CTV Measurement: Data Standards are Crucial Now & Integral to Future-Proofing
The IAB announced today its updated ad format guidelines for CTV (Connected TV), spurred by the rapid expansion of the sector and accompanying rise in ad spend. Interestingly, they also allude to the rising workflow inefficiency brands face in keeping up with CTV proliferation.
The goals of the update include:
- Minimizing the fragmentation of ad formats
- Standardizing formats and metadata
- Reducing the costs of producing assets in disparate formats
- Transitioning seamlessly from linear to CTV
- Staying agile to new CTV developments, such as interactivity
- Scaling both buy-side and sell-side with programmatic
- Enacting technical standards
The IAB has opened its standards for comment from the industry so it can align with existing “creative file requirements so that this guideline defines the baseline ad creative requirements to make it easier for brands and agencies to develop and submit ads for their available inventory.”
It’s a major step in the right direction. But it still falls short of:
- Enabling the pie-in-the-sky promises of Addressable TV
- Activating data standards — within the adtech/CTV industries nor within brands themselves
- Being future-proof and agile for whatever form of media, concern around privacy, or development in CX comes next
What we’re seeing in CTV measurement today is only the tip of the iceberg and it only concerns what’s available today — not what will develop in the future. Simply kicking back and relying on the IAB to dictate and activate data standards is almost as risky as not leveraging data standards at all.
Brands must instead take an in-house (or partner-aligned), proactive-ownership approach to their data standards (in marketing and beyond) to achieve a connected ecosystem that’s prepared for tomorrow’s challenges, too.
So let’s dig into these issues and how you can address them.
What's Going on with CTV Measurement?
Addressable? Eh, Not So Much
Matt Krepsik’s position via AdExchanger, The Truth About Addressable TV Is That It’s Not As Targeted As It Seems, pretty bluntly calls out the half-truths around addressable TV. It’s not quite the dream scenario it’s made out to be — or could be — when it comes to hyper-personalized marketing.
It may be bought and sold as a reality, but it’s far from realization, he argues.
Yes, we’ve seen major leaps in targeting capabilities from linear TV. But truly addressable TV marketing would enable a nearly endless amount of ad variations for possibly every single potential viewer — plus variations for each viewer such as by time of day or triggered by micro actions (also somehow measured).
One example of the current shortcomings are the supposedly personalized ads played between sections of our favorite shows on Hulu (et al) that we might see literally five times during the same episode. That’s not only poor pacing and wasting a lot of budget, but delivering a terrible customer experience.
This won’t change until we have a central ad exchange, standardized data network, and what he calls “smarter data” that can actually tell who’s watching (vs sharing an account) and even who’s actually looking at the screen. This would also encapsulate follow-up interactions such as landing page visits, QR code scans, and mobile app experiences.
Again, all just a dream for advertisers at the moment. But the cornerstone — CTV data standards that enable universal measurement — is rallying support across the industry. Even brands like LG, Samsung, and Vizio are becoming more forthright with their automated content recognition (ACR) data agreements. Or, they launch their own ad networks, such as LG Ads or Vizio’s Inscape ACR, and pledge transparency.
All three brands talk a tough game on fraud and other concerns, but that transparency is still in question — these are just more walled gardens, after all.
Quick Aside: Understand the Digital TV Jargon
Before we dive into the current state of CTV data standards, here’s a quick refresher on all this digital TV alphabet soup.
What most of us grew up with — think cable, satellite, or antenna TV, but also any standard broadcast channel you might stream via YouTube TV or Hulu’s live offerings. Ads will seem audience-relevant by channel (frosted cereal on kids’ networks, bran cereal on the news networks), region (snow tires in the Midwest, all-seasons in the Southeast), time of day (Sunday morning, Friday late-night), and other broad segments, but they’re not really personalized.
During one show’s screening on a single channel, a 75-year-old viewer is seeing the same commercials as an 18-year-old, despite the chasm between their buying preferences.
OTT (Over the Top)
While you may think this descriptor applies to nearly all modern programming — especially cable news — it’s actually a more literally rooted acronym. OTT refers to targeted multimedia played “over the top” of the streamable content on any device from TV to mobile to game consoles to, I guess even those smart fridges with TVs in the doors.
