capturing the viewer's attention till the end.
  • Branding: Helping them hear or visualize the brand.
  • Connection: Making them feel something about the brand.
  • Direction: Encouraging them to take action.

But each YouTube ad has a varying number of components, for instance, objects, background music or a logo. Each of these components affect the view through rate (which is referred to as VTR for the remainder of the post) of the video ad. Therefore, analyzing video ads through the lens of the components in the ad helps businesses understand what about the ad improves VTR. The insights from these analyses can be used to inform the creation of new creatives and to optimize existing creatives to improve VTR.


The Proposal

We propose a machine learning based approach for analyzing a company’s YouTube ads to assess which components affect VTR, for the purpose of optimizing a video ad’s performance. We illustrate how to:

  • Use Google Cloud Video Intelligence API to extract the components of each video ad, using the underlying video files.
  • Transform that extracted data to engineered features that map to actionable business questions.
  • Use a machine learning model to isolate the effect on VTR of each engineered feature.
  • Interpret and action on those insights to improve video ad performance, for instance altering existing creatives or create new creatives to be used in an AB test.

Approach


The Process

The proposed analysis has 5 steps, discussed below.

1. Define Business Questions
Align on a list of business questions that are actionable, for instance “does having a logo in the opening shot affect VTR?” We suggest taking feasibility into account ahead of time, for instance if a product disclaimer is necessary to have for legal reasons, there is no reason to assess the impact a disclaimer has on VTR.

2. Raw Component Extraction
Use Google Cloud technologies, such as the Google Cloud Video Intelligence API, and underlying video files to extract raw components from each video ad. For instance, but not limited to, objects appearing in the video at a particular timestamp, presence of text and its location on the screen, or the presence of specific sounds.

3. Feature Engineering
Using the raw components extracted in step 2, engineer features that align to the business questions defined in step 1. For example, if the business question is “does having a logo in the opening shot affect VTR”, create a feature that labels each video as either 1, having a logo in the opening shot or 0, not having a logo in the opening shot. Repeat this for each feature.

4. Modeling
Create an ML model using the engineered features from step 3, using VTR as the target in the model.

5. Interpretation
Extract statistically significant features from the ML model and interpret their effect on VTR. For example, “there is an xx% observed uplift in VTR when there is a logo in the opening shot.”


Feature Engineering


Data Extraction

Consider 2 different YouTube Video Ads for a web browser, each highlighting a different product feature. Ad A has text that says “Built In Virus Protection'', while Ad B has text that says “Automatic Password Saving”.

The raw text can be extracted from each video ad and allow for the creation of tabular datasets, such as the below. For brevity and simplicity, the example carried forward will deal with text features only and forgo the timestamp dimension.

 Ad

 Detected Raw Text

 Ad A

 Built In Virus Protection

 Ad B

 Automatic Password Saving

Preprocessing

After extracting the raw components in each ad, preprocessing may need to be applied, such as removing case sensitivity and punctuation.

 Ad

 Detected Raw Text

 Processed Text

 Ad A

 Built In Virus Protection

 built in virus protection

 Ad B

 Automatic Password Saving

 automatic password saving

Manual Feature Engineering

Consider a scenario where the goal is to answer the business question, “does having a textual reference to a product feature affect VTR?”

This feature could be built manually by exploring all the text in all the videos in the sample and creating a list of tokens or phrases that indicate a textual reference to a product feature. However, this approach can be time consuming and limits scaling.

Image of pseudo code for manual feature engineering
Pseudo code for manual feature engineering

AI Based Feature Engineering

Instead of manual feature engineering as described above, the text detected in each video ad creative can be passed to an LLM along with a prompt that performs the feature engineering automatically.

For example, if the goal is to explore the value of highlighting a product feature in a video ad, ask an LLM if the text “‘built in virus protection’ is a feature callout”, followed by asking the LLM if the text “‘automatic password saving’ is a feature callout”.

The answers can be extracted and transformed to a 0 or 1, to later be passed to a machine learning model.

 Ad

 Raw Text

 Processed Text

 Has Textual Reference to Feature

 Ad A

 Built In Virus Protection

 built in virus protection

 Yes

 Ad B

 Automatic Password Saving

 automatic password saving

 Yes



Modeling


Training Data

The result of the feature engineering step is a dataframe with columns that align to the initial business questions, which can be joined to a dataframe that has the VTR for each video ad in the sample.

 Ad

 Has Textual Reference to Feature

 VTR*

 Ad A

 Yes

 10%

 Ad B

 Yes

 50%

*Values are random and not to be interpreted in any way.

