Why Wasting Impressions Is Good For Performance

joel-nierman“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by Joel Nierman, marketing and media director at Critical Mass Chicago.

Improving the targeting of advertising has been a focus for the ad industry ever since John Wanamaker implied that it was a bad thing that he wasted half his money on advertising (he wasn’t sure which half).

This improved targeting movement has become hypercharged in the last decade or so because of digital ad tech. The problem with this march to waste elimination is that it is a self-fulfilling prophecy – eliminating waste will eventually eliminate everything.

John Wanamaker probably didn’t realize the impact his musing would have on the advertising industry a century or so after he said it. While not knowing which half works is a clever talking point, in reality it instilled a maniacal devotion to eliminating waste in every advertising professional; now the vast majority of the ad industry creates value for clients by reducing waste through improved targeting.

All the standard practices for buying ads hold this ideal at the core: Buying media against indexes is designed to minimize the number of people not in the target group from seeing a particular ad. Gross rating points (GRP) explicitly do not pay for waste. And practically every company on the Lumascape is focused on using data to target ads more precisely or to ensure they are targeted more precisely, which, said another way, is reducing waste.

And this makes logical sense. Brands and companies know who buys their products and services, and finding more people like their current customers – and fewer of the people not like current customers – is a logical way to build incremental business.

But there is a flaw in that thinking: the previously mentioned self-fulfilling prophecy. As advertising becomes ever more focused, it also becomes less scalable. Repeatedly cutting away the half of the advertising that isn’t working means that over time the addressable audience approaches zero; everyone in the target either becomes a customer or is cast off in the not working half.

This problem is solved by actually having waste, which is antithetical to Wanamaker’s implication.

The opposite of Wanamaker’s philosophical waxing is this: “The best advertising is written for a very small group of people, and the rest of us just happen to overhear it.”

It’s the second part of that idea that indicates the value of waste. Writing advertising for a select group allows for focus, passion and a point of view to be addressed in the advertising – all good things, and things the ad industry (sometimes) does already. But the second part allows for people who aren’t in the intended group to assume the attitude and perspective of that group, even if only for 30 seconds. To wit:

McDonald’s iconic “You Deserve a Break Today” campaign was written specifically for moms. But as non-moms saw those ads they decided that they too needed a break, and this propelled McDonald’s from a bastion of family lunchtime to the most popular restaurant in the world.

Apple’s iconic “1984” ad was actually written for early tech adopters and iconoclastic gear-heads. But then Apple ran it during the Super Bowl and a lot more folks beyond iconoclastic gear-heads figured they too wanted to get out from under Big Brother’s technological hegemony. That ad – specifically written and broadly distributed – laid the foundation for Apple to become the most popular consumer electronics company in the world.

These ads were so effective specifically because waste was unavoidable when these ads ran. In more modern times – with much less waste – these iconic ads would have been delivered to a much more precise audience, and McDonald’s and Apple would have needed a different watershed moment to build their business foundation, if that watershed ever happened at all.

Although the idea of measuring, understanding and maximizing waste is necessary, a century of advertising precedent has marginalized it. And with ad technology doing the bunny-reproduction thing, it is an idea that is not focused on by any of the emerging technology platforms, companies, agencies or clients that use these products.

This measurement and understanding does not necessarily have to be complex. We simply use the same depth of data and analysis with which we analyze our successful campaigns.

Instead of declaring unsuccessful impressions or users outside of the target as waste, we should look at that data, understand awareness, affinity and conversion behavior among the waste, and layer that against our “unwasted” impressions and in-target users. In this way we can understand the differences beyond the binary in/out dynamic, and address these differences with all the advertising tools we have at our disposal, just like we do with the successful parts of the campaign.

And then all our clients will become the biggest in the world.

Follow Joel Nierman (@FozzieBuyer), Critical Mass (@CriticalMass) and AdExchanger (@adexchanger) on Twitter.

Enjoying this content?

Sign up to be an AdExchanger Member today and get unlimited access to articles like this, plus proprietary data and research, conference discounts, on-demand access to event content, and more!

Join Today!


  1. Marc Rossen

    All good points Joel but this depends on how the “Target” is being defined of course. Any good Marketer in Ad Tech, whom I have been fortunate to working with many, know that upper funnel “reach” media feeds lower funnel remarketing. Where waste comes into play, as you noted, is if the consumer will not be receptive to your message. If the message will make some measurable positive brand impact then it is not waste. The problem is some people in Ad Tech are focused on pure direct measures and not looking at upper funnel impact or movement of cookies from upper funnel to lower funnel activities. If they did so they would stop defining “waste” as broadly as you say.


  2. I love the way you brought this back to creative, Joel. Wasn’t what I expected, but completely what I believe.

    From a media perspective, “wasted” impressions are like using a shotgun to hunt. Would you tell a hunter the pellets that didn’t hit the duck are a waste? There’s a reason cheap impressions cost $1.50 instead of a sharpshooter rifle $30 CPM for a premium placement. They’re designed to be a scattershot approach.

    Adam Kleinberg


  3. Interesting article, thoughtful musings and well-written, Joel.
    A couple of problems though: you assume the audience pool is static and repeatedly eliminating the “wasted” half of mis-targeted ad targeting will result in a declining pool of targets, at some point approaching zero. It’s not though.
    There are constantly new entrants into the pool – as people get older, more affluent, become moms (or dads) or take on whatever criteria defines your target audience.
    Moreover, good advertisers will also experiment and seek to expose new, previously untapped pools by either extending the product or message to appeal to new groups with similar traits (lookalike modeling is one approach for this) or merely altering some other component of the test, be it geography, time-of-day, context, etc. Good targeting is multi-dimmensional and good ads do not exist in a vacuum but, instead, rely on an eco-system of media, surrounding content, placement, frequency, supporting channels (social media, PR, events, OOH, etc.) and other factors. But ALL good ads – the EFFECTIVE ones – elicit SOME sort of response and that is usually measurable. I hear a lot of brand people dismiss solid metrics and rigorous analysis as the realm of direct response marketers, but that’s nonsense. If I see a “good” ad – one that appeals to me and reaches me at the right time, in the right place – I react. I will watch it again, share it, remember your product, visit the product Website, Google it, or, most measurable of all, BUY the damn thing!
    If I do any of these things, then the ad worked. If I don’t, then it’s by definition waste. If it worked then I was in the intended target audience, even though I may not have been formally included or targeted (though I should have been). If it did not, then I wasn’t or, at least, shouldn’t have been. Just because some marketers target poorly and fail, doesn’t discredit targeting. Similarly, others getting lucky by carpet-bombing doesn’t prove anything either.

  4. Totally agree with Marc Rossen here. The definition of “waste” is unclear in the article.

    Advertising cars to babies is a waste, but advertising them to teenagers is not, even though they cannot covert right away. Always measure the long-term effect. This is called exploration-exploitation trade-off and it is becoming more and more important when conversion attribution is being tackled.

  5. alejandro c.

    I liked your article. Perhaps lurking behind your assertion is the fact that many ads are ultimately aspirational. I don’t think the 1984 ad was targeted at early adopters…I think it was targeted at those of us who aspire to be that (i.e. practically everyone). I really like your re-statement of the wanamaker quote about “good advertising is for a small group of people and the rest of us overhear…” … I would add that, for brand advertising, the rest of us WANT to be part of that “small group” even though we may not be. So maybe for brand advertising, we should define anyone who “wants” to be in the target as the target. for DR, the target has to be more closed.