Editor’s Note: YuMe commissioned Nielsen to conduct the study. The original version of this article implied that it was a joint study by YuMe and Nielsen.
Consumers may tune into TV on a regular basis, but they mentally tune out rather fast in favor of the array of "second screen" options.
On average, television holds a consumer’s attention only 39% of the time -- a rate that pales in comparison to the attention rates that laptops (70%), tablets (76%) and smartphones (77%) command.
That’s according to a new report from Nielsen and YuMe, a digital video ad tech firm that commissioned the study, that is expected to be released Wednesday morning. MediaDailyNews got an early look at the data.
Over a two-month period, Nielsen and YuMe conducted in-lab observations on 200 consumers in Las Vegas. The consumers were told to engage with any of the devices (TV, smartphone, tablet and laptop) as they would at home for 20 minutes, and their actions were recorded. Nielsen and YuMe ended their experiment with 50 hours of video footage, and they claim the footage was then analyzed “second-by-second” to measure consumer attentiveness.
The television was on more than half the time (53%) during the experiment -- tops among all screens. Laptops (48%) were second, followed by tablets (38%) and smartphones (17%).
Tablets and smartphones are both passively "on" at all times, sending users notifications and vibrations as alerts that something new is happening. That passive "on" mode -- compared to televisions and laptops, which are of little to no use when off -- may help partially explain why the TVs and laptop screens were turned on for significantly more time.
On the flip side, TVs have a passive effect once they are on. After a consumer turns the TV on and chooses a channel, there is little interaction required. On a smartphone, unless a video is being watched, constant interaction is required if the consumer wishes to engage with new content.
Perhaps that’s why consumer attention to television rapidly deteriorated shortly after the screen was turned on during the test. Nielsen and YuMe note that attention to television dropped from over half in the first four minutes to under 20% in the final 16 minutes.
In addition to overall attentiveness, the report also notes how much consumers paid attention to ads on the respective screens.
When multitasking, consumers become even more focused on the “second screen.” In multi-tasking situations -- defined as situations in which consumers had at least two screens on at the same time -- ads on television were only paid attention to 30% of the time, compared to 71% on laptops, 93% on tablets and 100% on smartphones.
Paul Neto, director of research at YuMe, acknowledged that the smartphone sample size was low, and that smartphones are likely more similar to tablets.
“Ad load was not controlled during the experience, thus they would occur as they naturally do on the devices being used,” Neto said to MediaDailyNews. “Ad attention is when an ad occurs while they were paying attention to the device thus in attention view. For example, respondents were paying attention to the television 39% of the time it was on, and during that time, 30% of total ads were seen on average. Thus, 70% of ads were missed as the respondents attention was elsewhere.”
Neto said attention was defined “as when the respondent [was] looking at the screen.”
“No one is debating that consumers are multitasking. This ethnographic study was specifically designed to garner insights into users’ behaviors and preferences while multitasking,” Neto said in an earlier statement. “Despite distraction levels among consumers, it will be important for brand advertisers to continue running campaigns cross-screen, as viewers continue to show they are more attentive on laptops, tablets, and/or smartphones while ‘watching’ TV.”
The study also found there’s significantly more multitasking done in group settings than when alone; consumers multi-task four times more often when with other people.
Nielsen and YuMe also found that Millennials multitask 33% more than those over 35. However, among all age groups, the report found no significant relationship between gender and multitasking habits.
The report is expected to be released Wednesday morning at Videonomics’ ADAPT Summit in Scottsdale.
This type of finding is the expected one considering the "lab" context and the research design. It has little to do with the way ads or commercials are viewed in real world audience situations, however.
It is most unfortunate that Nielsen is apparently not ENGAGED in the basic work of producing accurate, reliable and useful TV ratings. Today, Nielsen seems to propose a business diet of cupcake icing at the expense of intellectual nutrition and commercial rigor. Arthur Nielsen would have said that that operating principle is always a mistake for research providers and clients in need of working knowledge. In fact, Mr. Nielsen urged his colleagues to "Employ every economy consistent with thoroughness, accuracy and reliability." Now, Nielsen's idea of ECONOMY is what some business wonks used to call "Work Out." Pass the critical responsibilities and hard work of panel management down to those who lack the discernment and discipline to know what RESEARCH QUALITY really is. Nielsen has would seem to have turned it's customers and respondents into lab rats that await the next cheesy research innovation such as the conversion of set-meter households into people-meter households at the wave of a magical wand worthy of Ollivander and Gregorovitch. In fact, as I read the study results, I am reminded of Ollivander's words to Harry Potter: “Curious indeed how these things happen. The wand chooses the wizard, remember...I think we must expect great things from you, Mr. Potter... After all, He-Who-Must-Not-Be-Named did great things — terrible, yes, but great.”
