Commentary

Key to Cross-Platform Advertising Is Creating A Single Metric

The proliferation of media devices may be a consumer’s dream, but it’s made conducting business more difficult for advertisers and publishers. Accurately accounting for audiences across TVs, computers, smartphones, tablets, and an ever-growing list of devices has created a measurement nightmare resulting in inefficient allocation of available ad inventory.

Fortunately, there is a path for our industry to make true cross-platform measurement a reality – one that can bring both buyers and sellers together through the use of a single, unduplicated metric on which both sides can transact.

Media companies would benefit by being able to sell against their entire audience, not just their digital or TV audience. Advertisers and their agencies would be able to plan more effectively and across screens, creating a unified media plan that leverages multiple touch points to reach target audiences rather than continuing to rely on inefficient and time consuming siloed campaign planning using unconnected data-sets.

advertisement

advertisement

How can this development of a single, unified audience metric be accomplished?

1. Develop a scalable way to measure across platforms. Traditional measurement panels are necessary but not sufficient for cross-platform measurement. While panels are fine for independently measuring each media platform, building a ‘single-source’ panel with overlapping samples across all measured media at a granular enough level is cost prohibitive for the industry.

Fortunately, unified methods of measurement that combine the benefits of person-level measurement from panels with the reporting granularity of census data hold the key to scalable cross-platform measurement. More importantly, this census data – such as tagged Web pages and apps in digital and set-top box data in TV – provides the granular touch points between platforms to produce audience de-duplication algorithms that work at the network, program and episode level.

This ability to de-duplicate audiences is fundamental to marketing. Advertisers want to know how many people they are reaching with their messages and how often, and media companies get paid according to how they can deliver on those propositions. Without the ability to de-duplicate audiences, accurate reach and frequency metrics remain a mystery.

Moreover, it leads to media being planned and bought in silos, which drives massive inefficiency in allocating dollars across the most appropriate platforms to effectively reach audiences. Efforts to estimate duplication are as limited as the approaches used. Passive, electronic, persons’ level measurement is the most accurate means to measure duplication. Once applied to a larger census based data-set, it unlocks the insights the industry needs.


2. Validate digital measurement to deliver a Human GRP comparable to TV. Digital measurement on the server-side can be messy. Cookies do not accurately measure audiences. Ad impression counts often get inflated due to non-human traffic. Many other ads are either delivered out-of-view, out-of-geography, or in non-brand safe environments.

Without the ability to validate server-side measurements and translate them into metrics that correspond to the behavior of actual humans, marketers can’t make decisions that align with their strategies. And perhaps more significantly, these metrics will not be comparable to TV audience metrics, which have historically met the reasonable standard of reaching humans with an ‘opportunity to see’ the advertising.

Therefore, digital measurement requires audience validation that cleans up the mess and can simplify it in terms of a single Human GRP that is comparable to TV. While the concept and usage of GRPs is not new in digital, recent advancements in viewability and NHT detection promise to make this metric even more precise, delivering a true Human GRP. An accurate Human GRP makes it easier for the industry to plan, buy and sell advertising across platforms.

3. Build a flexible system that can accommodate an evolving definition of cross-platform. As device proliferation continues – for example, Apple unveiled yet another screen with its new smartwatch – the challenges of cross-platform measurement have the potential to become more complex by relying on traditional measurement methods. By adopting a scalable system that can easily integrate new platforms into a unified view of cross-platform behavior, the media industry will be able to jump from screen to screen as seamlessly as consumers do.

This means avoiding multiyear R&D efforts to figure out how to incorporate new platforms into the mix and having the ability to quickly allocate dollars to the right media platforms. If there’s one thing we’ve learned in digital, it’s that the landscape changes constantly – why reinvent the wheel every time that happens?

Adopting this approach to cross-platform audiences and advertising enables a single metric on which media buyers and sellers can easily transact. It means media companies get paid for the full audience they deliver and advertisers can spend more wisely to achieve maximum impact from their campaigns, a natural win-win that the entire industry can rally around.

9 comments about "Key to Cross-Platform Advertising Is Creating A Single Metric".
Check to receive email when comments are posted.
  1. John Grono from GAP Research, October 1, 2014 at 10:11 a.m.

    Excellent post Manish and Jane.

  2. Skip Brand from bbd (branding big data), October 1, 2014 at 1:08 p.m.

    Manish and Jane.....good article, great timing and sorely needed.

  3. Ed Papazian from Media Dynamics Inc, October 1, 2014 at 1:34 p.m.

    Good article, Manish and Jane but the problem is how do you go beyond purely electronic indicators----the fact that an ad appears on a user's screen-----to human 'audience"----whether the user is actually paying attention or is "engaged" by the ad? While certain things can be defined, like time spent on a page, various actions the user takes, etc. as a rule, these account for a relatively small percentage of the potential audience. Many users may not give any indication of response to the ad, yet they see it; others, who also give no indication, completely ignore it?

  4. dorothy higgins from Mediabrands WW, October 1, 2014 at 3:08 p.m.

