When an ad should be counted as “viewable” and what impact different levels of viewability have are highly debated topics. These debates however have not benefited from a scientific perspective on the role viewability plays in actual ad effectiveness. In this session, the IPG Media Lab explores the relationship between the role of viewability and branding impact across a range of ad formats, including some never before presented insights on social video in a mobile environment.

1. © 2015 IPG Media Lab. Proprietary & Confidential
PUTTING
SCIENCE
BEHIND THE
STANDARDS

2. © 2015 IPG Media Lab. Proprietary & Confidential
HOW BIG
IS THE
VIEWABILIT
Y
PROBLEM?
2Source: Integral Ad Science
Statistic for the U.S.
57
%Of display ads are not
human/viewable

3. © 2015 IPG Media Lab. Proprietary & Confidential
WE HAVE A STANDARD TO GO BY…
3
STANDARD BANNER ADS RICH MEDIA ADS VIDEO ADS
At least 50% in view
For a minimum of 1 second
At least 30% in view
For a minimum of 1 second
At least 50% in view
For a minimum of 2 consecutive seconds

4. WE STILL DON’T KNOW HOW
EFFECTIVE ADS THAT MEET
THE MEDIA RATINGS
COUNCIL’S (MRC)
VIEWABILITY STANDARDS
ACTUALLY ARE.
4

5. 1st Media Trial
Standard Display, Rich
Media, Video on PC
2nd Media Trial
In-Feed, Auto Play Video
on Mobile
TWO ROBUST MEDIATRIALS
5

6. © 2015 IPG Media Lab. Proprietary & Confidential
OUR GOAL IS TO…
6
Present results from the first known study that scientifically ties
viewability to ad effectiveness by testing the following hypotheses:
OUR GOAL IS NOT TO…
Recommend changes to the MRC standard, or develop a new
viewability standard
1. The more viewable an ad is, the more consumers will see it
2. As viewability increases, so does ad effectiveness
3. There are strategies advertisers can employ to make less viewable ads more effective

7. © 2015 IPG Media Lab. Proprietary & Confidential
RIGOROUS SCIENTIFIC TESTING
7
Participants recruited from nationally
representative
online panel
Randomized into 1 of 189 viewability
test cells; Viewed webpage that
matched their typical consumption
habits
Subset of sample viewed webpage
while being eye-tracked
Answered post-exposure survey
for branding metrics

8. © 2015 IPG Media Lab. Proprietary & Confidential
189 DIFFERENT AD SCENARIOS WERE TESTED
8
*Two standard display ad sizes tested; 1 per industry vertical
**Logo placement test cells only included for 1 ad type for 1 brand
PERCENT IN
VIEW
TIME IN VIEW AD TYPE
INDUSTRY
VERTICAL
LOGO
PLACEMENT
AUDIO SHARE OF VIEW
CONTEXTUAL
RELEVANCE
25% .5 sec
Standard
Banner Ad
CPG Top On 1 of 1 ads In Context
30% 1 sec
Rich Media/
Large Format
Ad
Auto Not at top Off 1 of 2 ads
Out of
Context
50% 2 sec Video Ad 1 of 4 ads
75% 4 sec
100% 7 sec
Full
Exposure

9. WHAT VALUE DO ADS THAT
MEET THE MRC MINIMUM
STANDARD ACTUALLY
OFFER

10. © 2015 IPG Media Lab. Proprietary & Confidential
UNDER MRC STAN
DARD
AT MRC STANDAR
D
ABOVE MRC STAN
DARD
NATURALLY, HIGHER VIEWABILITY = MORE
EYEBALLS
10
Consumers
see ad
23%
Consumers
see ad
48%
Consumers
see ad
76%
Results include all ad types based on eye tracking data

11. © 2015 IPG Media Lab. Proprietary & Confidential
MRC STANDARD ISN’T A
MAGICAL THRESHOLD FOR AD
EFFECTIVENESS
11
* Statistically significant difference between control and test at >=90% confidence
^ Statistically significant difference between MRC standard and test at >=90% confidence
Results include all ad types
Ads that simply met the
standard did not have impact
on ad recall
+2%
+16%*^
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
AD RECALL
PERCENTCHANGEINADRECALL(DELTA)
AD RECALL BY MRC STANDARD (DELTA)
Met MRC Standard Above Standard

