TV Viewing in Internet Households
A Report by:
Data Sources:
National People Meter data - May 1999
Nielsen//NetRatings - May 1999
Executive Summary
The rapid growth of home Internet access in the U.S. and the continued capability of television to reach mass audiences is naturally giving rise to efforts by key media players to better understand how these two powerful media affect one another and how they can be used in concert as communications and advertising vehicles. Television is a fixture in U.S. households with virtually all U.S. households having at least one TV set, and three out of every four having multiple sets. Cable/satellite TV is available to 75% of households. In the average household, the TV set is on more than 7 ½ hours a day, every day of the week, every week of the year with individuals viewing anywhere from 3 hours a day (teens 12-17) to almost five hours a day (Women 18+). TV's penetration and high viewing levels has led advertisers to use this media extensively to reach their audiences. Over $45B is spent annually on TV advertising.
The Internet has been growing rapidly since 1995 and is currently available in approximately 38% of U.S. households. These households are a desirable segment for advertisers, as they tend to be more educated and affluent than the average U.S. household. Over 100 Million people have access to the Internet from home. Approx. 60 million of these use the Internet at least once a month (active users). The active users get on the Internet every other day for approximately 30 minutes per session. Average monthly usage for active users is about seven and a half hours. The Internet attracted $2B in advertising in 1998.
As more and more Internet companies start to leverage the reach of television to draw audiences to their web sites, and as the Internet becomes an increasingly important part of TV and cable networks' plans and delivery capabilities, the need to understand the interplay between the two media grows for Internet players and for broadcast and cable entities. With convergence knocking at the door, it may soon be commonplace for people to move seamlessly from one medium to another on the same delivery platform - TVs or PCs. Understanding how people use these two electronic media for information, entertainment, shopping, travel planning and all the things the Internet offers will enable the industry to be better positioned to deliver the types of programming and functionality people want and need.
Research done by Nielsen Media Research suggests that:
-
Internet homes are lighter TV viewers but analyses of the same homes before they had Internet access revealed that they were lighter TV viewers to begin with. There is currently almost no indication that Internet access cannibalizes television usage; instead, it offers a targeted vehicle to supplement advertising reach among these lighter television viewers. We are committed to ongoing research to measure and understand the evolving relationship between television and Internet usage.
-
While Internet homes are lighter TV viewers, they seem to watch certain types of shows and outlets more than non-Internet homes.
This report provides a look at viewing habits in homes with Internet access and is based on data comparisons from the 5,000 household Nielsen National People Meter Panel. It also draws from other research done by Nielsen Media Research to better understand cross-media consumption and its implications for sellers and buyers of advertising on TV and the Internet.
Manish Bhatia
Vice President - Interactive Services
Nielsen Media Research
The TV Viewing Profile
As of March 1999, there were 99.4 million TV households in the United States. Of those, 74 million had more than one TV set, and 75 million households had access to cable and/or satellite TV.
One of the most basic measures of television viewing is the number of hours per day the television set is in use - Households Using Television - or HUT. Overall TV usage has remained fairly constant over the past five years, averaging about 7 hours of TV usage a day. Seasonal differences are evident with increased viewing in the winter months (7 ½ hours in Feb.) and lower figures during the summer (6 hrs. 50 minutes in July).
Prime time (Mon-Sat 8PM - 11PM and Sunday 7PM - 11PM) continues to attract the most viewing activity. It is the daypart in which all segments of the audience spend the most time viewing. For March 1999, approximately 97 million persons watched TV during prime time per day with Sunday programming being the most watched (106 million people), and Friday and Saturday programming the least watched (90 million people).
Other dayparts reveal some clear differences by age group. For example, women 55+ spend the most time watching daytime television. Children watch more Saturday morning television and account for the fewest hours of prime time viewing. Sunday is the most viewed night of the week but not among all demographic categories. More children watch television on Friday nights than on Sunday night. Conversely, fewer men and women watch television on Friday and Saturday nights. Generally speaking, there are more women in the audience than men.
People who subscribe to premium services (e.g. cable, satellite, etc.) spend more time watching television than any other category of viewers - over 15 hours a week more than homes without premium services. Similarly, large families tend to watch more TV, and persons in higher income homes watch less TV.
The growth of cable and increases in the number of channels continues to increase the viewing options in the average TV household. By 1998, the average TV home was able to receive 57 channels. While the number of channels available in the TV home continues to grow, the number of channels actually viewed has not grown beyond 13. (The term `viewed' is defined as 10 or more continuous minutes per channel.) Coincidentally, this number is comparable to the number of unique sites Internet surfers tend to visit in a month (12 - described in detail in the following section). It should be noted that this list of 13 channels is not the same for every TV viewer. Each viewer has his or her own preferences that get reflected in what they watch, and each person has his or her own preferences for Web sites they visit
The Internet Surfing Profile
Latest estimates done by the Nielsen//NetRatings Internet measurement service puts the number of Internet homes in the US for May 1999 at approximately 38 million. These Internet households represent approximately 105 million people who have access to the Internet. It has been repeatedly pointed out in various studies that these households are more educated and have higher incomes than non-Internet households do. Of the people who have Internet access at home, 63 million people - about 2/3 of them - are active users, logging on to the Internet from home at least once a month.