    Echoing Ed and the need for some gauge of brand diagnostics for branded content and advertising for different channels and the desire response such as awareness, consideration, intent to purchase, recommendation, etc. Otherwise while we can add up all uniques who were served viewable ads we will have no idea how that display banner performed compared to the linear TV ad compared to the digital video pre-roll in driving long-term brand equity.

  5. Doug Garnett from Protonik, LLC, October 1, 2014 at 4:55 p.m.

    I don't agree that "single metric" is critical. Each medium plays (or should play when planned smartly) an entirely different role in the marketing/advertising mix. But unifying metrics implies comparability. It's simply not true. Platforms are not comparable apples to apples. Do we use rind thickness as the single metric for an apple and an orange? Or should we look at core thickness for an apple and an orange? It's our job advising client to keep them from the error of believing mediums are this comparable.

  6. John Grono from GAP Research, October 1, 2014 at 5:18 p.m.

    Hi Ed. It's the age-old conundrum. Are we measuring the medium, the medium's content, or the ad inserted into the content of the medium? Measuring just the medium is way too blunt I am sure we all agree. This article seems to be in the middle camp - measure the content on an audience basis. The ideal is measuring the 'engagement' or 'attention' to the ads inserted in that content on an audience basis. But we have to consider who pays for this measurement. For time immemorial the measurement of each medium is (correctly) funded by the owners of that medium as they stand to profit the most. Their basic advertising model is that they sell "opportunities to see" rather than "likelihood to see". So let's say that a TV station has a programme that delivers an audience of 10 million. In an ad-break there might be a break of half a dozen ads. Each of those ads will have its own level of engaged audience for reasons that are beyond the control of the station - the station can not control the 'quality' of the ad that appears within the programme whose 'quality' they are 100% responsible for. (This of course also applies to ads in a magazine, a cinema ad, a website home page for a day etc.). The other quandary is associated with measuring the 'engagement' of just the ad. Assuming you had a robust measurement of that ad's engagement, the power of that ad has to be taken in the context of the programme in which it is seen. There is an old saying - a bad media schedule will kill a good ad, but a good media schedule can never save a bad ad. So let's also discuss how such measurement systems of ads within the context of the media vehicle (page, programme, billboard etc.) could be funded or we might just end up designing a castle in the sky. I laud the direction, intent and call for common metrics and the goal for a single metric for the advertiser.

  7. Ed Papazian from Media Dynamics Inc, October 1, 2014 at 6:16 p.m.

    John, one of the main problems we will have in trying to measure engagement is the fact that this is a human response, which means that we must find a reliable way to go beyond electronic indicators, however useful, and evaluate the way audiences respond to the ads. I agree that it's not the medium's job to convince audiences to buy the advertiser's product. Though it should be the medium's responsibility to be ad-friendly, in the sense that it doesn't overload commercial breaks with so many messages that viewers tune them all out. Getting back to the question, I suspect that the ideal answer probably lies in some ongoing measurement of attentiveness or, if that is not feasible, of ad recall----by recall, I mean "verified recall", where the audience has to play back the basic selling proposition as well as demonstrating that the commercial was seen. I think that this is possible for TV, premium online videos, certain forms of digital out-of-home TV and many magazines. I am less certain that it can be done for radio, however. Other indicators, for TV, at least, are dial switching rates or ad zapping rates, which simply require much larger samples than are now in place. This is an area where new, set top box entities like Rentrak, which have huge panel bases, might step in and provide a very useful service. The main issue, however, is this, "Who's going to pay for this kind of research?"Up to now advertisers have tried to palm this off on the media, but I believe that a major rethink is needed and advertisers, who will also benefit, must pay their fair share.

  8. Doug Garnett from Protonik, LLC, October 2, 2014 at 2:08 p.m.

    Ed's right on about engagement. Had a friend whose company used digital activity surrounding their company product to find "engaged". Except those digits didn't reveal mind set. The company assumed it meant they were good prospects to buy the accounting software. In reality, they were people hoping to get jobs somewhere who were trying to learn enough about this particular software to sound intelligent in an interview. Digitally? Looking engaged. Mentally? Engaged...with trying to find a new job.

  9. Brooke Molineaux from ASU, October 7, 2014 at 12:55 a.m.


    This is was a great article to read. A few thoughts crossed my mind as I was reading though. As of now, how do you go beyond sole electronic indicators? Who knows if the user is actually looking at the ad or not? As it seems, online ad’s seem to provide a minimum amount of the actual business. There need to be ways to gauge whether or not the ad was useful or successful; if the user spent time looking at the ad and figure out how the audience is responding to the ad’s being posted. I don’t completely agree that the “single metric” is as important as the article makes it out to be. All mediums play a different roll and these platforms are unable to be compared. They are all very different. There can be many diverse ads during one break but only few may keep somebodies interest. I don’t think it is the medium’s job to be convincing to the audience but the medium should be ad-friendly and inviting. Using digital activity is tough because there are users looking at the ads and sites for different reasons whether it really be simply for interest with the intent to purchase whatever is being looked at or maybe to compare products. But another common reason is for those who are looking to be hired and are doing research on the company and products before an interview. As of now, we do not really know what the users’ intentions are.

Next story loading loading..