12. © 2015 IPG Media Lab. Proprietary & Confidential
WHICH MEANS EVEN SOME
IMPRESSIONS BELOW THE MRC
STANDARD HAVE IMPACT
12
* Statistically significant difference between control and test at >=90% confidence
^ Statistically significant difference between MRC standard and test at >=90% confidence
Results include all types
+6%*^
+2%
+16%*^
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
AD RECALL
PERCENTCHANGEINADRECALL(DELTA)
AD RECALL BY MRC STANDARD (DELTA)
Under Standard Met MRC Standard Above Standard

13. © 2015 IPG Media Lab. Proprietary & Confidential
6%*^
Under standard
BELOW MRC STANDARD BREAKOUT BY EFFECT ON
AD RECALL
MOSTLY
BECAUSE SOME
PARTIAL ADS
COULD BE IN
VIEW FOR A
LONG PERIOD
OF TIME
% VIEW-below
TIME-meets MRC
Results include all ad types
* statistically significant difference between control and test at >=90% confidence
^ statistically significant difference between % Met MRC standard and % Under standard at >=90% confidence
% VIEW-meets
TIME-below MRC
% VIEW-above
TIME-below MRC
% VIEW-
below
TIME-
Below
MRC
% VIEW-below
TIME-above MRC
16

14. © 2015 IPG Media Lab. Proprietary & Confidential
16%*^
Above standard
ABOVE MRC STANDARD BREAKOUT BY EFFECT ON
AD RECALL
ON THE FLIP
SIDE, SOME
ADS ABOVE
THE STANDARD
DON’T HAVE
IMPACT
Results include all ad types
* statistically significant difference between control and test at >=90% confidence
^ statistically significant difference between % Met MRC standard and % Under standard at >=90% confidence
% VIEW-above
TIME-above MRC
% VIEW-at
TIME-above MRC
Specifically, when time in
view is low
% VIEW-
above
TIME-
At MRC
% VIEW-
above
TIME-
At MRC
17

15. © 2015 IPG Media Lab. Proprietary & Confidential
BUT, WHAT IS MOST IMPORTANT IS THAT
VIEWABILITY IS HIGHLY RELATED TO
EFFECTIVENESS
15
* Statistically significant difference between control and test at >=90% confidence
Results include all ad types, % in view under the MRC standard excludes large format since it was not tested
5%
4%
9%*
11%*
0%
10%
20%
Under standard At MRC standard 75% 100%
PERCENTCHANGEINRECALL(DELTA)
PERCENT IN VIEW
AD RECALL BY PERCENT IN VIEW (DELTA)
Averaged Delta for All Ad Types
1%
3%
8%*
17%*
0%
10%
20%
Under standard At MRC standard 4 seconds 7 seconds
PERCENTCHANGEINRECALL(DELTA)
TIME IN VIEW
AD RECALL BY TIME IN VIEW (DELTA)
Averaged Delta for All Ad Types

16. © 2015 IPG Media Lab. Proprietary & Confidential
“TIME IN VIEW” IS KING
16
*This was only tested for Standard Banner and Video Ads, since Large Format was not tested with a lower % in view
Results include all ad types
TIME IN VIEW
PERCENTINVIEW
7 sec0
100%
EFFECT OF PERCENT AND TIME IN VIEW ON AD RECALL (DELTAS)
Percent in view higher, Time in view lower
+3.8%
Percent in view higher, Time in view higher
+14.0%
Percent in view lower, Time in view lower
+0.4%
Percent in view lower, Time in view higher*
+10.4%
AT MRC STANDARD

17. © 2015 IPG Media Lab. Proprietary & Confidential 17
LOGO AT TOP
Stronger impact after MRC
viewability standard for both
time in view and % in view
AUDIO ON
3x impact for ads under the
MRC viewability standard
LOW AD
CLUTTER
200% more impact once time
in view exceeds the standard

18. © 2015 IPG Media Lab. Proprietary & Confidential
PERSONAL TWITTER FEED
Users logged into their own
Twitter page and saw real
content from their feed, along
with the test ad
IN-FEED, AUTO PLAY VID
TWITTER PROMOTED V