Modeling is done using fixed effects, bootstrapping and ElasticNet. More information can be found here in the post Introducing Discovery Ad Performance Analysis, written by Manisha Arora and Nithya Mahadevan.

Interpretation

The model output can be used to extract significant features, coefficient values, and standard deviation.

Coefficient Value (+/- X%)
Represents the absolute percentage uplift in VTR. Positive value indicates positive impact on VTR and a negative value indicates a negative impact on VTR.

Significant Value (True/False)
Represents whether the feature has a statistically significant impact on VTR.

 Feature

 Coefficient*

 Standard Deviation*

 Significant?*

 Has Textual Reference to Feature

0.0222

0.000033

True

*Values are random and not to be interpreted in any way.

In the above hypothetical example, the feature “Has Feature Callout” has a statistically significant, positive impact of VTR. This can be interpreted as “there is an observed 2.22% absolute uplift in VTR when an ad has a textual reference to a product feature.”

Challenges

Challenges of the above approach are:

  • Interactions among the individual features input into the model are not considered. For example, if “has logo” and “has logo in the lower left” are individual features in the model, their interaction will not be assessed. However, a third feature can be engineered combining the above as “has large logo + has logo in the lower left”.
  • Inferences are based on historical data and not necessarily representative of future ad creative performance. There is no guarantee that insights will improve VTR.
  • Dimensionality can be a concern as given the number of components in a video ad.

Activation Strategies


Ads Creative Studio

Ads Creative Studio is an effective tool for businesses to create multiple versions of a video by quickly combining text, images, video clips or audio. Use this tool to create new videos quickly by adding/removing features in accordance with model output.

Image of sample video creation features in Ads creative studio
Sample video creation features in Ads creative studio

Video Experiments

Design a new creative, varying a component based on the insights from the analysis, and run an AB test. For example, change the size of the logo and set up an experiment using Video Experiments.


Summary


Identifying which components of a YouTube Ad affect VTR is difficult, due to the number of components contained in the ad, but there is an incentive for advertisers to optimize their creatives to improve VTR. Google Cloud technologies, GenAI models and ML can be used to answer creative centric business questions in a scalable and actionable way. The resulting insights can be used to optimize YouTube ads and achieve business outcomes.


Acknowledgements

We would like to thank our collaborators at Google, specifically Luyang Yu, Vijai Kasthuri Rangan, Ahmad Emad, Chuyi Wang, Kun Chang, Mike Anderson, Yan Sun, Nithya Mahadevan, Tommy Mulc, David Letts, Tony Coconate, Akash Roy Choudhury, Alex Pronin, Toby Yang, Felix Abreu and Anthony Lui.

Stable Channel Update for Desktop

The Stable channel has been updated to 120.0.6099.234 for Mac and 120.0.6099.224 for Linux and 120.0.6099.224/225 to Windows which will roll out over the coming days/weeks. A full list of changes in this build is available in the Log.


The Extended Stable channel has been updated to 120.0.6099.234 for Mac and 120.0.6099.225 for Windows which will roll out over the coming days/weeks.


Security Fixes and Rewards

Note: Access to bug details and links may be kept restricted until a majority of users are updated with a fix. We will also retain restrictions if the bug exists in a third party library that other projects similarly depend on, but haven’t yet fixed.


This update includes 4 security fixes. Below, we highlight fixes that were contributed by external researchers. Please see the Chrome Security Page for more information.


[$16000][1515930] High CVE-2024-0517: Out of bounds write in V8. Reported by Toan (suto) Pham of Qrious Secure on 2024-01-06

[$1000][1507412] High CVE-2024-0518: Type Confusion in V8. Reported by Ganjiang Zhou(@refrain_areu) of ChaMd5-H1 team on 2023-12-03

[$TBD][1517354] High CVE-2024-0519: Out of bounds memory access in V8. Reported by Anonymous on 2024-01-11


We would also like to thank all security researchers that worked with us during the development cycle to prevent security bugs from ever reaching the stable channel.


Google is aware of reports that an exploit for CVE-2024-0519 exists in the wild.

As usual, our ongoing internal security work was responsible for a wide range of fixes:

[1518006] Various fixes from internal audits, fuzzing and other initiatives


Many of our security bugs are detected using AddressSanitizer, MemorySanitizer, UndefinedBehaviorSanitizer, Control Flow Integrity, libFuzzer, or AFL.




Interested in switching release channels? Find out how here. If you find a new issue, please let us know by filing a bug. The community help forum is also a great place to reach out for help or learn about common issues.