Everyone sees the same streamable content, but not everyone sees the same OTT ads. Similar to display ads on Google or Facebook, OTT ads allow buyers to choose audience segments. So now brands can not only advertise specifically to those who watch baking shows, but bakers of certain ages, incomes, and other demographic variables.
CTV (Connected TV)
Chances are if you bought a television in the past few years, you had to configure it with your home WiFi network. Like OTT, CTV serves targeted ads atop streaming content, but while OTT refers to any device under the sun, CTV is specific to television sets.
There’s an argument that CTV is the most ideal place to advertise, because it has the largest screen and the fullest attention — no tabs or other apps running in the background to distract — although this is highly questionable given our proclivity to “second-screen” it: scrolling Instagram while watching a series, for instance. CTV is also limited in its ability to track the next action, such as click-through on a pre-roll ad.
But it can provide more trustworthy data on location and audience (given that it’s typically only used in one spot and directly linked to the user’s accounts via apps). The concept of a more “brand-safe” environment also arises, since the ad will be shown on the TV full-screen, without the risk of popping up against irrelevant, explicit, or otherwise questionable content contexts.
The dream of advertisers since the dawn of Timex — playing personalized ads suited to the individual viewer.
This is currently done via OTT and CTV on a household and/or account level, but it’s still just a slightly more refined iteration of the linear TV model of advertising. The segments are tighter, but to say it’s “individualized” is actually quite the monumental stretch. Hence, Krepsik’s perspective referenced earlier.
Frustration Around CTV Measurement and Sort-Of Solutions
Brands that spend millions on CTV ads are “becoming frustrated, unsure how they should measure audiences in these spaces,” reports The Drum. They’re not getting the ROI — or the value, period — they expect from the TV renaissance. One big player has a sort-of solution, but as large as Comcast is, it’s not the whole playing field.
And while fraud in CTV advertising has taken up a lot of the headlines lately, it’s far from the only problem or even the only challenge to ROI. But it is another aspect of CTV marketing affected by the complex, fragmented ecosystem and corresponding lack of transparency and consistency. While advanced hardware and “more sophisticated ad verification technologies” may be part of the crusade against fraud, the conversation again comes back to the adoption of industry-wide taxonomies, or data standards, to enable supply chain transparency.
AdWeek points out additional concerns from industry execs that lament the pacing/frequency issue, inaccurate IDs, and proving ROI. At the same time, they want to spend more and get even more targeted: local, contextual, individual. They also nearly beg for industry-wide data standards.
Can’t We Just Aggregate the Data? Well…
Starting with political customers (which value top-of-funnel audiences), Comcast’s FreeWheel made a grand step in their CTV measurement capabilities at the end of March by improving deduplication. The development of aggregated ad data from both linear and CTV allows for elevated segmentation abilities.
The advances in FreeWheel’s CTV measurement really come from advancements in how data is collected and leveraged between silos (linear and CTV), “because it’s able to extract log-level exposure data tied to linear addressable commercials,” and can then “ deduplicate reach at the household level.” That takes customer segments beyond zip code to track individual ad exposures by household.
FreeWheel’s pivot to aggregated data, they say, allows them to pivot from campaign stewardship to media activation. The shift from reactive stewardship to an active approach means — you guessed it — more data sprawl.
FreeWheel and its Comcast parent have an enormous range of channels (in both senses of the word) pumping out data to collect, share, aggregate, and activate. Yet those on the brand side — and even smaller media players, which, in the context of Comcast, is nearly everyone else — plus others on the sell-side, aren’t necessarily swimming in the same data ocean as Comcast.
Oops — Can’t Forget About Privacy and Compliance for CTV Measurement
Any CTV marketing strategy needs to be just as careful around privacy compliance as an email, social, or search marketing campaign. The IAB has, of course, considered this and uncovered a worrying gap in not only measurement but accountability and consent in general.
Solutions such as OneTrust OTT and CTV Consent provide packaged solutions, while enterprises with existing in-house compliance departments can integrate CTV customer experience data into their ecosystem. Regardless of the route a company takes to adhere to privacy guidelines such as CCPA and GDPR, having clean, usable, and standardized data remains crucial to make a successful integration and be future-ready.
