Understanding Audience Consumption of Media to Further Its ...

Understanding Audience Consumption of Media

to Further Its Measurement

Association of National AdvertisersJanuary 23, 2008

Richard ZackonCouncil for Research Excellence

Shari Anne BrillCarat

Michael Holmes Ball State University

Council for Research Excellence

Mission: To advance the knowledge and practice of

methodological research on audience measurement through the active collaboration of Nielsen Media Research and its clients

Council for Research Excellence

Values:

Transparency Inclusiveness Diversity Imagination Collegiality Practicality Client leadership

CRE chronology

CRE organization Chair: Mark Kaline, Ford Motor

Company Vice Chair: Mike Hess, OMD Secretary/Treasurer: Jessica Pantanini,

Bromley Communications

Steering Committee Research Committees Technical Advisory Panel

Facilitator: Richard Zackon

CRE membership Melva Benoit, FOX Shari Anne Brill, Carat Tim Brooks ARF, Lifetime Joanne Burns, 20th Television Fox Michele Buslik, AAAA Media

Research Committee; TargetCast TCM

Vicky Champlin, Anheuser-Busch Susan Cuccinello, TVB Henry DeVault, ABC, Inc. Paul Donato, The Nielsen Company Colleen Fahey Rush, MTVN

Research Nancy Gallagher, NBC Universal Don Gloeckler, P&G Bruce Goerlich, Zenith Optimedia Jean Goldberg, Warner Bros. David Gunzerath, NAB Mike Hess, OMD Worldwide

George Ivie, MRC Mark Kaline, Ford Motor Company Rick Keilty, Belo Corporation Pat Liguori, ABC Owned Television

Stations Jessica Pantanini, Bromley

Communications David Poltrack, CBS Lyle Schwartz, Mediaedge: CIA Ceril Shagrin, Univision Howard Shimmel, The Nielsen

Company Jonathan Sims, Comcast Spotlight Steve Sternberg, Magna Global Ira Sussman, CAB Jack Wakshlag, Turner

Broadcasting Alan Wurtzel, NBC Richard Zackon, CRE Facilitator

Four CRE Research Committees

Media Consumption and Engagement Committee

Co-Chairs Shari Anne Brill, CaratSteve Sternberg, Magna Global

Members:Melva Benoit, FOXTim Brooks, LifetimeJoanne Burns, 20th TV, MyNetwork TVVicky Champlin, Anheuser-BuschBruce Goerlich, Zenith Optimedia USA Jack Wakshlag, TurnerMike Pardee, Scripps NetworksDebbie Reichig, NBC Universal

Media Consumptionand Engagement

Mapping consumer media behavior: How consumers interact with different

media and how this changes over time Home versus workplace or other places Rate of consumer adoption

(early versus late adopters) What are the measurement implications?

MCE backgroundVerify or dispel commonly held beliefs:

“The 30 second spot is dead; no one watches commercials any more.”

“VCRs have gone away; everyone is ‘taping’ things on their DVRs.”

“People are watching less TV; they’re all on the web or downloading shows on their iPods.”

MCE study objectives

Dimension current consumption of media—focusing on television and video—and how it is

changing over time

Propose optimal forms of video media measurement

in order to

MCE RFP process Circulated a request for proposal document to the

industry

Reviewed over a dozen proposals

Met with three finalists

Selected research proposal submitted by Ball State University and Sequent Partners

Received approval from Full Council in Fall ’06 to move ahead with a pilot study of 50 participants

Currently negotiating larger-scale, national version

Measuring television is becoming a multi-platform, multi-place challenge

video

Responding to the challenge:

Observational studies to map media exposure

Mapping media exposure Reach (incidence of exposure) for each

medium Time/duration of exposure to each medium Media exposure across all locations Concurrent media exposure for all media Multi-tasking with non-media life activities By demographics, ethnicity, socio-

economic class, etc.