19. © 2015 IPG Media Lab. Proprietary & Confidential
R² = 0.852
0 1 2 3 4 5 6 7 8 9 10
%WhoRecalled
Viewability Index
Aided Ad Recall (Δ)
Linear (Aided Ad Recall (Δ))
The Viewability Index Score is an algorithm that takes into account both Time in View and Percent in
View and creates a composite value. The values are linear, so that an indexed score of four
represents an ad that is approximately twice as viewable as a score of two.
VIEWABILITY
MATTERS!
AIDED AD RECALL BY VIEWABILITY
INDEX

20. © 2015 IPG Media Lab. Proprietary & Confidential
1 SEC
3 SEC
4 SEC
6 SEC
All Unaided and Aided test-control deltas are significant
Control n=400, 1 sec n=241, 3 sec n=212, 4 sec n=175, 6 sec = 138, 15+ sec n=49
RECALL METRICS BY “TIME IN VIEW”
42%
55% 55%
60%
72%
49%
64%
60%
66%
73%
23%
31%
38%
35%
51%
Time at 100% in View
Unaided Ad Recall (Δ) Aided Ad Recall (Δ) Message Recall (%)
15+ SEC
THE LONGER
IN VIEW, THE
HIGHER THE
AWARENESS

21. © 2015 IPG Media Lab. Proprietary & Confidential
In-Feed, Auto-Play Video: Control n=400, 1 sec n=241, 3 sec n=212, 4 sec n=175, 6 sec = 138, 15+ sec n=49
2015 Webpage: Control n=103, 1 sec n=103, 2 sec n=104, 4 sec n=104, 7 sec n=101, Full Exposure n=103
VIEWABILITY BY VIDEO AD TYPE
AIDED RECALL (Δ)
COMPARING
VIEWABILITY
RESULTS
ACROSS
STUDIES
SEC
1%
5%
13%
23%
49%
64%
60%
66%
1 3 5 7 9 11 13 15
Time at 100% in View
2015 Webpage Study (PC, video ad, sound on)
In-Feed, Auto-Play (Smartphone, video ad within social media post, sound off)
73%
60%
SEC SEC SEC SEC SEC SEC SEC

22. © 2015 IPG Media Lab. Proprietary & Confidential
In-Feed, Auto-Play Video: Control n=400, 1 sec n=241, 3 sec n=212, 4 sec n=175, 6 sec = 138, 15+ sec n=49
2015 Webpage: Control n=103, 1 sec n=103, 2 sec n=104, 4 sec n=104, 7 sec n=101, Full Exposure n=103
VIEWABILITY BY VIDEO AD TYPE
AIDED RECALL (Δ)
COMPARING
VIEWABILITY
RESULTS
ACROSS
STUDIES
SEC
1%
5%
13%
23%
49%
64%
60%
66%
1 3 5 7 9 11 13 15
Time at 100% in View
2015 Webpage Study (PC, video ad, sound on)
In-Feed, Auto-Play (Smartphone, video ad within social media post, sound off)
73%
60%
SEC SEC SEC SEC SEC SEC SEC
OBVIOUS BRANDING
BEGINS (TWEET TEXT,
HANDLE, USERNAME
ABOVE VIDEO)
OBVIOUS BRANDING
BEGINS (WITHIN VIDEO)

23. © 2015 IPG Media Lab. Proprietary & Confidential
23
OBVIOUS
BRANDING
APPEARS BEFORE
“PERCENT IN VIEW”
MEASUREMENT
STARTS

24. © 2015 IPG Media Lab. Proprietary & Confidential
TO SUMMARIZE….
24

25. THE MRC
STANDARDS AREN’T
MEANT TO
GUARANTEE AD
EFFECTIVENESS.
THAT’S
THE AD’S
JOB.
25

26. ADS THAT EXCEED THE STANDARD
END UP HAVING THE GREATEST
IMPACT.
VIEWABILITY IS IMPORTANT,
BUT IS NOT THE END-ALL-BE-
ALL. IT ISN’T A KPI.
SO…
WHILE WE CAN’T COMPLETELY
CONTROL VIEWABILITY, WE CAN
CONTROL THE CREATIVE &
WHERE IT RUNS.

27. © 2015 IPG Media Lab. Proprietary & Confidential
THANK YOU