Case in point: CPRA, sometimes thought of as CCPA 2.0, has already expanded the protections afforded to Californians under the original privacy law. Having data standards in place allows for more reliable retroactive analysis and alterations to data collection practices to stay in compliance.
Data Standards for CTV Measurement, Marketing, and Activation
- Campaign impressions
- Total reach
- Completion rate
- View-through rate
- Ad viewability score
- Cost per mille (one thousand impressions)
- Cost per completed view
So how does the CTV world equip itself to provide reliable measurements and meaningful insights and activations, instead of risky suggestions based on small, imprecise data pools (and misinformed KPIs)?
Back in November 2021, Omnicom announced it “has launched The Connected TV Signal Standardization Initiative, calling for industry-wide support of a set of investment practices and protocols designed to bring the transparency, quality consumer experiences and safety of linear TV to Connected TV(CTV) environments.”
They reference concerns about safety, privacy, transparency, sentiment, and most costly, measurement — because fraudsters might be manipulating ad spends and falsifying results. So what do they recommend? Data standards that:
- Capture metadata around show, channel, genre, placement, and more
- Allow for third party validations on ads and spend
- Align with a governing body, such as the IAB’s Content Taxonomy 3.0 (remember — the IAB is who’s predicting measurement blackout)
- Enable a transition to household-level IDs and the phasing out of privacy-unsafe IDs like IP addresses
- Empower buyers and sellers to create first party data confluence within clean rooms
- Implement ads.cert 2.0 for supply chain transparency
- Lead to better CTV measurement
- Connect “devices, platforms, and services” for an aggregate approach to fraud prevention
The IAB actually laid our their own approach (CTV Advertising Standards: IAB Tech Lab Standards & Guidance) and it uses nearly identical categories.
Nicole Scaglione, global VP of OTT at PubMatic, isn’t waiting for Omnicom’s initiative to take hold. She’s urging: Publishers, Don’t Wait For CTV Standards – Act Now. Otherwise, you’ll fall victim in the meantime to “fraud, a lack of clear media-buying signals for buyers used to legacy processes and a hodgepodge of approaches to measurement.”
Scaglione emplores publishers to create and apply “clear and consistent standards when sharing data about impression placement and content.” She also agrees with Omnicom’s recommendation of using IAB’s content taxonomy and the necessity of clean rooms to fully realize data’s potential.
Seems easy enough, right? It can be.
Data standards can also include the household IDs both sources encourage, to enable third party verification and “provide measurements like incremental reach, frequency, on-target demo and other CTV/TV-specific metrics.”
She highlights the profitable position for publishers: “For example, if CTV publishers implemented the Open Measurement SDK, it would open up new demand because it would provide transparency and build trust with buyers.”
Arguably, a larger market, with greater trust and transparency, benefits everyone involved — buy-side, sell-side, and consumers.
Think Beyond Today’s CTV Challenges — Future-Proof for the Inevitable and Unpredictable
Inarguably, the role of data standards permeates the CTV marketing and measurement conversation. But brands need not wait for the CTV industry to make big moves. They can take a proactive approach to data integrity (which unlocks measurement across every channel, partner, team, and tool) by taking the reins of their own data standards.
Sure — go ahead and adopt the IAB’s CTV and content taxonomies! Then expand upon them to factor in metadata unique to your company, teams, workflows, partners, and tools.
And maintain the comprehensive standard as the organization scales and evolves.
And apply those data standards across every team, tool, and touchpoint in the enterprise.
And then connect your SaaS and activation and CX apps so that they can not only take in these standardized data inputs but feed their own new data out and back into your ecosystem with the same standards applied.
If that sounds intense — it is. If it sounds impossible — it isn’t.
Bringing a data standards solution to life in the marketing organization (and permeating it throughout the enterprise) is the task The Data Standards Cloud is built to tackle.
Don’t treat data standards as only a solution for today’s challenges of CTV measurement (or even marketing measurement more broadly). Let the data integrity experts at Claravine activate data standards that future-proof your entire organization for any and all of the unpredictable data-centric challenges of tomorrow, next year, or next decade and beyond.
Because inevitably, the sky will fall again. And again. (And… again.) But your data will be (standardized) and ready.