Comparing media: Apples to oranges
Right now there is no simple way for a marketer to compare the value of an advertising spot on television to radio, or to social media, or to the many other paid media options available to advertisers. Research to date has provided limited guidance on which media advertisers should use now, when media choices have become increasingly complex (e.g. more channels, more devices, more ad formats).
Lack of comparability and commonality of media metrics has been a long-standing industry-wide concern (Council for Research Excellence, 2015). Many metrics are unique to a single medium (e.g. circulation figures for print, click-through rates for internet), which prevents direct comparisons across media. Even when similar metrics are used, they are often based on data collected with different methods (e.g. self-report versus passive measurement). More troubling is the lack of common definitions for important metrics. Terms such as impressions, exposures and opportunity-to-see (OTS) can mean different things for different media, stemming from how the audience data is collected and classified, for example, time-spent-with-the-ad. Current standards for digital media identify an OTS for desktop display ads when at least half of the ad’s pixels are on-screen for at least one second, whereas for desktop video ads, at least two seconds of video must be played (Interactive Advertising Bureau, 2015; Media Rating Council, 2014)¹. Meanwhile, an OTS for TV ads is only counted when the ad is played from start to finish by the broadcaster. Such inconsistencies make it difficult for marketers to gauge the relative performance of an advertising exposure in one medium compared to another. This then makes it nearly impossible to assess the relative value of money paid for those exposures (i.e. “what is a spot worth?”).
Most media is purchased on an OTS basis, which means advertisers pay for the potential to reach consumers. This approach to selling/buying media raises two key questions to consider:
- How many of the OTS purchased in a particular media will catch the attention of viewers and be seen or heard (i.e. real exposures²)?
- Then, what is the impact of these real exposures?
It is obvious that watching a video is a different experience to listening to radio, which is different again from seeing an online display ad – but what is less obvious is the impact that these different exposures will have on consumer memory and behaviour.
Tapping advertisers’ knowledge
Despite these challenges, marketers and media agencies must still regularly make media choices; they must consider their options based on incomparable or incomplete information. Outcomes of these decisions, if measured in some way, create feedback that will influence future decisions. Marketers inevitably develop some understanding of, or a decision framework about, media performance which shapes how they work and how the industry works as a whole. We set out to study the perceptions of media decision-makers from both sides of the media buying and selling process (i.e. brand/advertiser and media/advertising agency perspectives). From this point onwards we simply refer to these decision-makers collectively as advertisers.
Finding common ground: Using CPM as a framework
Cost Per Mille (CPM) is a useful metric because it standardises OTS across media as the amount paid to potentially deliver an ad 1,000 times.
We sought to capture advertisers’ perceptions of the two components of CPM that drive media performance and determine true value:
- How many OTS are actually seen/heard? Higher conversion rates³ means greater value.
- Of the OTS converted to real exposures, what is the impact of those real exposures? More impact means greater value.
Deconstructing the CPM metric in this way allows marketers to make more direct comparisons across media.
To address these two questions and give insight into the relative value of different media, we spoke to more than 100 advertisers from different countries. We asked about a comprehensive range of media, incorporating the consumer device (e.g. desktop or mobile), specific platform (where relevant, e.g. Facebook or Instagram), and ad format (e.g. audio-visual capabilities, skipping capacity).
How often does your OTS get attention in different media?
Key Finding 1 – Advertisers believe that video-based media deliver more exposures because audiences are thought to be less distracted and video formats are thought to draw more attention.
We asked advertisers to estimate the proportion of OTS that are actually seen or heard by viewers for each media, knowing that sometimes ads will be missed because people are distracted and/or deliberately avoid ads. We also asked advertisers to explain the reasons for their estimates.
Advertisers believe that video-based media – especially cinema, non-skippable desktop pre-rolls, and primetime TV – have the capacity to gain more real exposures (i.e. are genuinely noticed by the audience) compared to static image and/or text-based media, such as magazines, or static Facebook newsfeed posts. Average conversion rates for each media are presented in Table 1.
Advertisers predominantly justified their conversion rates on the basis of audience behaviour in the advertising environment (i.e. what else people are doing at the same time), which can negatively impact how much attention is paid to the specific media. Someone who is listening to the radio when driving or cleaning up at home, for example, was generally considered a more distracted media consumer than someone who is sitting in a dark cinema with a very large screen.
There were other reasons. Format itself was a contributing factor; video-based ads were considered to be more likely to be seen and/or heard than a static format. Viewers’ ability to interact with the media type was another concern (i.e. the ease of skipping or scrolling past ads). Static online ads were perceived to be easier to skip or scroll past than other online formats, lowering conversion rates.
Table 1: Conversion rates of OTS to exposure for different media
How much is a (real) exposure worth across media?
Key Finding 2 – Exposures on primetime TV and cinema are the most impactful and, therefore, valuable; most other media are valued at about half or less, relative to these top media.
Following the conversion estimates, we then asked advertisers to place a dollar value on a fixed amount of real exposures. We provided a benchmark to make it an easier task for participants. They were presented with the following example: ‘If 1,000 validated TV exposures to a 30 second primetime ad costs $100 (the benchmark), what would you pay for 1,000 validated exposures in other media types?’ Average values for each media are shown in Table 2.
