X

The Oracle Data Cloud blog highlights the latest data-driven insights and trends in digital marketing and ad tech.

Brand safety has evolved, and so should your ad blocking strategy

At its core, brand safety is a set of measures that protects a brand from appearing in unsafe or unsuitable environments. It continues to be a top priority among brands and publishers, but one of the key challenges is that brand safety is not one-size-fits-all. What is considered "brand-suitable" to BuzzFeed may not be considered the same for Clorox, making brand safety a tricky business. 

While ad verification services have made it easier to protect brands by safeguarding digital ad investments, understanding when to use each service is key to making advertising more accountable and effective. 

What is post-bid blocking? 

Blocking is a general term that is used throughout ad tech, and it can mean a number of different things–from pre-bid segments on demand-side platforms (DSPs) that prevent bidding on invalid traffic (IVT) and brand-unsafe impressions to post-bid solutions that stop an ad from rendering based on real-time brand protection technology. 

At Oracle Data Cloud, we use the term post-bid blocking to refer to real-time decisioning, after the bid has already been won, which then prevents an ad creative from appearing in an environment the advertiser wants to avoid. 

How does post-bid blocking work? 

To set up post-bid blocking, an advertiser generally needs to “wrap” their creative tags with blocking wrapper tags. These tags are then uploaded into an advertiser’s ad server or DSP so, instead of loading the creative directly onto a page, the wrapper tag loads first and can run a series of checks to determine whether the creative should be loaded at all. 

These checks vary by vendor but we protect impressions using four criteria:  

  1. brand safety (both standard and custom segments) 

  1. domain blocklists or allowlists  

  1. geographic area 

  1. IVT detection 

An advertiser can choose to block on any or all of those criteria. (And soon, they will be able to set separate criteria on an individual campaign or line item level as well.) For example, an airline may want to avoid all standard brand-unsafe categories as well as a custom segment that comprises any pages that mention the Boeing 737 Max 8 plane. A bank or financial services company may want to block any users that are not within the geographic areas served by their products, and so on. 

For display ads that are blocked, the ad slot may be left empty or replaced with a generic image containing no reference to the brand. For blocked video ads, the video is simply skipped, and the player moves onto the next asset in the queue. The net effect across both formats is the same: your brand is protected from showing up where it shouldn’t and your Moat Analytics dashboard reports out on how many impressions were blocked under each of the criteria. 

 

Is it better to use pre-bid targeting, post-bid blocking, or both? 

The short answer: it depends. It’s important to keep in mind that the use cases and objectives for pre-bid targeting (even exclusion targeting, which some in the industry refer to as “blocking”) and post-bid blocking are quite different. 

Pre-bid targeting (including exclusion segments such as high-IVT or brand-unsafe pages) is primarily designed to increase ad budget efficiency. The segments are generally set up within a DSP and allow advertisers to set parameters around their campaigns or line items in order to bid only on the most relevant incoming requests. 

Because this targeting is applied prior to a bidding decision (and thus before any ad has loaded on the page), it is inherently probabilistic: you can’t know with certainty, for example, if an impression will be MRC (Media Rating Council) viewable before it happens. There will always be some degree of false positives and negatives, even with sophisticated modeling. Advertisers understand that there is a built-in level of uncertainty in pre-bid segmentation. It is still an extraordinarily valuable tool for decreasing budget waste and ensuring that ad dollars are allocated to the most promising opportunities.  

Post-bid blocking, on the other hand, occurs downstream from pre-bid segments and is primarily deterministic: it occurs after the impression has been won and thus the blocking decision is based upon real-time signals. In pre-bid, for example, IVT avoidance is based on historical patterns observed against web pages and users; in post-bid blocking, however, IVT is detected in real-time and a blocking decision is thus made based on what is happening in the moment. Similarly, if the DSP doesn’t have reliable data on the URL of the incoming bid request, it will not be able to apply pre-bid targeting correctly; in a post-bid environment, the URL is detected from JavaScript running directly on the page. 

Another important distinction lies in where these products are generally used. Pre-bid is almost exclusively applied within programmatic environments: this is why pre-bid segments are most often found within a DSP’s user interface. Post-bid blocking can be applied anywhere -- not just on programmatic buys but on direct ones, too. If an advertiser is working directly with a high-quality news publisher, for example, they may still wish to avoid certain news topics or even users from a certain geographic location. In this case it can be useful to apply post-bid blocking even when pre-bid segments are unavailable. 

Finally, typically an advertiser’s reasons for using pre-bid and post-bid blocking differ. Pre-bid segmentation is primarily used to spend ad budgets as efficiently as possible. Post-bid blocking may save money (because the advertiser’s ad server is never called on blocked impressions), but the primary motivator for post-bid blocking is less about efficiency and more about protecting the brand’s carefully-cultivated image in a dynamic online environment. 

 

How should I choose my brand’s strategy in relation to pre-bid segmentation and post-bid blocking? 

Every marketer should apply these two steps to building more effective brand protection strategies: 

1. Define your objectives. 

Defining which objectives are most important to your strategy will help you identify which solution to leverage. 

Objective #1 – Maximize programmatic investments 

Solution: Use pre-bid targeting to avoid buying fraudulent impressions and allocate advertising dollars to the most promising opportunities 

Objective #2 – Avoid unsuitable and unsafe environments 

Solution: Use post-bid blocking to safeguard your brand from unsafe, unsuitable environments and gain better insights to optimize campaigns 

2. Ensure your partner’s capabilities meet your high standards. 

Before you execute, do your due diligence in whose technology you are employing.  

A good pre-bid solution should have: 

  • Low level of false positives and negatives 

  • Sophisticated modeling 

  • High scale 

A good post-bid solution should have: 

  • A built-in infrastructure for detecting both GIVT and SIVT across platforms 

  • Customizable, robust brand safety blocking rules that can be modified at any time without re-trafficking tags (even mid-campaign) 

  • The ability to set blocking criteria for geographic areas and domain lists 

How can I use post-bid blocking? 

If you are an advertiser looking to protect your brand from brand-unsafe and inappropriate environments, IVT, and incorrect geographic targeting, contact your Oracle Data Cloud account manager today to get post-bid blocking set up. 

To get more information about protecting your ad spend, read our white paper on invalid traffic.

***

Author: Jay Pinho

Jay is a principal product manager at the Oracle Data Cloud focused on pre-bid and brand safety solutions. 

Join the discussion

Comments ( 1 )
  • Andi Herlin Wednesday, December 18, 2019
    Thank u
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.