• +770-330-5446
  • rashchris@gmail.com

Tag Archives: Click Fraud

Click Fraud Vs. Viewability: The Difference and Why They Matter

rashchris No Comments

The digital media industry is buzzing with discussions of ‘viewability’ and ‘click fraud,’ but it’s remarkable how few media buyers truly know what these terms really mean, why they are important, and how these factors truly impact marketing spend and return on investment. Lets begin with the basic definitions:

Viewability: The definition of a viewable impression, according to the Internet Advertising Bureau (iab), varies based upon the type of ad unit.

Ad units are considered viewable if:
ŸŸ – Video Ads – 50% of the pixels are in view (on screen) for a minimum of two seconds
Desktop Display Ads – 50% of their pixels are in view for a minimum of one second
Large Format Ads – 30% of pixels in view for one second

Click Fraud: Digital Advertising Fraud is defined as the deliberate (illegal) practice of attempting to fraudulently profit from served ad impressions or reported clicks that are not initiated by a human user.

As you can see, click fraud and viewability are clearly related; a common form of ad fraud called ad stacking is based upon the generation and sale of non-viewable impressions. However, the two are fundamentally different elements, each with its own specific measures and strategic impacts.

Considering Ad Viewability

Much of the discussion today around Desktop Ad Viewability is centered around the relatively low average viewability rates of the different programmatic media channels. Currently, programmaticdesktop display viewability hovers anywhere between 40% and 60%. The iab has set the standard viewability guidelines for Desktop Display (all, not just programmatic) at 70%. So you can see, the programmatic space still has some room for improvement.

Many of you are likely asking, “Why only 70% when my television and radio advertising is 100%”? A large part of this discrepancy is due to limitations of the measurement technologies and not necessarily the quality of the ad units or placements. The Media Rating Council (MRC) has said that “100% viewability is currently unreasonable because current technology can’t consistently measure all online impressions. Different ad units, browsers, ad placements, vendors, and measurement methodologies currently yield wildly different viewability statistics.”

So, with 70% as the target and 50% as the average, why would you even consider programmatic media? Or how should we address these issues in the future?

There are currently two schools of thought on pricing and determining value from programmatic campaigns, knowing that 50% of your purchased inventory is potentially non-viewable, or at least immeasureable:

  1. Pricing for Viewability – Viewable Cost-per-Thousand (vCPM)
  2. Optimizing for Engagement

Pricing for Viewability

This concept is based upon the theory that viewable impressions are the only valuable impressions and thus pricing should be based on that percentage of impressions that meet the iab’s measured viewability standards.

A simple and fairly typical CPM versus vCPM comparison looks like this:

$40,000 Budget $40,000 Budget
$4.00 CPM $8.00 vCPM (assuming 50% viewable inventory)
10,000,000 Impressions 5,000,000 Viewable Impressions

As can be seen above, pricing for viewability guarantees only measureable/viewable impressions, achieving the advertisers stated purpose, but it does not actually increase ‘eyeballs’ seeing the ad units or drive greater value from the campaign because the vCPM price is typically increased in direct proportion to the non-viewable inventory ratio. This perspective misses the potential benefit of non-measureable but viewable impressions and thereby underachieves by comparison.

Optimizing for Engagement

The flip side of pricing out the non-trackable, potentially non-viewable inventory is the proactive realization that non-viewable inventory is already factored into the rate card price of most programmatic media vendors. Knowing this, savvy digital marketers focus their campaigns on optimizing placements for campaign performance metrics that require user engagement (and thus viewable ad units). The thinking in this paradigm is that a campaign, optimized for user engagement, will auto-adjust to reduce the amount of wasted, non-viewable impressions based upon the programmatic decisioning logic that drives the optimization algorithm, essentially gaming the system to achieve better results.

Considering Ad Fraud

Ad fraud is an intentional and illegal activity that is unfortunately rampant in the programmatic media space and as such it has received significant editorial attention over the last few years. We have dedicated our own resources to helping our clients understand the various types of ad fraud that are most common as well as what the Interfuse platform and team of fraud fighters are doing to combat this trend.

For the purposes of this article we will focus less on the types of ad fraud, as it has been adequately covered elsewhere, and more on how fraudulent activity impacts digital media performance.

Non-Human Actions

Perhaps the most pervasive and damaging result of the fraudulent activities in the industry is the effect they are having on traditional benchmark metrics like the Click-Through-Rate (CTR). Lauded for its simplicity and effectiveness, the CTR has been the primary metric of success for display campaigns almost since their inception. However; today’s sophisticated fraudsters can now create “bots” that not only artificially ‘view’ ads but click on them as well, increasing the CTR but denying the advertiser a valid human interaction. This practice is doubly troubling in the programmatic space as this bot-generated click is seen as a success by the basic optimization algorithm and the defrauding site is then rewarded with increased weighting in future ad serving opportunities further compounding the problem.

Fighting this effort requires constant vigilance and proactive human oversight of the algorithmic optimization engine in order to identify and “black-list” or remove suspicious sites (sites delivering unreasonably high CTRs). Unfortunately, many of the programmatic vendors in the space are grossly negligent in this area and instead are passing the artificially inflated CTRs on to client advertisers as valid metrics. This practice is essentially poisoning the well for future, more scrupulous vendors whose carefully curated algorithm consistently delivers a more accurate, but lower, human Click-Through-Rate (hCTR).


"Conversations are Free"

If you'd like to discuss your specific marketing challenges feel free to reach out any time.