Brand Safety in CTV Advertising

Part of the blog post series: CTV Advertising Standards: IAB Tech Lab Standards & Guidance

Executive Summary:

Brand safety is an important aspect of ad delivery, but is increasingly complex.
Tech Lab enables brand safety through features in Content Taxonomy, OpenRTB, VAST and Open Measurement.

Company	Display	Video	Mobile Web	Mobile Inapp
Cyber Communications Inc	Impressions Clicks	-	Impressions Clicks	-

The Details:

Brand safety has been highlighted in 2020 as a result of COVID, protests, elections, and more, but it has been a concern for advertisers for a long time. In fact, one of the attractions of digital advertising is the ability to provide marketers with controls that granularly manage the context in where their ads appear. This includes keyword-based exclusion/inclusion and sentiment analysis tools.

Brand safety (and its lesser known but equally important sibling, “brand suitability“) are difficult technical challenges due to the complexity of understanding sentiment, as well as the sheer volume of content that needs to be processed, much of it in real time. Brand Safety vendors in the space are working to resolve these problems. At Tech Lab, we are focussed on building the plumbing that supports the industry: the ability to articulate the concepts, communicate the required metadata, and take actions based on the data.

The brand safety features within Tech Lab’s standards fall under three main groups:

1. Categorization:

The Content Taxonomy provides a standard way to describe site/app content. Its use allows marketers to work across various platforms and vendors, and consistently articulate the types of content they want / do not want their ads to appear on, through a “common language“. The richness of the taxonomy allows flexibility in how granularly (or not) the controls can be configured. The categories in the taxonomy can be used both to target specific types of content or to define suitability restrictions on various content categories.

In addition, we are currently working with the Global Alliance for Responsible Media (GARM) & the 4As/APB (Advertiser Protection Bureau) to include an update to the Content Taxonomy to incorporate the concepts of “Brand Safety Floor” and “Brand Suitability”. This will help marketers take their concepts of brand values and risk tolerance, and apply them consistently across platforms.

Finally, we are working on an “Ad Product Taxonomy” that will enable a consistent categorization of the ad creatives — what’s being advertised in the ad itself — and help “match” the ad to the right content.

2. Communication

VAST4, OpenRTB, and Open Measurement all enable brand safety checks by providing mechanisms to pass the page URLs, categories, and other metadata about the content of an impression. This basic metadata is necessary for brand safety vendors to analyze content and help buyers make brand safety/suitability decisions.

It is important to note that challenges with post-bid processing on CTV devices mean that publishers and tech vendors should do their best to support pre-bid scenarios where possible. We have heard from members that the percentage of content that can be checked in a pre-bid manner (across all platforms) is very low. We recommend that this be addressed by the industry — a pre-bid solution might be a cleaner way to address brand safety/suitability without having to deal with potentially messy post-impression ad blocking and reconciliation.

3. Action

In many cases, a brand safety decision can only be reached post-bid, when an ad is being delivered. In cases where content is considered unsuitable for the brand, buyers prefer to stop the ad from being rendered. The ability to block or swap ads that were deemed “unsuitable” has historically been done either via proprietary methods, or via VPAID for video ads (only on the Web). As we encourage eliminating VPAID (for efficiency and safety reasons), the Open Measurement team is working on supporting this in a clean and transparent manner. This is not a capability typically available on mobile or CTV at the moment, but is worth being aware of for the longer term.

In addition to the above, a final brand safety consideration is ad fraud. Advertisers do not want to have their ads show up in channels that are fraudulent or associated with malware — and none of us should want to fund fraudulent and misleading content. We will however discuss that under the anti-fraud section of this series.