Observational studies

Computer-assisted observation

Alphasmarttm Dana smart

keyboards equipped with

a custom “Media Collector”

program

Observers categorize media exposure, life activities, location and attention, via touch screen & keyboard

Data are logged every 5-15 seconds, depending upon study objectives

Category systems

Category systems: media

CRE Pilot – Latino, 327:30 a.m.

11:30 p.m.

TV

TV

Any media Any screen media

Radio

Radio

Radio Web

Email

Web

Email

Software

DVD

Cell

Cell

Cell

Teen media exposure (HSMToo)

15 teens averaged 7 hours, 5 minutes of media exposure

7:00am 12:00am

Television

Radio Any ComputerAny Online

WebE-Mail

Inst.Msg

Software

All Phone

Telephone

Mobile Phone Music

Print Media

Newspaper

Magazine BookAll Video

VCR

DVDGame Console

Other

0%

20%

40%

60%

80%

100%

0 60 120 180 240 300Daily Duration: Average Minutes per User

Daily Reach:Incidence of Use Among 350 Adults

Daily Reach and Duration for Various Media

Among 350 Adults in the Middletown Media Studies II

Media daily reach and duration

CRE MCE Pilot Study

Understanding Audience Consumption of Media

to Further Its Measurement

Pilot study acknowledgments

We would like to acknowledge and thank: The members of the Subcommittee Howard Shimmel, Calvin Chan and the

team at The Nielsen Company

Pilot recruitment process Target: 50 participants in Indianapolis

metro area Ball State recruited 10 high-tech and 5

Spanish-dominant participants Purposive sample

Nielsen recruited 35 participants Former Nielsen cooperators Alert letters mailed (w/ brief study description) Recruiting calls ‘Agrees’ were turned over to Ball State for

scheduling Incentives:

$150 for observation $50 for returning self-administered survey

Pilot measurement process Trained observers are gender- & language-

matched to participants Two observers per observed day Observations February – April, 2007 in

Indianapolis 17 media 15 life activities 6 locations 10-second increments

Pilot measurement process Reminder call before observation day Each observation day begins by capturing

any pre-observation media usage Observation goal: 14 hours average

observed day Self-administered satisfaction, uses and

gratifications surveys are left behind Next day callback captures any post-

observation media usage

Success criteria, in detail Sample representation; Target groups of interest

represented proportionate to the local universe: Spanish-language dominant (target: 4;

achieved: 5), African Americans (target: 4; achieved: 5), Inner-city subjects (target: 4; achieved 5), Advanced video technology users (target: 8-

10; achieved: 10). People Meter Forced Turn Over (FTO) HH

individuals (target: 3-5; achieved: 4). Approximately equal samples by day of week

(target: 5-9 per day; achieved: 7.1).

Distribution of participants in Indianapolis metro area

Neighborhood incomes: light=lowest, dark=highest

Success criterion: advanced video technology users

Target: 8-10 … Achieved: 10

HDTVDigita

l DVR PVP BBWirele

ss DVDSat.

Radio

1

2

3

4

5

6

7

8

9

10 HDTV: high definition Television BB: Broadband Digital: digital cable or satellite

DVD: DVD player DVR: digital video recorder, including TivoPVP: portable video, video iPod, video cell phone

Success criteria, in detail Data capture Robust average observational day

Target 14 hours; achieved 13.5 - 16 hours (13.5 hrs observed, 16 hrs with reconstructed data)

Virtually all observed subjects complete the observation day

Target 45; achieved 50 Demonstrate collection of useful information on

the unobserved earliest and latest parts of the day

Target: data obtained and usage reconstructed for 40 respondents; achieved 50 responses, with additional media use collected for 48

Success criteria, in detail

Video exposure captured across multiple platforms and multiple locations.

