Thinking Inside the Box Phase 2: The Addressable TV Opportunity

We’re excited to be able to unveil the second phase of our research – Thinking Inside the Box. GroupM Nexus commissioned this research with DRG to explore the medium of TV, what it means to viewers and how they consume and respond to the content delivered within the TV environment.

We launched this on 29th September where DRG’s Phoebe Casey-Miller joined by UCL Professor of Cognitive Neuroscience, Joseph Devlin, presented the key findings followed by an industry panel to discuss some of the key themes raised. Thank you to those who joined the live session.

If you missed it or couldn’t make it, you can catch up and watch the recording of the session here.

We also have a couple of infographics also available to download here.

We ran out of time to answer all of your questions in the Q&A. Here are the ones we missed and the answers from our experts:

Which audiences are addressable ads better for according to the neuroscience experiment? Was there a pattern for sectors that this works better for?

Joseph Devlin, Professor of Cognitive Neuroscience, UCL:   In this module of the experiment, we only tested four categories of addressability:   cars, gender, mobiles, and family. In our data, the car and gender categories showed the strongest advantages for addressability, followed by mobiles and finally family ads. These latter two categories were much less ‘specific’ in their audience and messaging too and presumably trying to garner a broader appeal, so the upshot of this is that the better you know the audience you are trying to engage, the more valuable and effective addressable becomes.

Audiences who are interested in more relevant adsIs that 22% of 52%, i.e. 10% of the overall survey?

Phoebe Casey-Miller, Research Manager, DRG:    This was 22% of the 52% to explore the differences between personalised and relevance. Amongst the total sample, the % who said they would be more likely to watch TV ads if they were relevant to them was 33%.

What’s the line we draw between ‘targeted’ and ‘personalised’ or ‘relevant’?

Clive Page, Head of Product, Data & Analytics, GroupM Nexus:    There is a blurred line between personalisation and relevancy as this something that is subjective, depending on the individual. Personal or sensitive data should not be used, nor anything that can be applied to an individual user, which is where online advertising has come unstuck – you looked at this table on a website but didn’t buy it. Viewers should feel that that advertising is more applicable to them, without it being obvious that they have been singled out to see this particular campaign.

Following that comment from the Research director, is it now possible to frequency cap via GroupM Nexus?

Kristian Claxton, Head of Engagement, GroupM Nexus:    We have the ability to cap by supplier and can now monitor and optimise total campaign reach and frequency via our integration with AudienceProject. This allows for a measure of viewing across all supply partners (broadcasters) for each of the campaigns. The output of the measurement can be viewed in GroupM Nexus Report, enabling brands to better understand how their campaigns performed with regards to R&F. This is a market-leading solution that has created some exciting new opportunities to measure Linear + On-Demand in combination (GroupM Nexus Total TV).

What proof do you have that the cost premium of addressable ads is more effective in terms of driving sales for a brand versus broader targeting?

Clive Page:    The premium applied to addressable TV advertising should only be justified based on the benefit the additional level of precision it affords to the campaign delivery. If the segment definition is x10 more effective at reaching and influencing those who are being targeted, then it could be argued the x10 price increase is justified. Of course, this is a hugely oversimplified assertion, as there many factors that are also at play when considering pricing and premiums – the quality of audience planning, the level of affinity, the value associated to the broadcast non-target audiences (wastag0, the relative cost of other media. However, using the previous example of effectiveness of addressable TV campaigns, it would be unlikely that the price set would be anywhere near a x10 premium, as most in-market pricing would be keen to demonstrate clear efficiencies in usage.

How is the industry overcoming the challenges you have posed for addressable TV?

Kristian Claxton:    Collaboration is key, this will need to continue if we are to overcome the challenges posed for Addressable TV. Clients do not want to have multiple buying points, varying audience definitions and separate measurement solutions for each buy line. We need to simplify this complex ecosystem for clients. Unified measurement through BARB is essential and something we would all welcome – especially as they have been a driving force behind the growth of linear TV over the years. Measurement & Effectiveness is key to building clients understanding of how addressable adds value to their business, proving out benefits is key for gaining trust and showing its delivers against core business objectives (e.g sales, shift in brand perception, reach).

Did you ask consumers what they thought of the ad breaks? My experience is that breaks often cut ads off before they finish, have the same ad in consecutive breaks, or indeed no ads in it at all. The technology failings make the “Addressable TV” industry and advertisers look unprofessional to the consumer, and leaves advertisers cynical to the medium. The theory sounds great, the reality is poor.