Onwards & upwards
Looks like pressure on TV pricing is on all angles, regardless how much real truth the reality is. Personally I have CNN or CNBC turn on, in the background all day long, and I looked over at the TV when there is some breaking news, so I am incline to think that such behavior is increasing since a decade or two ago.
Vegas...perfect place to gauge what people are paying attention to...
No doubt, Nielsen will shortly claim that it's work has been misrepresented. Nonetheless, the damage has been done as the monopoly seeks to be all things to all clients. It's just not possible, nor is it right. As Albus Dumbledore said "It is our choices… that show what we truly are, far more than our abilities."
Something doesn't add up here. n=200. 20 minute duration. Total of 4,000 respondent minutes. Equates to 66 hours 40 minutes. Yet only 50 hours of video was captured, or only three quarters of what was planned. Setting that issue aside, 'engagement' (which seems to be measured as viewing duration) is generally correlated to content quality and duration - that is you will look more at something 'good' but the longer it goes for the greater the chance you will drift away. I wonder if either of these factors were controlled for in the experiment design, or factored into the analysis. Also, hardly surprising that the 'second screen' grabbed higher attention ... it is the 'new screen' that has just been introduced to the viewing environment. Try the experiment in reverse (primary screen is the tablet and second screen is the TV) and see what happens.
Even intuitively these facts seem right. Don't we all turn on TV and mess around with our laptops, tablets, smartphones? The attention span - isn't it what the advertisers were looking for? They should thank YuMe and Nielsen for this study. Think about this line in the article "...respondents were paying attention to the television 39% of the time it was on, and during that time, 30% of total ads were seen on average. Thus, 70% of ads were missed as the respondents attention was elsewhere..." Quite possible I would say.
YuMe's calling this an "ethnographic" study, when it is described as a lab study, casts some doubt on the conclusions.
Ram, do you really believe that this lab-style study replicates the typical TV viewing scenario of a normal person?
Ed P, I think the lab study replicates real life. Look at this line "...ads on television were only paid attention to 30% of the time, compared to 71% on laptops, 93% on tablets and 100% on smartphones.." Even in real world watching this happens because the smartphones are more in your face and TVs are out there few feet way and the other devices are kind of in-between. So we tend to ignore the farther devices when more than one device is on. TV is the content-and-ad-only devices, whereas the other devices show content-ad as well as those distracting info like messages, emails, apps etc.
Ram, let me approach this in another way. According to Nielsen, an average person spends 5+ hours watching TV daily, including daytime, early and late evening, as well as prime. How much of this viewing time do you think the study in question replicated...approximately?
Does anyone know the actual method of determining 'attention' ? Did they wear 'eye-gaze; glasses ? Maybe they were video recorded and 'post-hoc' the footage was then analysed frame-by-frame - and if so, how were the screen images captured? I suspect the former. While such a system works well - physiologically - it is imperative that (I) the respondent must know NOTHING about the purpose of the research (2) that it be done in-situ (i.e. their own home) (3) that the initial footage (generally the first 10-15 minutes) is discarded as it is the period the respondent was getting used to the device (4) that the device is properly calibrated (across all distances) to capture eye-gaze (5) and then what minimum gaze duration was used in the analysis phase.
John, I am always suspect of "custom studies" of this nature, especially when they are sponsored by a party with an ax to grind. In this case we are led to believe that the "finding" that 70% of the TV commercials were skipped, or totally ignored, with the implication that this is typical of real world viewing behavior.And, sure enough, that's what the article seems to be saying in its headline and opening paragraphs. Yet, the same company---Nielsen----operates another division which measures commercial recall and other ad-impact metrics, with thousands of studies in its database---all based on real world viewing situations. According to this research, a typical TV "30" is remembered---with some degree of proof offered----by about 40-45% of the program viewers. Naturally there are highs and lows, surrounding the norms, but one has to ask how it can be that average ad recall levels are 40-45% if it is really true that only 30% of the audience even "looks at the TV screen when an average commercial is presented?
Totally agree Ed. In Australia we have another very sophisticated piece of audience research. You may have heard of it. It's called "the pub test". If you ask half a dozen of your mates in the pub does that make sense. This report fails the pub test.