After years of discussion and months of working in earnest, the world’s leading advertising authorities early this morning released a “framework” and a set of principles for creating the next generation of cross-media measurement. Importantly, it is designed to be future-proof measurement in a way that deals with both technological, platform business models and with self-regulatory changes that have already undermined third-party browser cookies and are likely to do the same to persistent mobile identifiers.
The group, organized by the World Federation of Advertisers, with participation with the Association of National Advertisers and its European counterparts, is intentionally designed to be advertiser-centric, but it also provides a “north star” for media, as well as data, technology and research providers to begin building new methods that account for new media forms, as well as the evolution of media and the societal forces that govern them.
The framework is not an edict or a set of global rules, per se, and is intended to be applied based on the local standards of each advertising market, but there are some guidelines that should make their development relatively consistent, if not universal.
For one thing, they seek to unify a historical paradox of media measurement: the kind of legacy panel-based measurement (think Nielsen) that has been the currency for traditional media, as well as the kind of “census logs”-based measurement that has been the coin of the digital realm.
The result will be a Secure Universal Measurement Identity, or the industry’s next big acronym -- SUMID -- which will allow advertisers and the media to both ensure consumer privacy and control, while enabling the industry to control for reach and frequency.
This last part is key, because the goal of the new cross-media measurement systems is to enable the ad industry to finally fulfill John Wanamaker’s conundrum (I know that half my advertising works, I just don’t know which half), while also creating a better consumer experience (for example, fewer, more relevant and better-targeted ads).
The framework is painfully technical, but based on my initial read of it, mobile technology will play a vital role, because the basic segments of its stack include the location and device of consumers, as well as basic demographics.
It’s not clear exactly how this will be achieved in a privacy-compliant way, but the framework coins yet another new vital acronym -- VIDs, or virtual people IDs -- intended to enable the industry to target people, and control for reach and frequency, without actually knowing who they are.
The development of these VIDs will no doubt be the most challenging and hotly debated component of the framework, because they need to be both accurate in terms of achieving the industry goals, as well as ensuring they don’t violate consumer privacy regulations, including non-re-identification: the process by which records in a de-identified consumer data set can be linked to specific individuals, by combining with records in another dataset.
Phew, good luck with that one.
Lastly, it looks as if the initial phases of this framework will still be TV-centric, based on my read, and that the specific industry self-regulatory bodies like the U.S. Media Rating Council will play a lead role in auditing and accrediting their methods and compliance.
I think that’s a good thing, but it will no doubt create some political tension among the various segments of the “cross-media” marketplace, especially mobile, given that the mobile marketing industry has been among the most vocal critics of the MRC’s efforts to establish the first building block of cross-media measurement: a duration-weighted viewable impressions.
Fasten your seatbelts. It’s going to be a bumpy ride.
Agreed! A very bumpy ride indeed!
Beyond all the highly complex technical issues, this document is referred to under the guise of cross-media measurement. However, the WFA unabashedly and understandably noted that this an "advertiser-centric framework". This raises a fundmantal question: "Does this proposed approach, which actually addresses Ad Effectiveness Measurement (the advertiser’s fully understood Holy Grail of which Media is merely an important driving component of many), burden the media with far more than they are responsible for or should be accountable for?
That accountability is no more than, nor no less than delivering actual exposure to the content and/or the ads to the specified target group in a "suitable" and non-toxic environment. i.e. delivering "contacts" and not, underlined not, "gross impressions" however "impressions" might be misguidedly defined. And, beyond that, the most important media metric deliverable to advertisers, unduplicated reach of those exposed in the target group within defined time periods.
Media's deliverable is reflected by Level 3, Advertising Exposure in The ARF Media Model. Ultimate brand effects and/or sales are reflected by Level 8, Sales Response! Achieving Levels 4, 5, 6, 7 and 8 are driven by numerous powerful elements in addition to the media vehicles used in the brand's marketing campaign. Are you ready for the ride? Stay tuned.
And the beat goes on. I agree, Tony, but no matter how often a few of us point out the inherent fallacies, we are hopelessly outnumbered and our voices will be drowned out by the pressure to develop one-size-fits- all "metrics" that are based almost exclusively on devices being on and content being on-screen, without concern---aside from lip service---about anyone being there or actually watching an ad message. Also, the only path that can ever bear fruit is one that is fair and reasonable for both buyer and seller. Trying to force sellers to guarantee results---website visits, clickthroughs, ad awareness, sales?---- isn't fair or even logical as that is not the business the sellers are in.It's like demanding that the U.S. Postal Service guarantee responses---or sales?--when it delivers direct mail pieces to lists of prospective customers.
Thanks Tony and Ed.
I'll chip in a thought or two.
First, I am pleased to read that it is "intentionally designed to be advertiser-centric". By that, I read that it will be advertiser funded. (OK, OK ... I know ... it is snowing in hell).
Second, have a peek at what our TV currency provider here downunder, OzTAM, is working on and planning to release next year. It is called VOZ - Virtual Australia. It is a meld of panel data (good for cross-device and duplication, as well as the base for broadcast), server-side data (digital 'impressions'), and Australian Bureau of Statistics Census data (our most accurate count of people, age, gender, location). Flipping it all on its head, what if we had a data base of our 25 million population that were 'pseudo-respondents' (i.e. it will always add up to the official population) onto which we overlaid TV viewing data via any device that is 'average minute', time-based, and preserves the duplication to allow robust R&F calculation? Yep, it is looking very promising. Caveat, it is broadcast TV centric at this stage - because they are paying for it.
Third, I agree with 'what constitutes an ad impression'? There are two major factors here. First, how many people potentially exposed to the ad-break. But in IMHO, the singularly most important factor is the ad itself. Run a crap ad and you will lose your audience. Whose fault is that? Hardly the broadcaster! So should ads that under-perform to the adbbreak norm, and lose 'ad-break audience' be surcharged for losing audience to other ads and the programme itself. Advertisers ... just be careful what you wish for.