Phoebe Casey-Miller:    We did. Amongst the online survey, we found that 16% of people explicitly said they enjoy watching TV ads. Within the qualitative work, there seem to be 3 key groups when it comes to perceptions of ad breaks. 1) Complete TV ad rejectors – they want control over their TV viewing and don’t want to watch any ads / find them an intrusion or an annoyance. 2) TV ad break acceptors – they accept TV ad breaks as something that’s part of life, they don’t mind them, they might go and make a cup of tea when the ad break is on. 3) People who explicitly like TV ads as a way to stay up to date. There was no insight into the tech issues around ad delivery – although some people clearly see ads as an annoyance and feel bombarded/think TV breaks are too long, tech issues didn’t come up in any of the viewer workshops or ethnography.

A bit of a geeky question – sorry – but how does the research account for the Hawthorne Effect, when individuals modify an aspect of their behaviour in response to their awareness of being observed?

Joseph Devlin:    The Hawthorne Effect comes into play when people are aware of an intervention and are conscious that it is supposed to do something, making them more likely to change their behaviour. In this experience, participants were not aware of the addressability dimensions — they simply saw some ads during a TV show so there was nothing to create a Hawthorne Effect.

If targeted ads drive increased relevance, how does this change the way we buy and measure campaigns with GroupM Nexus?

Kristian Claxton:    Something that makes GroupM Nexus unique is the robust measurement framework that goes above and beyond post-campaign reporting. We understand that Addressable TV is a new and evolving space, and therefore want to prove the impact of it for our clients. In addition to the traditional TV metrics, such as brand uplift or incremental reach, we can measure shorter-term objectives such as sales/ROI to determine the success of a campaign. We have a huge number of measurement & effectiveness studies (200+) working with independent research companies such as Kantar, LiveRamp, YouGov & in-house (client or agency) econometrics team. We are starting to explore new trading mechanics with regards to guaranteed outcomes (shift in brand consideration, sales and reach) – however, these are still very early conversations and would welcome agency and advertisers alike to come and speak to us with any ideas.

Do we know what category of brands the ads in this study were from or was it a mix? Surely certain categories will suit a personalised ad far more than others and therefore results would vary greatly based on this?

Phoebe Casey-Miller:    The neuroscience study included four diverse categories (automotive, gender, family and mobile) using various targeting data – from interest in the topic to in the market for the product or household make up to try to make the study as wide-ranging as possible. The aggregated findings combine all categories so that it isn’t just based on one. Within the wider research, the qualitative work discussed various brands and ads that viewers mentioned and that came up during in-home observations.

You found that creative is of the utmost importance. Did you speak to any creative people within your expert panel, and if so what did they have to say?

Phoebe Casey-Miller:    As creative was something which came up spontaneously as an important theme (rather than an initial objective), the interviews spanned media agencies, advertisers and industry bodies. There are clearly more interesting ideas to explore here. More details on viewer perceptions can be found in the whitepaper.

Do you think this research shows anything significantly different to what has already been presented to the market in regards to BVOD / Addressability?

Harry Harcus:    Yes, the research stands out in two ways. First, the high level of focus on Addressability in TV. Secondly, it has deployed such a wide range of robust research techniques and scientific experimentation. One major takeaway from the research is that there is still so much to learn and understand evolving TV viewing habits and the effectiveness of advertising on those viewers.

If I can work directly with broadcasters to create bespoke contextual creative, which has been the case for the past 20 years why is addressable now the answer to targeting? This sort of thing has been the driving force behind channels such as Cartoon Network and the Viacom channels for years.

Kristian Claxton:  It could be said that addressability and contextual creative are two separate things, but when used in combination can be extremely powerful. In order for contextual (using the question as the example) to be impactful, it means that you have to be around specific content that is relevant to your advertisement – this works where the genre of programming fits perfectly for relevant brands (e.g HASBRO for Cartoon Network). Where addressability is different is it uses the audience as the proxy (HH data), to determine if whether or not to show a brands advertisement e.g Luxury shopper, Pet food buyer). Addressable advertising doesn’t make the assumption that “you are what you watch”, instead it takes relevant data from companies such as Experian, MasterCard and nectar to determine the audience definition. We are now starting to see clients use contextually relevant creative to these addressable audiences (e.g. dealership messages to auto intenders within a 30 minute drive time) – this is when creative and addressability can we be used in an extremely powerful way.

Why are we unable to use first-party data sets to target in a “pure” sense regardless of scale? There’s a heavy premium for addressable TV and if you use e.g. Mastercard 1st party data to identify a target audience based on transactional data it appears the best that can then happen is to build lookalikes from that to deliver impressions or delivering postal sectors with high indices. This undermines the value of the first-party data and the point of spending significant premiums for the e.g. Mastercard data in the first place undermining the sell to clients.

Clive Page:  First party datasets can be targeted within addressable TV. It is true to say that direct matches are not easy, as there are some limitations imposed upon the platforms by the suppliers privacy positions and by the regulatory environment (GDPR). However, even at the current level of availability, there are significant benefits in accuracy and precision that can be realised from the use of client and advertiser customer data.