On average, these active users tend to log on every other day (16 sessions a month). They tend to visit about 12 unique sites and view over 300 pages of content per month. This 12-site number is very close to the 13 channels viewed in TV households and may suggest that irrespective of how many TV channels or sites there are out there, people tend to find favorites and stick with them. Per session, they tend to view about 20 pages. They spend about 7 ½ hours a month on the Internet from home with each session lasting a bit less than half an hour.
The power of portals is evident from the fact that despite thousands of sites available, 90% of the online audience visits at least one of the top 10 sites in a given month.
Over half the people log on to the Internet during the hours of 5PM and 10PM from home with activity during the 1AM-4AM dropping to less than 10% of users. Teenagers (12-17) are more likely to be on the Internet during the hours of 3PM-5PM. People 55+, incidentally, are also more likely to be surfing the Internet during the same hours. The 25-54 group tends to go online during the 8PM-10PM hours.
Researching cross-media consumption since 1997
Nielsen Media Research first looked at TV viewing trends in Internet homes in the Fall of 1997 when every household in the entire 5,000-household National People Meter Sample was classified as Internet enabled or not. We looked at viewing data in Internet households and compared them to viewing in non-Internet households. The data for the reporting month of Oct. 1997 suggested that Internet homes were lighter TV viewers. This was confirmed in two subsequent updates - one in Apr. 1998 and the other in Oct. 1998.
While research done by Nielsen Media Research confirms that Internet homes watch less TV and that Internet homes were lighter TV viewers even before they got Internet access, based upon current research, there is little evidence that suggests that the Internet is directly impacting television viewing. Homes that got Internet access were lighter viewers even before they got Internet access. We are committed to ongoing research to measure and understand the evolving relationship between television and Internet usage.
What begins to emerge from current research is a media consumption profile for people living in Internet households that is far more revealing than the relatively simple `light' or `heavy' viewership classifications. People in Internet households seem to consume media differently than persons living in non-Internet households. In some cases, they consumed less of other media; in other cases they consumed more of other media.
Broadcast TV viewing in Internet Households
The following analysis is a snapshot based upon May 1999 TV viewing data from the Nielsen Media Research National TV panel.
-
Overall, Internet HHs watch less TV than all TV HHs
On average, 29.7% of all US TV households had the TV set on at any given point in time during the day for May 1999. Only 26.9% of Internet households had the TV set on resulting in a 10% lower TV usage compared to all HHs.
-
Differences in TV viewing between people living in homes with Internet access and non-Internet homes are greatest during weekday daytime.
-
Viewing levels in prime-time were comparable for Internet homes and non-Internet homes.
Total Day |
All TV HHs |
Internet HHs |
Index (Base=100) |
HHs |
29.7 |
26.9 |
90 |
Men 18-49 |
13.7 |
11.9 |
87 |
Women 18-49 |
16.2 |
13.9 |
85 |
Mon-Fri 9AM-4PM |
|
|
|
HHs |
25 |
20.2 |
81 |
Men 18-49 |
7.6 |
5.6 |
74 |
Women 18-49 |
13 |
10.4 |
80 |
Prime Time |
|
|
|
HHs |
57.8 |
56.1 |
97 |
Men 18-49 |
32.3 |
31.5 |
98 |
Women 18-49 |
35.1 |
33.2 |
95 |
As is apparent from the above data, total day viewing in Internet homes is about 10-15% lower than non-Internet homes (a late 97/early 98 study pointed to 16% lower viewing levels in Internet homes). However in prime time, viewing levels are fairly comparable.
Looking at viewing to specific networks, the following trends emerge:
-
ABC and NBC over deliver in Primetime for male 18-49 and female 18-49 demos in Internet homes
Broadcast Network Viewing in Internet Households for Prime Time is as follows:
ABC |
All TV HHs |
Internet HHs |
Index (Base = 100) |
Men 18-49 |
3.7 |
4.0 |
109 |
Women 18-49 |
4.9 |
5.1 |
104 |
CBS |
|
|
|
Men 18-49 |
2.7 |
2.5 |
93 |
Women 18-49 |
4.3 |
4.0 |
93 |
FOX |
|
|
|
Men 18-49 |
4.2 |
4.4 |
103 |
Women 18-49 |
4.6 |
4.5 |
98 |
NBC |
|
|
|
Men 18-49 |
4.7 |
5.3 |
112 |
Women 18-49 |
6.4 |
7.1 |
111 |
PAX TV |
|
|
|
Men 18-49 |
0.2 |
0.1 |
58 |
Women 18-49 |
0.3 |
0.2 |
73 |
UPN |
|
|
|
Men 18-49 |
1.0 |
1.0 |
95 |
Women 18-49 |
1.1 |
0.9 |
81 |
The WB Network |
|
|
|
Men 18-49 |
1.1 |
1.0 |
91 |
Women 18-49 |
2.3 |
2.1 |
90 |
While networks like UPN and The WB Network under deliver in Internet homes, some of their shows do deliver high ratings in Internet homes. Felicity and Dawson's Creek, (The WB Network) StarTrek- Voyager and Dilbert (UPN), 60 Minutes and Chicago Hope (CBS) are examples of such shows.
