Limited Time Offer: Save 30% on PRO with code GETRESULTS »

Real-World Education for Modern Marketers

Join Over 600,000 Marketing Professionals

Start here!
Text:  A A

Skip the Cookies: Value-per-Impressions Analysis Provides Better Data

by Jim Moar  |  
October 8, 2013

Many advertisers have a love-hate relationship with third-party tracking cookies. At best, cookies are a reliable way for marketing and advertising professionals to create targeted ad campaigns that help them reach particular audience segments based on ideal behavioral characteristics.

However, cookies crumble when applied to one of today’s digital advertising imperatives: measuring the cross-channel value of advertising. Today’s advertisers are all too aware of the need to go beyond direct, in-channel measurement of ad campaigns to include an understanding of their advertising’s cross-channel effects. Only by looking through both of these lenses can an advertiser can understand the actual value of his ads and optimize his ad spend accordingly.

Cookies, which are used to track individual purchase paths across the web, are the foundation upon which attribution schemes attempt to measure cross-channel ad value. The problem is that third-party cookies cannot track across devices---a fatal flaw in today’s multi-device world. With the increasing likelihood that any given person is browsing the Web on a tablet, looking through specific product catalogs on a mobile phone, and ultimately making a purchase on a laptop, cookies can’t keep up. In fact, Google’s ad sales chief recently reported that 90% of today's consumers shift seamlessly between devices to complete a task, with 500% more people simultaneously using multiple devices than were three years ago. Also, third-party cookies are themselves under increasing fire, with a growing portion of the browser market blocking them by default. Plus, rumors abound that Google may even be planning to replace cookies altogether.

The bottom line is that the purchase paths cookie-based attribution schemes deliver are fraught with potholes and pitfalls. It’s a risky proposition to base ad spend decisions on such a shaky foundation.

Reach for VPI Instead of Cookies

Instead of relying on incomplete cookie paths, marketers can and should instead measure value by using data that spans devices---data that, in other words, provides a solid foundation upon which to make sound ad bidding and buying decisions. Ad-level impression data, sitting reliably right at the fingertips of every digital advertiser, fits the bill. Impression data focuses on the number of people exposed to any one ad, rather than futilely trying to tie together one person’s journey through a variety of ads on different channels and devices. It’s a natural source for measuring how advertising in one channel, such as display, impacts sales through another channel like search, where consumers go with the intention of finding product info and buying.

By identifying the relationship between a particular display ad’s impression volume and sales through a particular search keyword, advertisers can determine the value of each ad, measured as value per impression (VPI). VPI is calculated through modeling at the atomic (individual ad and keyword) level that first determines whether a particular display ad’s impression volume is one of the top variables driving a particular keyword’s predicted performance and then measures the magnitude of the impact.

VPI analysis can be done for each and every display ad, as well as other brand-centric ads such as those on Facebook or Twitter. It delivers what cookie-based attribution cannot: a meaningful and actionable understanding of a particular ad’s cross-channel value across devices. It gives advertisers a metric that can be applied across all ads in their mix to help them understand which ads are truly valuable... and which are not. Moreover, it can be translated directly into buying and bidding terms, guiding advertisers on precisely how much to invest in each ad to optimize their spend for maximum financial return across channels.

When measuring cross-channel value, cookies have been leading marketers down the wrong path, one that leads to an incomplete, broken set of data upon which to base ad investment decisions. The good news is that VPI analysis can provide a complete view of ad value across devices and channels, which in turn enables smarter ad investment and better-than-ever returns.

Sign up for free to read the full article.Read the Full Article

Membership is required to access the full version of this how-to marketing article ... don't worry though, it's FREE!


We will never sell or rent your email address to anyone. We value your privacy. (We hate spam as much as you do.) See our privacy policy.

Sign in with one of your preferred accounts below:

Jim Moar is CEO of OptiMine. He has held several CEO and COO positions in technology and industrial companies, both public and private, accumulating over 25 years of top-level experience. Jim most recently served as CEO of Reshare Commerce, a start-up SaaS company. Previously, he was COO and President of Identix, a global leader in biometric systems, where he led a major turnaround doubling revenue to $90M.

Rate this  

Overall rating

  • Not rated yet.

Add a Comment


  • by Tom Metro Sun Dec 22, 2013 via blog

    So if I'm following this article correctly, the original goal was to use cookies to track individual users, as they roamed around the web, so you could 1. avoid counting the same user multiple times, 2. build a profile of the user's interests inferred from the collection of sites they visit, and 3. collect various metrics, like how many exposures a typical user needs to an ad before they take action.

    But now marketers are saying that's too hard to pull off for technical reasons, so instead they gave up, and are now simply tracking the volume of ad impressions (it sounds like without filtering out repeat impressions to the same user) and are measuring the success of a campaign based on its impressions to sales ratio.

    "When measuring cross-channel value, cookies have been leading marketers down the wrong path, one that leads to an incomplete, broken set of data upon which to base ad investment decisions."

    Yes, but that's not because you couldn't pull it off for technical reasons, but because way more significance was given to the sorts of information you could extract from such tracking.

    One of the first problems marketers ran into was treating cookies like they were a precise measuring tool for examining the behaviors of individuals, rather than treating them as a tool for statistical sampling. (Back when cookies were first introduced, all the talk was that they'd be used sporadically for statistical sampling.)

    For example, if out of 10,000 ad impressions, you served up only 1000 with tracking cookies, and the data showed that your ad was 10 times more apt to be clicked on if the user previously visited an auto parts web site, then you've inferred some useful macro behavior. On the other hand, trying to control things at the ad network level and customize the ads shown to a specific user based on their specific browsing history, then I think the accuracy becomes far more tenuous.

    You can clearly see where social networks are going to have an advantage when it comes to this sort of tracking and profiling. Not only will they have a much deeper and more accurate understanding of the user's preferences, but because they encourage the user to maintain a logged-in state on all platforms they use, they can track user behavior everywhere, thanks to the "+" and "Like" buttons appearing on most sites.

    However, anecdotally, Facebook seems to be doing a poor job of leveraging this. I frequently hear talking heads on tech shows describe the poorly targeted ads appearing on their FB pages. The ads are definitely targeted - they reflect some signal (a purchase, a like) - but they promote a product the viewer is no longer or never was interested in. So despite all the rich data, it sounds like it is still hard to interpret it in an effective way.
    (Can't say too much about G+ yet, as they're still building their network, and don't seem to be trying to use it for generating revenue yet.)

    "The good news is that VPI analysis can provide a complete view of ad value across devices and channels..."

    Calling a metric that you've had since the first days of online advertising, and have always had, something that gives you a "complete view" is a joke. At best, it is simply a metric that is less misleading and less prone to marketers inferring greater meaning on it than it deserves.


MarketingProfs uses single
sign-on with Facebook, Twitter, Google and others to make subscribing and signing in easier for you. That's it, and nothing more! Rest assured that MarketingProfs: Your data is secure with MarketingProfs SocialSafe!