The Association of Commercial Television

EXPOSING THE BLACK HOLE OF ONLINE VIDEO ADVERTISING test

There were lots of interesting things to catch the eye in the Advertising Association/Warc annual adspend figures for 2018 – not least the rapid growth of Broadcaster VOD advertising, up 29% to £390m and forecast to grow at a similar rate for the next two years. Total TV investment was stable year on year at £5.1bn.

But what really caught my eye, and then violently thrust my eyebrows up several notches, was the amount of adspend now going to online video.

In 2018, it was £2.3bn, up from £1.7bn in 2017. Take BVOD’s reported £390m away from that and you have a non-broadcaster online video ad market in the UK of £1.9bn – 26% of the total video ad market.

Why the raised eyebrows? Well, when you compare this share of video adspend with the relative share of video ad viewing, everything gets weird. Very weird, indeed. And eye-wateringly expensive.

TV including BVOD accounts for 95% of video advertising time, according to analysis of industry data.

Online video such as YouTube, Facebook and the long programmatic tail together account for 4%. Cinema is the remaining 1%.

In other words, the average person watches 17 minutes of TV advertising a day (the 95%) and one minute of non-broadcaster online video advertising (the 4%).

Yet AA/Warc figures show that £1.9bn was poured into that single minute a day and £5.1bn into the 17 minutes.

That means 4% of video ad viewing took 26% of video adspend; the other 95% of video ad viewing took 70%.

How are your eyebrows doing? Mine have escaped my forehead and are heading for the skies. And we haven’t even factored in the quality of the different ad exposures; this is purely based on quantity of advertising seen.

Let’s turn this into the comparable measure of cost per completed views. Online video ads are usually priced on a cost-per-start basis, but that is obviously the lowest of benchmarks, so let’s look at the average cost for 30 seconds of video ad viewing for a meaningful like-for-like comparison:

As you can see, the average cost across TV advertising (linear and BVOD) for 30 seconds is just over £6. For non-broadcaster online video, it goes up to a whopping £45.

So, the potentially brand-unsafe, often small-screen, often partially viewable world of online video costs advertisers seven times more than TV. This doesn’t sound like value for money.

How has this happened? How is this black hole sucking in such a disproportionate amount of money?

What seems likely is that advertisers are often making decisions about online video based on its superficially attractive cost per view (start), rather than assessing the price based on completed views.

What to do? First, please return my eyebrows to me if you see them. Second, I would suggest advertisers request a detailed breakdown from their agencies of the like-for-like cost per completed (human) viewable view in their audiovisual mixes. If their eyebrows can take it.

Source: https://www.thinkbox.tv/news-and-opinion/blogs/exposing-the-black-hole-of-online-video-advertising/

WHY IT’S TIME TO RE-EVALUATE TV PLANNING test

As BARB releases its latest Viewing Report, Wavemaker’s Emma Moorhead discusses how new multiple-screen data is changing the way that agencies plan television

The media industry is in search of the Holy Grail: a single-source measurement of TV viewing across all screens and channels. In 2018, we got one step closer to this with the launch of multiple-screen viewing figures, the first stage of BARB’s Project Dovetail initiative. While these additional data are very much welcomed, how do they change the way we deliver the best outcomes for our clients?

Overnight viewing figures have formed the bedrock of how we plan, buy and optimise television campaigns. The launch of a new campaign goes hand-in-hand with securing a kick-off spot in a top-rating programme to reach a mass audience. If a programme over- or under-performs, next-day negotiations ensue to secure the desired number of exposures. But in recent years, displacement of viewing has made this task more complicated.

Linear television viewing is now regularly timeshifted. On average, people watch 29.3 daily minutes of television timeshifted, and this timeshifted viewing accounts for an average increase of 15% on a typical overnight rating. Sometimes it can be more; the launch episode of Shipwrecked on January 28th 2019 was watched live by 219,000 people, but the consolidated 7-day figure more than tripled to 685,000. Consolidated ratings used as a proxy for reach become progressively difficult to estimate, making the buyer’s job increasingly complex.

Some content is still watched live by the majority of its audience, in particular event programming – think of England in the World Cup. BARB data break down the numbers watching live or timeshifted within seven or 28 days, so we can buy the right spot when that live first-look is still the key metric; we’re starting to see greater nuances in how we plan our television content.

In addition, while it’s well-documented that linear television consumption is in decline, this is partially compensated for by the growth of BVOD services. The appetite to consume high-quality content for several hours a day remains, albeit fragmented across screens. BARB’s new multiple-screen viewing data give us insight into how programmes are viewed across TV sets, PCs, tablets and smartphones.

For example, in 2018, Love Island gained up to 27% incremental viewing uplift from non-TV devices, while Family Guy gained up to 8% uplift. Nonetheless, as a whole, non-TV set devices add less than 2% to TV set viewing; the TV set remains the favoured means of viewing.

And despite these additional data on device consumption, we remain none the wiser as to the incremental reach BVOD can offer to a linear television campaign across all screens.

