It’s (Finally) Time To Stop Data Abuse

AbderrezakKamelThe Sell Sider” is a column written by the sell side of the digital media community.

Today's column is written by Abderrezak Kamel, chief technology officer at Technorati.

One of the largest inefficiencies about the current ad tech setup is the industry’s rapid-fire approach toward data pixels.

In addition to being offensively inefficient, the approach does a terrible job of serving the publishers and harms the experience of the very users that advertisers are trying to reach.

While this is not a new problem, it is getting worse and calls for better standards and enforcements.

The most common pixels being fired are for data collection and user synching. This occurs because data providers and advertising technology platforms collect information to build user models by tapping into the traffic patterns and user data from publishers’ sites. Additionally, all parties want users synced with the maximum coverage to transact with their partners.

However, certain players are abusing the systems by either collecting data without compensating publishers or firing too many pixels without restraint, adversely affecting user experience.

Losing Control

When the primary driver behind pixel firing is user syncing, there's really no reason for a platform to synch a single user more than once per the lifetime of the cookie. When it happens more often, users feel the pinch through increased page load times and script failures. These subsequent syncs also allow other players to access publisher data without being monitored, restricted or audited by the publisher. In addition to the user ID, platforms can access additional data such as screen size, browser, operating system and geolocation. This data is valuable and publishers are giving it away for free.

The biggest problem for publishers is that they have little control over who ends up placing pixels on their page. Publishers often monetize impressions through platforms where they're sold and resold via multiple vendors. These vendors often use any impression as an opportunity to sync with multiple partners, and those partners may sync with others as well, adding additional pixels to each impression. Publishers may have a say with the first vendor but not the rest. The pixel chain can get so long without the publisher realizing it, and it could then have a negative impact on page load times and user experience.

Deteriorating User Experience

Beyond the already snowballing privacy concerns of users, there is the issue of performance of the sites they visit and the total bandwidth being used. The number of calls per impression is crazy. It slows the page down and increases the chance that one of the scripts may fail. In a world where a larger number of users access sites via mobile plans that have data use caps, these calls are using a large chunk of users’ limited data plans over the course of a month.

All these inefficiencies stems from the fact that there is little in the current web standards that have been purposefully designed to enable digital advertising. For most advertising needs, we end up repurposing mechanisms for things they were not designed to do. And when there were standards, they were used to thwart advertising, such as Do Not Track and the Platform for Privacy Preferences (P3P).

The best solution would be user ID standardization through browser manufacturers and a governing body that everybody trusts, perhaps through the IAB, Network Advertising Initiative or Digital Advertising Alliance. Creating, enforcing and rewarding adherence to these standards – similar to mobile’s AdID – would be a great step in the right direction.

Standardization and, more importantly, enforcement of the usage of data pixels is needed now more than ever. Without it, the problem will continue to get worse. The proliferation of data-driven ad platforms and the sprawl of programmatic buying and selling will only continue growing. In addition, users are increasingly consuming their web content through mobile devices.

User ID standardization will go a long way toward improving the overall user experience, creating better opportunities for advertisers, properly compensating publishers for their data and cutting out anyone who continues to abuse the system.

Follow Technorati (@technorati) and AdExchanger (@adexchanger) on Twitter.


Popular On AdExchanger Right Now:

4 Comments

  1. Jay Pinho

    Nice piece! The part about ID universalization is interesting, but I think the problem there is that the players most able to help standardize IDs are the ones with the most to lose by doing so. In the browser world, Chrome is owned by Google, whose competitive advantage is largely due to owning their own massive ID space (across devices). Safari is owned by Apple, and they've started to carve out a niche as the anti-ad-targeting champions (especially with their upcoming Content Blocker capability). Mozilla sells its own ads on the home page as well. The other major player with an ID graph is Facebook, which has no incentive to share that as well. One alternative may be for the many smaller players to band up and start their own data collective to challenge the big dogs.

    Reply
  2. Amen! For the average publisher, there is very little that they can do to curb this abuse. This leads to higher quality publishers limiting access to video for the sake of their reader experience which leaves only lower quality content sites available for advertisers.

    Reply
  3. Very nice article! I work with digital marketing and I know how this is difficult, but I agree with Jay Pinho. The smaller players should to collect their own data. I'm one of these players and I think this is the best alternative!

    Reply
  4. Amen! For the average publisher, there is very little that they can do to curb this abuse. This leads to higher quality publishers limiting access to video for the sake of their reader experience which leaves only lower quality content sites available for advertisers.

    Reply

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>