Srinivas Sista
Google Chrome

Launch the Lucidspark whiteboarding app directly from Google Meet Series One Board 65 and Desk 27 devices

What’s changing 

In September 2023, we announced the next phase of digital whiteboarding for Google Workspace. At that time, we announced that new whiteboarding tools would soon be integrated with Series One Board 65 and Desk 27 devices. By the end of the month, Lucidspark by Lucid Software can be launched directly from these devices. With this integration, users will be able to share and participate in a Lucidspark whiteboard session in a Meet call, either initiated from the Series One Board 65 and Desk 27 or a remote participant on the call.



Who’s impacted

Admins and end users


Why you’d use it 

Integrating third-party collaboration tools is a key component of the new digital whiteboarding experience in Google Workspace. Specifically, we’ve heard from our customers that whiteboarding tools like Lucidspark by Lucid Software help their teams work better together with features like infinite canvas, templates, and diagramming. Today, Lucidspark can already be launched inside a Meet call from your laptop. Now, you also have the option to use Lucidspark on the Board 65 and Desk 27 devices within a Meet call.


Later this year we’ll introduce support for launching FigJam and Miro directly from Board 65 and Desk 27 devices. We’ll provide an update here on the Workspace Updates blog with more information at that time. For more information on the future of digital whiteboarding in Workspace, please refer to our initial announcement.


Getting started


Rollout pace


Availability

  • Available to all Google Workspace customers with Google Meet Board 65 and Desk 27 devices


Google Workspace Updates Weekly Recap – January 12, 2024

4 New updates

Unless otherwise indicated, the features below are available to all Google Workspace customers, and are fully launched or in the process of rolling out. Rollouts should take no more than 15 business days to complete if launching to both Rapid and Scheduled Release at the same time. If not, each stage of rollout should take no more than 15 business days to complete.


Version history limits for Apps Script projects 
For all new scripts, you’ll be able to create and save up to 200 versions of your script. If needed, you can permanently delete a script version from the project history page. | This is available now to all Google Workspace customers. | Learn more using our developer documentation on working with Apps Script versions. 


Share a link to a specific time in a Google Drive video 
We’re adding new functionality to the Drive sharing button that lets you share timestamped links to specific parts of a video. On web, simply navigate to drive.google.com > find and open a video file > play the video (you can pause the playback before performing the following steps) > select the dropdown on the “Share” button in the top-right corner > select “Copy link to this time” > send the link. | Rolling out now to Rapid Release and Scheduled Release domains. | Available to all Google Workspace customers and users with personal Google Accounts. | Learn more about copying a specific time in the video. 
Share a link to a specific time in a Google Drive video

Introducing dropdown options on the sharing button in Google Docs, Sheets, Slides and Drawings We’re adding a new feature that ensures a seamless sharing experience across Workspace. In Google Docs, Sheets, Slides and Drawings, you will now see a dropdown on the Share button that surfaces quick actions, such as pending access requests and the “Copy link” option. | Rolling out now to Rapid Release and Scheduled Release domains. | Available to all Google Workspace customers and users with personal Google Accounts. 
dropdown options on the sharing button

Using functions in Connected Sheets for BigQuery 
Today, Connected Sheets for BigQuery supports 23 Sheets functions, such as AVERAGE and XLOOKUP. However, all of these functions behave somewhat differently than their native counterparts. Thus, to help Connected Sheets users write better functions, we now display context-aware Help Center content in Sheets. The ‘formula help’ shows descriptions for Connected Sheets functions when writing a formula that would query BigQuery, and otherwise shows descriptions of native Sheets functions. | This is available now to all Google Workspace customers and users with personal Google Accounts. | Learn more about the XLOOKUP function.
Using functions in Connected Sheets for BigQuery


Previous announcements

The announcements below were published on the Workspace Updates blog earlier this week. Please refer to the original blog posts for complete details.


Updates to metrics in Google Drive Apps Reports and Reports API 
We’re making some updates to the Google Drive metrics in the Admin Console Apps reports and the Reports API. As a result of these improvements, admins who analyze metrics will have more reporting clarity and can better understand activity trends within their domain. | Learn more about metrics in Drive Apps Reports and Reports API. 

Easily share Google Drive files to Google Calendar meeting attendees 
We’re introducing the option to share any file with all meeting participants on a Google Calendar invite via the sharing dialog within a file. | Learn more about sharing Drive files to Calendar. 

Google Meet is now available on Logitech Android appliances 
Google Meet is now supported on Logitech’s Rally Bar and Rally Bar Mini Android-based appliances for collaboration rooms and spaces of just about any size. After initial setup, admins can easily enroll, manage, and monitor these devices using the Google admin console. Google Meet on Logitech Android-based devices is supported on CollabOS v1.11 as a video conferencing provider | Learn more about Meet on Logitech Android appliances. 