Viewing on computers or portable video devices

Measurement of video media exposure in situations currently unmeasured

Success criterion:Multiple video platforms

4% of video viewing was on computers or portable video devices

No use of video on mobile phones was observed

Note:“Video playback” = DVD, VCR, DVR“Web video” = partial-screen video player embedded in a web page“Video on computer” = DVD and stored or streaming full-screen videoMethodological pilot, illustrative data only – not conclusive or projectable and not to

be used for analysis

Success criterion:Multiple video locations

11% of video viewing was out of the home

The largest block of out-of-home viewing was “other” (friends’ homes, restaurants & bars, etc.)

Methodological pilot, illustrative data only – not conclusive or projectable and not to be used for analysis

Measured TV/Video,

87%

Potentially unmeasured TV/Video,

13%

The reason for doing the study:Potentially unmeasured video (pilot data) 13% of video viewing was in currently unmeasured forms or

settings Measured

At home: TV, DVD, VCR, DVR Potentially unmeasured

Out of home: TV, DVD, VCR, DVR; Portable video; Other computer video:

DVD, stored, streaming.

Methodological pilot, illustrative data only – not conclusive or projectable and not to be used for analysis

Success criteria, in detail Adult in-home television usage to compare

logically to NSI Indianapolis DMA data for dayparts well covered by observation, excluding early morning and late night (significant correlation by time period).

Data obtained from companion instruments for measurement of relevant video media attitudes, perceptions and uses and gratifications provide insights that would be useful, if drawn from a larger sample.

Comparison to NSI

A18+ PUT MCE Pilot Compared to Indianapolis NSI

010203040506070

7:0

0AM

8:0

0AM

9:0

0AM

10:0

0AM

11:0

0AM

12:0

0N

1:0

0PM

2:0

0PM

3:0

0PM

4:0

0PM

5:0

0PM

6:0

0PM

7:0

0PM

8:0

0PM

9:0

0PM

10:0

0PM

11:0

0PM

A18

+ P

UT

NSI MAY NSI FEB MCE Jan-April

R2 = .94 for MayR2 = .97 for February

What does the pilot “suggest” about TV and video exposure?

“Suggest” because the pilot is too small and too local for powerful conclusions, and it purposely includes extra “high-tech users” in the sample

How this would play out nationally and over time would be answered in a full national study

Suggestions from the pilot Video-capable portable devices are getting

a lot of use, but not much of that use is video viewing

Live TV is still a dominant presence, but video playback showed significant daily reach (62%) and average duration per user (1 hr 40 min)

Video playback on computers is still small (<10% daily reach); average duration per user was greater for desktops than laptops

Suggestions from the pilot The great majority of video viewing was at

home; “other” locations (other’s home, retail settings, etc.) accounted for 2/3 of the rest, with work/car/school providing the remainder

Mobile phones had extensive reach (68%) but were used for talk and texting; no mobile video exposure was observed

The next step: A full study

Not yet for public release

CRE MCE full study Five regionally dispersed DMAs Nielsen-assisted recruitment Targeting former cooperators (FTO,

LPM conversions, etc) 350 individuals (based on power

analysis) Each observed twice In spring and fall Plus 100-person media acceleration

panel

CRE MCE study DMAs

Core study: Atlanta Chicago Dallas Philadelphia SeattleAcceleration module: Indianapolis

Media Acceleration Process™ Designed to avoid “early adopter” trap Targets devices/services in market but

in early adoption phase Provides steep discount (50%) to

participants on catalog of selected items to accelerate adoption to “early majority” phase when shelf prices are falling

Participants are observed before and after acceleration

CRE MCE study timeline 2008

Feb March April May June July Aug Sept Oct Nov

Recruitment

Obs. (2 DMAs)

Obs. (3 DMAs)

Accel recruitment

Obs. (accel DMA)

Acceleration

Then we will have a framework for learning…

How should weunderstandconcurrentviewing?

How will this affect media planning?

How are mobile devicesbeing used for video?

How will playbackimpact liveviewing?

Where do we next followthe video?

How can wemap a strategy for

multi-platformviewing measurement?

Thank you!

www.researchexcellence.com