Advertisers were willing to pay a similar amount for real exposures in cinema to primetime TV, identifying these two advertising media as the most valuable exposures of all media options. These ads, once seen, are presumably considered more likely to deliver on their intended purpose. Non-skippable desktop pre-roll was valued at 60% of TV and cinema, which indicates that delivering video in an online context is perceived as less impactful by advertisers. Meanwhile, newspaper, outdoor and magazine exposures were valued at about half that of TV and cinema, and most static online exposures involving text and/or images were valued at about a third or less of TV and cinema.
Table 2: Value of 1,000 exposures across different media
To provide a quick comment on data out quality – given that cinema exposures were perceived as essentially the same as TV exposures, this indicates high face validity. Similarly, that 30 seconds of radio was valued substantially less than 30 seconds of video altogether suggests that participants understood the task that was asked of them for this study.
Advertisers assign very different values to the same media
Key Finding 3 – Large variation in advertisers’ estimates of media value indicates subjective opinions (rather than facts).
The averages we have shared are just that… averages. Ranges and standard deviations reported in Tables 1 and 2 do show high levels of variation around those averages, which indicates that advertisers have diverse ideas about which media are the most attention grabbing or most impactful, or there was lots of guessing.
Different experiences will inevitably lead to some differences in knowledge; the people we spoke to do work on different categories, have seen different campaigns working well in different media, and may access different data sources to support their ideas. To explore how a buying versus selling background might have influenced our overall results, we compared the responses from brand/advertisers to media/ad agency decision-makers. Media/ad agency participants generally gave higher conversion rates and were willing to pay more for exposures; their responses were 14 and 12 percentage points higher on average across media respectively, with larger differences for online media. The finding suggests that media/ad agency professionals are generally more positive about all media’s ability to gain attention, and also consider exposures in non-TV media to be more effective than marketers do. This appears to be a classic endowment effect, which is when people have a tendency to overvalue the things that they own.
Summary
The intent of our much larger research agenda is to provide marketers with some evidence-based values for advertising spots in different media, in different conditions, and for different product/service categories.
This particular study provides an approach to establish what media decision-makers presently think about the relative value of different media. The overarching finding being that advertisers perceive there is more value gained from advertising in video-based media than static or audio-based media.
That said, our sample of media decision-makers provided very different estimates for both conversion rates and willingness-to-pay for real exposures in different media. Though we don’t expect everyone to agree all of the time, such variability speaks to a lack of evidence-based media planning. Being evidence-based is of great importance, and is an area where there is scope for the media and advertising industry to develop better bases for decision-making.
Whenever possible, advertisers are strongly encouraged to consider media options using a comparable metric (e.g. CPM), rather than looking at each media on a case-by-case basis and relying on media-specific metrics. Doing so should prevent marketers from being distracted by the sales arguments used by media providers.
Considering these findings and the prior work by the Institute (see recommended reading list at the end of this report), we now provide our current best-practice guidelines for media placement decisions.
Guide to Media Selection
Step 1 – Reach Potential
1. What level of reach does my media schedule deliver?
Choose the largest reaching media that allows for a continuous presence over time, e.g. as many weeks on-air as possible in the planning period.
2. Will adding different media increase total campaign reach?
Consider including more than one media and/or varying the time of day when ads are aired to increase the level of reach.
Step 2 – Opportunity-to-See Versus Exposures
3. What constitutes an OTS?
Criteria for an OTS are very different across media types – one minute viewing for TV, 50% of pixels for at least two seconds for online video – so carefully consider the impact this will have on the likelihood of the ad you have paid for being seen (i.e. deliver a real exposure).
4. What is the level of clutter?
Is the ad the dominant stimuli when it is presented or how fiercely is it competing for attention (both with content and other ads)? It makes sense that when ads are larger and shown alone on screen they have a better chance of capturing attention. Assess the ad environment carefully.
5. What ad avoidance behaviours are occurring?
People avoid ads across all different media; actively, like changing the channel, or passively, like looking away or multi-tasking. Stay up-to-date with the latest research on avoidance and take the opportunity to watch people interacting with your media to give a better idea of the behaviours your ad is likely to encounter. The Institute is currently studying ad avoidance and will issue further evidence in due course.
6. What device are consumers using – e.g. TV, laptop, mobile?
There is some support for big screen effectiveness but also support for the intimacy of consumption on mobile. Think about your message and which device will provide the best context for your goals.
Step 3 – Impact on Consumer Memory
7. What is the ad format – i.e. audio and/or visual, static or dynamic?
The type of stimuli will impact the level of mental processing. Studies show that our ability to retain visual information is superior to audio. The Institute is currently studying the impact of different exposures and will issue further evidence on this in due course.
¹ Noting that a recent survey found about a third of brand and agency professionals do not adhere to these standards (Benes, 2018).
² By real exposure we mean that some part of the ad was noticed or registered by a person in the advertising environment.
³ Advertisers were asked to consider the average conversion rate of a media type rather than taking into account the impact of creative performance to grab attention.
