Chumbox Quality Quotient

A New Consumer Sentiment Index for Bottom-feeders?

David Carroll
May 3, 2017 · 5 min read

An industry insider, who would prefer to remain anonymous, ran a paid Google Survey and donated it to me for open research. It tested responses to so-called chumbox ads, those grids of sponsored links that bait what industry often calls sourced traffic. This pilot survey tested how people feel about ads from bottom-feeders RevContent and; these are pervasively embedded on hoax-fraud hyperpartisan propaganda “fake news” sites. They supply guaranteed revenue streams for websites, which helps explain their ubiquity across both legit and illegit sites.

Not a real chumbox. But you wouldn’t click on any of those, would you? Via Awl

Presented to consumers as Sponsored Links or From Around the Web this sub-species of adtech goes by other aliases in various sub-industries. They generally prefer content recommendation platforms. When you talk to publishers, I’ve heard them discussed as recirculators because a publisher can re-invest their guaranteed revenue into ads for their articles appearing on other chumboxes elsewhere to draw in audience extensions. Meanwhile, advertisers and agencies tend to refer to these as sourced traffic because they inflate publisher traffic beyond their normal audience, and skew the targeting demographics accordingly.

Just like out there on the world wild web, no one knows exactly how much of the traffic is non-human or as they prefer to say…invalid. Fake users clicking fake ads for fake news are making real money for arbitrageurs. Meanwhile the humans have weird algorithmic encounters with the uncanny. It gets Freudian at times.

The Awl coined the term chumbox in its seminal taxonomy, published before it shutdown its website and outsourced its publishing software to Medium.

Two companies each worth $1B are likely to merge into one company worth $1B.

Since then, business is brisk in the trafficking of soul-searching and moral panic thinkpieces, hot-takes, and ‘anguished long-reads’ about the decline of online advertising and how it supports publishing. As merger and public-offering chatter circulates around the biggest operators, Outbrain and Taboola, Digiday’s Editor-in-Chief, Brian Morrissey, ominously tweets:

So what happens when you click on the bait?

The non-profit conducted a methodical study of the chumbox recirculator sponsored link business and came up with some stark figures. Of the Top 50 news sites, 41 (82%) ran recirculators; 16 used Outbrain (39%), 23 used Taboola (56%), and 1 each used RevContent and Adblade.

They clicked on 312 links and sorted the results into discreet categories.

More than a quarter of the traffic is pure chum. Only 15% appears to be recirculated re-investment by publishers. Smart buyers of premium impressions probably pressure publishers to go light on the sourced traffic.

More than a quarter of the trafficked sites are anonymously registered domains. This is this an indication of how unaccountably ads get arbitraged on these networks.

Enough teasing. Here it is.

On March 3, 2017, a Google Survey ran in the United States and attracted 373 responses with an 11% response rate across 3,285 impressions.

Here’s what the user experience of taking the survey might have been like for the respondents. They were incentivized with “premium content” for taking the survey.

Survey simulation by Google Surveys. Sequences accelerated for illustration purposes.

The respondents graded the quality of three sample RevContent and grids. Less than a majority of respondents willingly admitted that they recognize these as ads. 4-3% of respondents admitted that they found them trustworthy. 4–12% of respondents liked the ads.

The majority of respondents who finished the survey gave the ads one sad star. It went downhill from there.

Here’s data on the sampling bias segmented into age and geography.

Here’s full access to the Google Survey results.

People have strong opinions about chumboxes. Most are appalled by the bottom-feeders of the industry. They serve as a somewhat reliable signal of low-quality publications for more than a third of respondents.

Avoiding the uncanny Freudian byproducts of the clickbait algorithm appears to improve consumer ad quality sentiment for publishers. The third sample earned a higher consumer quality quotient compared to the other two samples which are appreciably weirder and slightly disturbing.

The worst chumbox offenders are the companies that guarantee revenue for toxic publications who profit by propagating hyperpartisian disinformation, hoax-fraud falsification, and counterfeit scam sites.

I’ll be looking for ways to expand this study to compare how consumers rate quality and trust across a broader range of chumbox recirculators.

Questions we might want to ask:

  • What percentage of users can distinguish between Outbrain and RevContent, for example?
  • Is this distinction a reliable signal of publisher quality?
  • What innovations in consumer-centric ad quality rating methods can be iterated from this pilot study?

For a more scholarly study of how users understand sponsored-link content, see Journal of Advertising, Volume 45, Issue 2, 2016 (Wodjynksi & Evans 2015).

Anonymously donating Google Surveys is an interesting concept. You can find me on Twitter or at

David Carroll

Written by

Associate Professor of Media Design at Parsons School of Design @THENEWSCHOOL

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade