IMPACTS OF SCROLL VELOCITY
FEED READ VS.
I MPA C T S O F SC R OLL VE LOC I TY
AJ MATHEW, VP RESEARCH, KARGO
ABSTRACT The digital ecosystem brings speed and convenience to consumers’ lives and enables advertisers to promote products and services to them across a variety of platforms. The growing digital, and more specifically mobile, ad industry has to adapt to these new distribution channels with an eye on how these environments impact ad effectiveness. Today, advertisers face challenges that were not necessarily issues in the days of traditional media. For example, in TV or print advertising an advertiser had to worry about their ad within that one medium, reducing the variables to consider. In the mobile world, advertisers face many variables—multiple operating systems, platforms and distribution channels— that can impact the delivery, attention, and effectiveness of their marketing efforts. Add to those challenges the varying mindsets consumers bring to specific mobile activities across these environments, and we have to ask—how does it all impact ad performance?
TELEVISION
DESKTOP
TABLET
MOBILE
ANDROID
APP 3
IPHONE
MOBILE SITE
APP 2
APP 1
MOBILE
TELEVISION
FIA
ANR
AMP
SNAPCHAT
APP 3
MOBILE SITE
APP 2
APP 1
APPLE WATCH
DESKTOP
BACKGROUND Smartphones are highly task-oriented devices—users engage with them via specific platforms to pursue specific activities. It’s likely these mobile environments impact ad engagement and effectiveness. Users’ mindsets and intentions within specific environments can influence whether an ad was given any attention or remembered. Therefore, behavioral and attitudinal measures were utilized to determine the level of visibility, attention, and effectiveness of ads within editorial, social, and search environments to help answer the following: 1. How visible are ads in these different environments? 2. How many ads are actually seen by the human eye in each environment? 3. How well are ads in these environments remembered?
METHODOLOGY Partnering with Nielsen, 100 smartphone users were recruited to experience premium editorial, social, and search content in a live in-market test. Each participant was given 10 minutes within each environment to experience the content as they normally would in a use session. Eye tracking and post-exposure survey tools were used at Nielsen’s Media Lab facility at Television City in Las Vegas, NV. Testing/Analysis Summary Devices: Participants’ personal smartphones: iOS & Android sample representative of the US mobile population1 Premium Editorial Content: A simgle publisher’s mobile website (same publisher for all, freedom to explore at participants’ leisure) Social: Participant’s own social media feeds Search: Cued search categories, freedom to explore at participants’ leisure Analysis: Findings are based on 90% statistical significance
EXECUTIVE SUMMARY > Significantly more fully-visible ads appear in long-term reading environments than in social feeds. > Editorial remains the visibility leader as time progresses. More fully-visible ads remain on screen over time, while significantly less of them stay on screen in social during the same time span. > Fully visible ads are visible in editorial content for an average of 44 seconds, compared to 6 seconds in search, and 3 seconds in social. > Overall ad recall is competitive among all tested environments, however brands in editorial in-article ad units are remembered significantly more than those in social. > Time viewed and ad recall suggest that editorial is the most efficient environment for media investment on mobile, with ads needing much less view time to be remembered.
USAGE & ATTENTION Consumers state that they are using all three environments for expected purposes: > Editorial is used most to catch up on current events and learn about breaking news > Social is used to learn different perspectives on news and to kill time > Search is used to look up information and discover new products EDITORIAL (A) 70% BC
To catch up on current events
69% BC
To learn about breaking news To hear different perspectives on news stories
35%
To kill time when bored
33%
To look up information about products or brands To discover new products or brands
SOCIAL (B) 47%
SEARCH (C) 54%
35%
54%
50% AC 85% AC
32% 43%
14% A
2%
21% A
5%
B
93% AB 67%
AB
18-34 y.o. 67% 35-54 y.o. 33%
In our results, we saw a notable distinction in how younger and older demographics use social for seeking out different perspectives on news stories—67% of 18-34 year olds vs. 33% of 35-54 year olds. That latter percentage aligns more closely with how all ages use editorial and search for the same purpose.
VISIBILITY Ad visibility tells us if consumers had the opportunity to see an ad. In this study eye tracking tools provided the opportunity to measure what was on each smartphone screen and for how long. ANY VISIBILITY VS. FULL VISIBILITY All three environments succeeded in displaying at least partially visible ads for at least a second. This means that only some of the ad is on screen and not necessarily key parts like messaging and branding. Therefore, our focus remained on “fully visible” ads. An ad being fully visible on screen assures that a consumer had the ability to see all branding, imagery and messaging. When evaluating ads that were 100% visible on screen, differences emerged. Editorial was more successful in displaying fully-visible ads, with nearly all (96%) ads being 100% on-screen. Meanwhile almost 4 out of 5 ads on social and 9 out of 10 on search were fully visible.
IMPACTS OF SCROLL VELOCITY Time is, of course, important for marketers to communicate their message to consumers. Differences between ads in each environment become more apparent as we evaluate visibility based on time. Significant differences emerged between fully-visible ads in each environment that are present for 1+ seconds and 2+ seconds. Fully-visible ads in editorial maintain visibility, with 92% and 82% of ads being fully visible for 1+ and 2+ seconds, respectively. Ads in social experience a significant visibility drop off, with only 44% of ads fully visible for 1+ seconds, and less than a third (32%) visible for 2+ seconds. These findings reflect the differences in consumer scroll speeds between editorial content and social feeds. 99% 98% 98% 82% 67%
32% VISIBLE 1+ SEC
FULLY VISIBLE
FULLY VISIBLE 1+ SEC
EDITORIAL
SOCIAL
SEARCH
FULLY VISIBLE 2+ SEC Statistical difference with editorial at 90% CI
TIME IN VIEW The scroll velocity factor of these environments not only results in different volumes of full ads on-screen, but also the length of time that ads are visible. Ads in editorial are at least partially visible for an average of 37 seconds. Ads that are fully visible in editorial are on screen for an average of 44 seconds. There are significant differences when compared to ads in social and search. 44
AVERAGE SECONDS
37
10
9
6 3
EDITORIAL
SOCIAL VISIBILE
SEARCH
EDITORIAL
SOCIAL FULLY VISIBLE
SEARCH
AD VIEWS Once an ad is visible, we must determine if it was seen by human eyes. Eye tracking tools provided data to demonstrate the varying levels of visual attention that ads received within the tested environments. There are distinct variables that impact natural visual attention. People visit editorial sites to consume journalistic content. Ads there are surrounded by the sought-after content, which occupies visual attention and makes it easier to consciously ignore them. Contrarily, ads within social feeds are not only blend in with a user’s regular feed, but they also take up much of the screen space, and in some cases, extend beyond the screen borders. Therefore, it’s nearly impossible for a user’s eyes to miss the ad. Search is based on a user actively looking for something, so their eyes will naturally skim through the search results—and the ads seamlessly integrated within them.
This is clearly observed in the eye-tracking data for ads that were actually viewed. Nearly three-quarters of ads in editorial were seen, significantly less than ads in social and search.
72
%
EDITORIAL
99
%
SOCIAL
93
%
SEARCH
MEMORY The field of psychology outlines for media researchers how memory works. There are three mental processes involved—encoding information, storing it and retrieving it. These are measured through recognition, cued recall and free recall.2
ENCODING
STORAGE
RETRIEVAL
RECOGNITION
CUED RECALL
FREE RECALL
Free recall, or cognitive retrieval, is the strongest measure of memory since it includes all three processes. It works without any cues to the participant, and is where brands strive to land—with their messages retrievable by stored information and remembered without any assistance. OVERALL AD RECALL
Overall, significantly more ads in search were recalled than in editorial or social—an expected outcome given that ads in search are delivered based on a user’s explicit search terms, creating a strong association between sought-after information and branded messaging (cognitive storage phase of memory).
5% AB 3% 2%
EDITORIAL (A)
SOCIAL (B)
SEARCH (C)
Letters indicate significant difference at 90% confidence
There was an exception, however. In a 2015 study by Kargo and Media Science, adhesion ad units—those that appear and stay at the bottom of the screen as users scroll down the page— received significantly lower visual attention compared to in-article units—those that appear within the flow of editorial content. In addition, in-article ads received the lowest biometric response among other tested ad units in editorial—indicative of a better user experience.3 % LOOKED
TIME LOOKED (SEC)
# OF FIXATIONS
# OF REVISITS
4.17
38%
ADHESION IN-ARTICLE
3.45
3.13
63%
4.23
2.67 1.64
ADHESION IN-ARTICLE
ADHESION IN-ARTICLE
ADHESION IN-ARTICLE
When isolating editorial in-article units for recall in the current study analysis, there was significantly higher recall for ads in social, and alignment with recall for ads in search.
AD RECALL (EDITORIAL IN-ARTICLE) 6% 5%
2%
EDITORIAL IN-ARTICLE(A)
SOCIAL (B)
SEARCH (C)
Letters indicate significant difference at 90% confidence
EFFICIENCY
AD RECALL (EDITORIAL IN-ARTICLE)
Previous research shows that viewers spend roughly 3 seconds viewing an ad.4 By analyzing both the time an ad was viewed, and its ability to be remembered, we can evaluate the level of efficiency of different environments. The editorial environment proves to be very efficient compared to social and search, requiring much less time to pass a message to users. Editorial averaged 1.6 seconds for ads that were successfully recalled, compared to nearly 11 seconds for social and 2.8 seconds for search.
1.6s EDITORIAL
2.8s SEARCH
10.7s SOCIAL
ADVERTISING PERCEPTIONS Environments effect the perceptions of brands that advertise within them. In this study, brands that advertise in editorial environments are perceived as reliable, high quality, intelligent, and trustworthy. Meanwhile, brands advertising in social are associated with being fun, youthful and cool.
EDITORIAL (A) 52%
Reliable Intelligent
46%
BC
Trustworthy
46%
BC
37%
High Quality Fun Youthful Cool
SOCIAL (B) BC
18% B
4% 9%
12% 15% B
2% 7%
21% B 62% AC
5% 6% 4%
SEARCH (C)
56% 48%
AC
AC
6% 2% 10%
CONCLUSION Editorial ad environments provide an exceptional value to marketers. As seen in comScore’s 2016 study on the halo effect of premium publishers, “the overall effectiveness of an ad is the combined effect of its increased likelihood of being seen and the value of its surrounding context…ads appearing on premium publishers were significantly more effective in driving brand lift. While some of this effect was due to higher ad viewability on premium sites, the more significant driver was the halo effect of appearing on these sites.”5 The visibility, or average length of time on screen, for ads in editorial helps brands communicate their message. These premium environments prove to deliver not only longer visibility times, but also the greatest efficiency when it comes to recall. It takes consumers less view time to remember ads delivered in editorial than it does in social or search. Millward Brown Digital’s research showing that viewers spend roughly 3 seconds viewing an ad (Millward Brown Digital Eye Tracking Norms, 2012), aligns with findings from research by Kargo and MediaScience ( Captivate vs. Aggravate, 2016). With Feed vs. Read, we have learned that editorial has a higher percentages of ads being fully visible for over 2 seconds after inception and that it takes less than 2 seconds of view time to recall ads in that environment—all substantiating the value of advertising in premium editorial. comScore: Mobile Metrix, 2016 MediaScience, 2008 3 Kargo & MediaScience: Captivate vs. Aggravate, 2016 4 Millward Brown Digital: Eye Tracking Norms, 2012 5 comScore: “The Halo Effect: How Advertising on Premium Publishers Drives Higher Ad Effectiveness,” 2016 1
2
BIG IDEAS FOR SMALL SCREENS
I MPA C T S O F SC R OLL VE LOC I TY
AJ MATHEW, VP RESEARCH, KARGO
ABSTRACT The digital ecosystem brings speed and convenience to consumers’ lives and enables advertisers to promote products and services to them across a variety of platforms. The growing digital, and more specifically mobile, ad industry has to adapt to these new distribution channels with an eye on how these environments impact ad effectiveness. Today, advertisers face challenges that were not necessarily issues in the days of traditional media. For example, in TV or print advertising an advertiser had to worry about their ad within that one medium, reducing the variables to consider. In the mobile world, advertisers face many variables—multiple operating systems, platforms and distribution channels— that can impact the delivery, attention, and effectiveness of their marketing efforts. Add to those challenges the varying mindsets consumers bring to specific mobile activities across these environments, and we have to ask—how does it all impact ad performance?
TELEVISION
DESKTOP
TABLET
MOBILE
ANDROID
APP 3
IPHONE
MOBILE SITE
APP 2
APP 1
MOBILE
TELEVISION
FIA
ANR
AMP
SNAPCHAT
APP 3
MOBILE SITE
APP 2
APP 1
APPLE WATCH
DESKTOP
BACKGROUND Smartphones are highly task-oriented devices—users engage with them via specific platforms to pursue specific activities. It’s likely these mobile environments impact ad engagement and effectiveness. Users’ mindsets and intentions within specific environments can influence whether an ad was given any attention or remembered. Therefore, behavioral and attitudinal measures were utilized to determine the level of visibility, attention, and effectiveness of ads within editorial, social, and search environments to help answer the following: 1. How visible are ads in these different environments? 2. How many ads are actually seen by the human eye in each environment? 3. How well are ads in these environments remembered?
METHODOLOGY Partnering with Nielsen, 100 smartphone users were recruited to experience premium editorial, social, and search content in a live in-market test. Each participant was given 10 minutes within each environment to experience the content as they normally would in a use session. Eye tracking and post-exposure survey tools were used at Nielsen’s Media Lab facility at Television City in Las Vegas, NV. Testing/Analysis Summary Devices: Participants’ personal smartphones: iOS & Android sample representative of the US mobile population1 Premium Editorial Content: A simgle publisher’s mobile website (same publisher for all, freedom to explore at participants’ leisure) Social: Participant’s own social media feeds Search: Cued search categories, freedom to explore at participants’ leisure Analysis: Findings are based on 90% statistical significance
EXECUTIVE SUMMARY > Significantly more fully-visible ads appear in long-term reading environments than in social feeds. > Editorial remains the visibility leader as time progresses. More fully-visible ads remain on screen over time, while significantly less of them stay on screen in social during the same time span. > Fully visible ads are visible in editorial content for an average of 44 seconds, compared to 6 seconds in search, and 3 seconds in social. > Overall ad recall is competitive among all tested environments, however brands in editorial in-article ad units are remembered significantly more than those in social. > Time viewed and ad recall suggest that editorial is the most efficient environment for media investment on mobile, with ads needing much less view time to be remembered.
USAGE & ATTENTION Consumers state that they are using all three environments for expected purposes: > Editorial is used most to catch up on current events and learn about breaking news > Social is used to learn different perspectives on news and to kill time > Search is used to look up information and discover new products EDITORIAL (A) 70% BC
To catch up on current events
69% BC
To learn about breaking news To hear different perspectives on news stories
35%
To kill time when bored
33%
To look up information about products or brands To discover new products or brands
SOCIAL (B) 47%
SEARCH (C) 54%
35%
54%
50% AC 85% AC
32% 43%
14% A
2%
21% A
5%
B
93% AB 67%
AB
18-34 y.o. 67% 35-54 y.o. 33%
In our results, we saw a notable distinction in how younger and older demographics use social for seeking out different perspectives on news stories—67% of 18-34 year olds vs. 33% of 35-54 year olds. That latter percentage aligns more closely with how all ages use editorial and search for the same purpose.
VISIBILITY Ad visibility tells us if consumers had the opportunity to see an ad. In this study eye tracking tools provided the opportunity to measure what was on each smartphone screen and for how long. ANY VISIBILITY VS. FULL VISIBILITY All three environments succeeded in displaying at least partially visible ads for at least a second. This means that only some of the ad is on screen and not necessarily key parts like messaging and branding. Therefore, our focus remained on “fully visible” ads. An ad being fully visible on screen assures that a consumer had the ability to see all branding, imagery and messaging. When evaluating ads that were 100% visible on screen, differences emerged. Editorial was more successful in displaying fully-visible ads, with nearly all (96%) ads being 100% on-screen. Meanwhile almost 4 out of 5 ads on social and 9 out of 10 on search were fully visible.
IMPACTS OF SCROLL VELOCITY Time is, of course, important for marketers to communicate their message to consumers. Differences between ads in each environment become more apparent as we evaluate visibility based on time. Significant differences emerged between fully-visible ads in each environment that are present for 1+ seconds and 2+ seconds. Fully-visible ads in editorial maintain visibility, with 92% and 82% of ads being fully visible for 1+ and 2+ seconds, respectively. Ads in social experience a significant visibility drop off, with only 44% of ads fully visible for 1+ seconds, and less than a third (32%) visible for 2+ seconds. These findings reflect the differences in consumer scroll speeds between editorial content and social feeds. 99% 98% 98% 82% 67%
32% VISIBLE 1+ SEC
FULLY VISIBLE
FULLY VISIBLE 1+ SEC
EDITORIAL
SOCIAL
SEARCH
FULLY VISIBLE 2+ SEC Statistical difference with editorial at 90% CI
TIME IN VIEW The scroll velocity factor of these environments not only results in different volumes of full ads on-screen, but also the length of time that ads are visible. Ads in editorial are at least partially visible for an average of 37 seconds. Ads that are fully visible in editorial are on screen for an average of 44 seconds. There are significant differences when compared to ads in social and search. 44
AVERAGE SECONDS
37
10
9
6 3
EDITORIAL
SOCIAL VISIBILE
SEARCH
EDITORIAL
SOCIAL FULLY VISIBLE
SEARCH
AD VIEWS Once an ad is visible, we must determine if it was seen by human eyes. Eye tracking tools provided data to demonstrate the varying levels of visual attention that ads received within the tested environments. There are distinct variables that impact natural visual attention. People visit editorial sites to consume journalistic content. Ads there are surrounded by the sought-after content, which occupies visual attention and makes it easier to consciously ignore them. Contrarily, ads within social feeds are not only blend in with a user’s regular feed, but they also take up much of the screen space, and in some cases, extend beyond the screen borders. Therefore, it’s nearly impossible for a user’s eyes to miss the ad. Search is based on a user actively looking for something, so their eyes will naturally skim through the search results—and the ads seamlessly integrated within them.
This is clearly observed in the eye-tracking data for ads that were actually viewed. Nearly three-quarters of ads in editorial were seen, significantly less than ads in social and search.
72
%
EDITORIAL
99
%
SOCIAL
93
%
SEARCH
MEMORY The field of psychology outlines for media researchers how memory works. There are three mental processes involved—encoding information, storing it and retrieving it. These are measured through recognition, cued recall and free recall.2
ENCODING
STORAGE
RETRIEVAL
RECOGNITION
CUED RECALL
FREE RECALL
Free recall, or cognitive retrieval, is the strongest measure of memory since it includes all three processes. It works without any cues to the participant, and is where brands strive to land—with their messages retrievable by stored information and remembered without any assistance. OVERALL AD RECALL
Overall, significantly more ads in search were recalled than in editorial or social—an expected outcome given that ads in search are delivered based on a user’s explicit search terms, creating a strong association between sought-after information and branded messaging (cognitive storage phase of memory).
5% AB 3% 2%
EDITORIAL (A)
SOCIAL (B)
SEARCH (C)
Letters indicate significant difference at 90% confidence
There was an exception, however. In a 2015 study by Kargo and Media Science, adhesion ad units—those that appear and stay at the bottom of the screen as users scroll down the page— received significantly lower visual attention compared to in-article units—those that appear within the flow of editorial content. In addition, in-article ads received the lowest biometric response among other tested ad units in editorial—indicative of a better user experience.3 % LOOKED
TIME LOOKED (SEC)
# OF FIXATIONS
# OF REVISITS
4.17
38%
ADHESION IN-ARTICLE
3.45
3.13
63%
4.23
2.67 1.64
ADHESION IN-ARTICLE
ADHESION IN-ARTICLE
ADHESION IN-ARTICLE
When isolating editorial in-article units for recall in the current study analysis, there was significantly higher recall for ads in social, and alignment with recall for ads in search.
AD RECALL (EDITORIAL IN-ARTICLE) 6% 5%
2%
EDITORIAL IN-ARTICLE(A)
SOCIAL (B)
SEARCH (C)
Letters indicate significant difference at 90% confidence
EFFICIENCY
AD RECALL (EDITORIAL IN-ARTICLE)
Previous research shows that viewers spend roughly 3 seconds viewing an ad.4 By analyzing both the time an ad was viewed, and its ability to be remembered, we can evaluate the level of efficiency of different environments. The editorial environment proves to be very efficient compared to social and search, requiring much less time to pass a message to users. Editorial averaged 1.6 seconds for ads that were successfully recalled, compared to nearly 11 seconds for social and 2.8 seconds for search.
1.6s EDITORIAL
2.8s SEARCH
10.7s SOCIAL
ADVERTISING PERCEPTIONS Environments effect the perceptions of brands that advertise within them. In this study, brands that advertise in editorial environments are perceived as reliable, high quality, intelligent, and trustworthy. Meanwhile, brands advertising in social are associated with being fun, youthful and cool.
EDITORIAL (A) 52%
Reliable Intelligent
46%
BC
Trustworthy
46%
BC
37%
High Quality Fun Youthful Cool
SOCIAL (B) BC
18% B
4% 9%
12% 15% B
2% 7%
21% B 62% AC
5% 6% 4%
SEARCH (C)
56% 48%
AC
AC
6% 2% 10%
CONCLUSION Editorial ad environments provide an exceptional value to marketers. As seen in comScore’s 2016 study on the halo effect of premium publishers, “the overall effectiveness of an ad is the combined effect of its increased likelihood of being seen and the value of its surrounding context…ads appearing on premium publishers were significantly more effective in driving brand lift. While some of this effect was due to higher ad viewability on premium sites, the more significant driver was the halo effect of appearing on these sites.”5 The visibility, or average length of time on screen, for ads in editorial helps brands communicate their message. These premium environments prove to deliver not only longer visibility times, but also the greatest efficiency when it comes to recall. It takes consumers less view time to remember ads delivered in editorial than it does in social or search. Millward Brown Digital’s research showing that viewers spend roughly 3 seconds viewing an ad (Millward Brown Digital Eye Tracking Norms, 2012), aligns with findings from research by Kargo and MediaScience ( Captivate vs. Aggravate, 2016). With Feed vs. Read, we have learned that editorial has a higher percentages of ads being fully visible for over 2 seconds after inception and that it takes less than 2 seconds of view time to recall ads in that environment—all substantiating the value of advertising in premium editorial. comScore: Mobile Metrix, 2016 MediaScience, 2008 3 Kargo & MediaScience: Captivate vs. Aggravate, 2016 4 Millward Brown Digital: Eye Tracking Norms, 2012 5 comScore: “The Halo Effect: How Advertising on Premium Publishers Drives Higher Ad Effectiveness,” 2016 1
2
BIG IDEAS FOR SMALL SCREENS