Using BVOD to supplement reach is centred in a linear television-first approach to planning. Interestingly, another aspect of BARB’s data suggests that alternative approaches could be sensible to deliver impact in a fragmented viewing world. BARB can now measure viewing to programmes on BVOD services before they are broadcast.

For example, 1.15m people chose to watch the second episode of Save Me via Sky On Demand pre-broadcast on TV sets; this was more than half of the total TV set audience of 2.19m. This increased throughout the series, with the final episode watched by 83% of people via Sky On Demand pre-broadcast.

Understanding how programmes are consumed influences how we approach planning. A linear-first approach misses the opportunity to reach audiences when the content is at its most valuable; when it’s providing the watercooler moment. We should instead be moving towards an audience and content-first approach – buying the right programming, at the right time for greatest impact amongst viewers.

The need to re-evaluate our planning approach becomes even more prominent when we consider that changes in consumer behaviour are starkest among younger audiences. Ad-supported YouTube and SVOD services such as Netflix and Amazon remain the top challengers to television’s incumbent media owners.

BARB’s measure of unidentified viewing – where the TV set is used to do something other than watch a BARB-reported channel or BVOD service – includes viewing to SVOD services and online platforms. In 2018, unidentified viewing accounted for 48 daily minutes for all individuals, rising to 71 minutes for 16-34-year-olds. Greater insight into this consumption would be very much welcomed.

YouTube has expressed a willingness to be part of BARB’s Joint Industry Currency (JIC) model, but in its own words it wants to be “represented appropriately and fairly”. Here remains a fundamental problem with reconciling television and online viewing.

Television is measured by impacts in units of 30 seconds, whereas online is measured by impressions with no time exposure element. For the two to be comparable, we need to use duration-based measurement for online viewing. Everything needs to be clearly labelled so we aren’t comparing apples and pears.

Meanwhile, addressable television is on the rise but is far from ubiquitous. In order to realise the long-promised future where television is a more efficient, targeted and digital-like medium, we need to reach a point where content and distribution are more vertically integrated.

In this future, new measurement opportunities may complement the data offered by BARB through the likes of set-top box data. A more digital-like television future offers the opportunity to deliver precision at scale.

Trusted and accurate measurement remains essential to accountability, planning

and optimisation, and increasingly so in a world where we see displacement, fragmentation and disruption. Ultimately, we need to understand the value that each exposure drives for advertisers. The outcomes are what are important; measurement allows us to link exposure to value.

The industry must come up with a measurement solution enabling better understanding of viewing patterns across all screens and channels. This is still some years away, even in the most advanced markets.

BARB’s Project Dovetail in the UK is setting the example, although we must remain patient before we achieve multiple-screen advertising campaign performance. Regardless, the JIC principles underpinning Barb should not be weakened or compromised.

Despite everything, linear television has sustained advertiser demand, giving the impression that it is as effective and essential as ever, but for how long, and in what balance relative to the alternatives?

Source: https://mediatel.co.uk/newsline/2019/05/20/why-its-time-to-re-evaluate-tv-planning

TV ADS STILL MORE RELEVANT TO CONSUMERS THAN STREAMING VIDEO ADS test

What does this say about programmatic video ads?

Despite marketers’ efforts with advanced programmatic and data targeting for video ads, consumers still find that they are more likely to be served a relevant ad on linear TV.

According to a survey conducted by Adobe in February 2019, 49% of US internet users said that TV was one of the mediums where they were most likely to see a relevant ad, while just 12% said the same about streaming video.

Valuable ad placements and the ability to reach a mass audience have kept TV ads relevant in the digital age. But digital video allures marketers with advanced targeting through programmatic advertising, which uses audience data that TV doesn’t have.

This year, we estimate $29.24 billion will be spent on programmatic video ads—accounting for 81.2% of digital video ad spend. For TV, just 4.0% of ad spending in the US will be programmatic.

There are many perceived advantages to programmatic advertising, most of which revolve around lower costs and the ability to harness data. In response to a November 2018 survey by Digiday Research, 56% of US agency and brand media buyers said that increased targeting and optimization was the biggest advantage of programmatic advertising. But if viewers still find TV ads to be more relevant, then programmatic may not be as effective as some marketers perceive it to be.

Consumer sentiment does indicate that ads are becoming more relevant overall. In the Adobe study, 46% of consumers felt that the ads they saw currently were more relevant than those they were served a year prior. But, most marketers can agree, there’s still a long way to go before personalization is perfected.

Only 32% of marketers believe their industry is delivering personalization effectively, according to a survey conducted in February and March 2019 by Evergage and Researchscape International. When asked how satisfied they were with the level of personalization in their own marketing efforts, 50% said they were either not satisfied or slightly satisfied, and 34% said they were moderately satisfied.

While marketers may have faith in programmatic’s potential to target the ultraspecific, it is possible that the accuracy of the data just isn’t sufficient—and that’s why these ads aren’t resonating with viewers.

Many marketers would agree, per the Evergage/Researchscape International report. Nearly half of respondents (45%) felt they didn’t have sufficient data and insights for effective personalization.

Source: https://www.emarketer.com/content/tv-ads-still-more-relevant-to-consumers-than-streaming-video-ads