Philip M. Napoli Professor and Area Chair, Communication & Media Management

Co-Director, Center for Communications Graduate School of Business

Fordham University

This research was conducted in partnership with, and with the assistance of, The Collaborative Alliance

Plan

� Background: Audience Evolution and the Post-Exposure Audience Marketplace

� Research Questions and Method � Overview/Assessment of Social TV

Analytics Methodologies � Comparative Analysis �  Implications

Audience Evolution

New Audience Information Systems Evolved Audience Transformation of

Media Consumption

Stakeholder Resistance and Negotiation

The Post-Exposure Audience Marketplace

exposure

appreciation

response recall

interest

Research Questions

� What are the key methodological issues that arise around social media-based constructions of television audiences?

� How do competing representations of social media-constructed television audiences compare to each other and to traditional ratings?

� How might the institutionalization of social TV metrics affect programming?

Methodology

�  Textual analysis �  Interviews (21) � Participant observation

�  Industry events/meetings �  Social TV analytics training sessions (3)

�  Limited analysis of social TV and traditional ratings data

What Do Social TV Analytics Do?

�  Program Performance Assessment �  Quantity of online comments �  Share of online comments �  Sentiment �  Involvement

�  Content Performance Assessment �  Plot/characters

�  Advertisement Performance Assessment �  Affinity Tracking

�  Brand ßà Program �  Program ßà Program

�  Trend Analysis �  Audience Analysis

�  Demographics (limited) �  Influence/reach

How Do They Do It?

�  Data Gathered from Online Social Media Sources �  Twitter �  Facebook (public pages) �  TV Check-in platforms �  Other online communities �  Online news media

�  Analyzed via Language Processing Algorithms �  Synchronized with Program Schedule/Content

Data �  To Produce a Wide Range of Analytical

Outputs

Points of Differentiation

� Data sources � Algorithm/search terms � Measurement period � Granularity

Methodological Issues: The Redistribution of Cultural Influence

�  Layers of possible misrepresentation �  Access �  Participation ○  10% of Twitter accounts produce 90% of

tweets ○  Services generally don’t include Facebook

� Overrepresentation of young males � Overrepresentation of television enthusiasts

Methodological Issues: “Black Box” Audiences

�  “They never tell you what’s in the black box. There’s no transparency.”

�  “People who are qualified to come up with solutions are coming from a completely different direction from traditional media research people. There’s no overlap in terms of skill sets, pedigree, etc.”

Methodological Issues: Compatibility with Established Practices � Demographics � Dayparts �  Local markets

A Crowded Marketplace

% Overlap in Top 25 Programs

Bluefin Trendrr.tv General Sentiment

Nielsen HH Nielsen 12-34

Bluefin 40% 40% 32% 32%

Trendrr.tv 36% 28% 40%

General Sentiment

16% 32%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

% Overlap

Program Ranking

Bluefin-Trendrr-General Sentiment Comparison: Top 25 Programs

Bluefin-Trendrr GS-Bluefin

GS-Trendrr

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

% Overlap

Program Ranking

Bluefin-Nielsen Comparison: Top 25 Programs

Bluefin-Nielsen HH

Bluefin-Nielsen 12-34

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

% Overlap

Program Ranking

Trendrr-Nielsen Comparison: Top 25 Programs

Trendrr-Nielsen HH Trendrr-Nielsen 12-34

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

% Overlap

Program Ranking

General Sentiment-Nielsen Comparison: Top 25 Programs

GS-Nielsen HH

GS-Nielsen 12-34

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

% Overlap

Network Ranking

Bluefin-Trendrr-General Sentiment Comparison: Top 25 Networks

Bluefin-Trendrr

GS-Bluefin

GS-Trendrr

Implications: More or Less Diversity of Content?

�  + Greater diversity of success criteria �  Away from the “tyranny of 18-34” �  Beyond exposure ○  Conversation/appreciation

�  - Programming primarily to/for those

active on social media?

Thank You!

For more, visit:

http://audienceevolution.wordpress.com