Everything you need to know from IAB Europe’s definitive CTV guide

It was assembled with help from Comscore, Kantar, Magnite, MediaMath, PubMatic, Samsung Ads Europe and ShowHeroes Group, and splits TV viewership into three fields – traditional linear, broadcaster video-on-demand (BVOD) and native CTV apps or channels.

Magnite research claims that eight out of 10 EU5 households watch a mix of TV and CTV content. Almost half of the CTV content is ad-supported. Marketers will have to get to grips with CTV, whether they like it or not.

Measurement challenge

While there’s more inventory available by the day across an array of channels, standardization of measurement remains front of mind. “The CTV territory is a fragmented one, which makes measurement challenging,” it admits.

The TV panels of linear days no longer provide the depth of detail they once did as viewers fragment. Set-top boxes added a degree of data upon adoption, and past that, the smart TVs themselves bring a lot to the table.

The report reads: “Smart devices have brought a level of sophistication not seen before in the TV landscape. Whether this is the development of content curation on the device itself, the delivery of ads in video on demand (VOD), linear or even the Electronic Program Guide (EPG), or the ability to understand overall viewing across the device, to create reporting/insights to showcase these viewing habits, not on a household level but by device.”

Targeting

The report points out: “This is a large-screen environment, and coupling it with specific targeting options such as context and identifiers makes it very compelling for brands that are searching for new ways to reach potential customers.”

There’s a rich array of data that can now be actioned. For example, those watching on games consoles are probably gamers. Contextual or automatic content recognition (ACR) techniques can place ads against relevant content (Nike ads to heavy sports viewers, for example). And in cases where users log in to apps or BVOD players, there’s consumption and demographic data that can often be layered with third-party data brought in by the client.

When targeting by device, advertisers are targeting households, not individuals. Marketers can’t be certain exactly who in the house they’re reaching. But it’s often more than one person.

ACR also operates through smart TVs to understand what people watch. It analyzes identifiers in content to understand consumption habits. That could be a sonic frequency imperceivable to the human ear – an audio barcode.

Here’s an example of why this would help. “Light linear TV viewers can be reached, or a user or users who have seen a particular TV show, for example reaching only users who have been watching F1 content with an ad featuring an F1 sponsor.”

More of this data is readily available as smart TV purchases increase.

Contextual targeting is a little more difficult. “It tends to rely on what is passed in the bid stream by the publisher and SSP. Whether through a third-party solution, which is able to ingest video content and attach metadata (i.e. show or genre), or through the app or publisher themselves.”

Most of the identifiers don’t make it through to the demand-side platform from the supply side yet. “Specific inventory or channels can also be targeted in the same way where app publishers or CTV channel providers create identifiers by channel, by genre or by context for targeting purposes. This is still in its infancy but is one of the fastest-growing areas across the CTV landscape.”

There’s further work ahead to better understand how genre and meta-tagging can be understood and deployed.

“This can move an advertiser forward from just targeting ‘auto’ content to targeting programs around innovation, future focus, sustainability etc. This level of content understanding is in its infancy and as with all CTV targeting solutions does have scale limitations, but as publisher content libraries grow and advertisers’ strategies need more granularity.”

Until that scale is reached, CTV lives to extend the reach of linear TV campaigns, with many buyers deploying test-and-learn approaches. One day that dynamic may flip. Even linear advertising is becoming addressible, however, so the distinction is becoming less relevant.

CTV and BVOD can be used to reach viewers who saw the linear campaign and drive further recall, using the aforementioned identifiers to monitor frequency and who is being reached.

Best practice

Supply Path Optimization (SPO) is encouraged to ensure the DSP and SSP are sharing the correct information – particularly with a wealth of metadata needed for contextual buys.

• Consider connecting with Original Equipment Manufacturers (OEM) such as Samsung Ads to utilize ACR where possible.

• Seek transparent reporting across the whole supply chain where possible.

• Consider curated marketplaces where premium CTV inventory is collated at scale.

• DSPs connected to publishers and SSPs offer extra layers of targeting.

• Seek partners with multiple CTV content identifiers that enable campaign scale, particularly those with stronger contextual offerings.

• The most efficient and value-driven metric to understand performance on a cost level is cost per completed viewable view (CPCVV) giving “a true measure of value against the cost of media.”

What’s next?

Publishers are being encouraged to secure as much user consent in their databases as possible. “Once this is more frequent, IP addresses can be used to connect IDs from other elements of the supply chain. In the US device graphs map users using mobile ID and house IP connected to CTV.” The theory is that this approach can be adopted in Europe too, as will cookieless targeting solutions.

Next, buyers need to be able to measure all types of viewing in the one interface, whether linear, BVOD or CTV. Barb in the UK is helping develop such a solution.

“Technically this brings the advent of new hybrid measurement approaches combining the best of worlds including ACR – be it via audio fingerprinting or watermarking, with passive panel metering technologies such as people meters or portable people meters, along with digital metering, potentially via linked mobile devices or via home router level meters, and accompanied with further third- or first-party census feeds.”

Read the full report here.