Google Meet hardware devices from Poly now support interoperability with Cisco Webex and Zoom 
We’re expanding the existing interoperability between Google Meet, Cisco Webex, and Zoom to include Android-based Meet hardware devices from Poly. Specifically, these devices include: Poly Studio X30, X50, X52, and X70. | Learn more about support interoperability with Cisco Webex and Zoom.

Extending Trusted Types to Gmail
We’re excited to announce the expansion of Trusted Types to Gmail. This will provide a defense against DOM XSS and further enhances our advanced data protection controls to keep users and data safe across more of the apps they use everyday. | Learn more about Trusted Types.


Completed rollouts

The features below completed their rollouts to Rapid Release domains, Scheduled Release domains, or both. Please refer to the original blog posts for additional details.

Rapid Release Domains: 

Scheduled Release Domains: 

Rapid and Scheduled Release Domains: 

For a recap of announcements in the past six months, check out What’s new in Google Workspace (recent releases). 

Extending Trusted Types to Gmail

What’s changing

Last year, we improved the client-side security of Google Docs, Sheets, Slides, Forms, Sites, Drawings, Drive, and Calendar with Trusted Types. This browser-based runtime feature limits the uses of Document Object Model (DOM) APIs that are used by the apps listed above or third-party extensions. Trusted Types also reduce the possibility of Document Object Model Cross Site Scripting (DOM XSS), which continues to be one of the most critical threats to web security. 

DOM XSS occurs when a cyber attacker injects malicious code into a web page, which can then be executed by the victim's browser. This can allow the cyber attacker to steal cookies, hijack sessions, and even take control of the victim's computer. 

To defend against this, we’re excited to announce the expansion of Trusted Types to Gmail. This will provide a defense against DOM XSS and further enhances our advanced data protection controls to keep users and data safe across more of the apps they use everyday. 


Who’s impacted 

Developers (relying on any Chrome extensions that modify DOM APIs.) 


Additional details 

This new enforcement mode will require third-party extensions to use typed objects instead of strings when assigning values to DOM APIs. Once Trusted Types are fully enforced, the Trusted Types directive will be present in the Content Security Policy (CSP) header: 

Content-Security-Policy: require-trusted-types-for 'script';report-uri https://mail.google.com/mail/cspreport 


Getting started 

  • Admins: There is no admin control for this feature. 
  • Developers: 
    • To make code Trusted Types compliant, signal to the browser that data being used within the context of these DOM APIs is trustworthy by creating a Trusted Type special object. 
    • There are several ways to be Trusted Types compliant, such as removing the offending code, using a library (such as safevalues or DOMPurify), or creating a Trusted Types policy. To ensure a seamless experience for users, we recommend employing these techniques before Trusted Types enforcement is rolled out. Failure to make code Trusted Types compliant may cause feature breakages for third-party extensions as their DOM manipulations will be blocked by the browser. 
  • End users: There is no end user setting for this feature. 

Rollout pace 


Availability 

  • Available to all Google Workspace customers and users with personal Google Accounts 

Resources 

MLK Day and the pursuit of equitable internet access

As Martin Luther King Jr. Day approaches, we are reminded of the ongoing pursuit for equity, justice, and the civil rights leader’s legacy and vision of a world with equal access to opportunities. Google Fiber aspires to be a part of this dream, by helping to bridge the digital divide and foster inclusivity in our increasingly connected world. 


Everyone deserves fast, reliable internet at an accessible price, and the ability to put that internet connection to use - to connect to opportunity. We’re grateful to work with many organizations across the country that put that principle to work every day, helping their clients and constituents get more out of their lives, both online and beyond.  


Thumbnail


Here are a few ways our incredible community partners are marking this important day across the country:





GFiber is proud to be a small part of these efforts, and others, working towards a more equitable and just world.


Of course, there is still more work to be done. We will continue to push forward to make the internet more accessible and to help others harness the power and opportunity of the internet. You can help too! One small way to act right now  — the Affordable Connectivity Program helps make high speed internet available to millions of US households through a $30 a month subsidy, but this program will end in April 2024 unless Congress acts to allocate additional funding. There is a bill before Congress to extend funding for this critical program. Take a moment to let your Congresspeople add your voice on the Affordable Connectivity Program Extension Act of 2024 to keep that connection strong for everyone.


Subscribe to get the GFiber Blog in your email.


Posted by Jess George, Head of Digital Equity & Community Impact