While ABC and NBC's prime time schedules over deliver in Internet homes, audiences are selective about what they watch on these networks also. Despite strong prime time delivery, Boy Meets World (ABC) and World's Most Amazing Videos (NBC) under deliver in Internet homes.
Cable TV Viewing in Internet Homes
Cable shares some common themes with the Internet. Not too long ago, cable was the `new media.' It offered viewers niche programming tailored to their individual tastes. The early adapters of cable were affluent households that were more likely to have kids. Cable also promised the advertisers the ability to deliver targeted audiences.
However, just like broadcast viewing, people in Internet homes have a different cable consumption profile than that of people in non-internet homes. Certain cable networks do better in Internet homes compared to non-Internet homes. Here also, we look at prime time viewing as that is when most of the TV viewing occurs. Below are the top 10 cable networks that over indexed for the key demos in Primetime:
Demo/Network |
Men 18-49 (Rank) |
Women 18-49 (Rank) |
CNBC |
139 (1) |
156 (1) |
HGTV |
108 |
123 (8) |
History |
109 |
126 (5) |
Discovery |
115 (8) |
122 (9) |
E! Entertainment |
134 (4) |
125 (7) |
FX |
102 |
121 (10) |
CNN |
123 (6) |
134 (3) |
ESPN |
113 (10) |
109 |
Fox News |
124 (5) |
126 (5) |
Headline News |
135 (3) |
132 (4) |
MSNBC |
137 (2) |
140 (2) |
Sci-FI |
114 (9) |
91 |
VH1 |
118 (7) |
110 |
Syndicated TV Viewing in Internet homes
The following syndicated shows over deliver to the Males 18-49 and Female 18-49 demo in Internet homes in May 1999.
Females 18-49
|
Index |
TV.COM |
152 |
ACCESS HOLLYWOOD (AT) |
117 |
LIVE-REGIS & KATHIE LEE |
111 |
ROSIE O'DONNELL SHOW |
111 |
SAVE OUR STREETS |
110 |
G.MICHAEL SPORTS-MACHINE |
106 |
WHEEL OF FORTUNE-WKND |
106 |
ENTERTAINMENT TONIGHT |
105 |
ROSEANNE SHOW |
105 |
TRIBUNE PRIME NEWS |
105 |
MMN HOME TEAM BASEBALL |
104 |
SISKEL AND EBERT |
103 |
STAR TREK:DEEP SPACE NINE |
103 |
ENTERTAINERS |
102 |
YOUR NEW HOUSE |
102 |
FRIENDS-SYN (AT) |
101 |
LIVING BETTER W C. WIATT |
101 |
Males 18-49
|
Index |
STAR TREK:DEEP SPACE NINE |
137 |
TV.COM |
116 |
CLICK |
114 |
EARTH: FINAL CONFLICT |
111 |
ACCESS HOLLYWOOD (AT) |
109 |
QUICK WITZ |
108 |
AMERICA OUTDOORS |
107 |
ENTERTAINMENT TONIGHT |
106 |
HOWARD STERN RADIO SHOW |
106 |
S.O.F.:SPCL OPS FORCE |
106 |
HOLLYWOOD SQUARES |
105 |
FRIENDS-SYN (AT) |
101 |
X-FILES-SYN (AT) |
101 |
In Conclusion
The data outlined above are one slice of the research Nielsen Media Research has been doing to help our customers better understand TV viewing patterns in Internet homes. The analysis holds valuable lessons for all players in the TV and Internet arenas in the following ways:
-
Internet companies are increasingly relying on TV to pitch their products and services. This information can help them develop more effective media campaigns and deliver their messages on programs and dayparts that deliver the Internet enabled audience.
-
Broadcast and Cable Networks are increasingly wearing two hats by producing both TV and Internet content. As such, they can use this information to better `manage' their audiences in getting them to flow from one medium to the other.
-
As sellers of advertising time to Internet companies, broadcasters and cable networks can use this information to offer the Internet companies schedules that better deliver on their target audiences.
Understanding media consumption profiles of people living in Internet households can better position the TV and Internet industry to develop and deliver information, entertaining and advertising content in the post convergence world.
Next Steps
In addition to providing the television and Internet industries with quality research, Nielsen Media Research has been activity participating in various convergence trials around the country. We are committed to contributing to the body of research and helping our clients better understand how people flow from one media to another and back. Additional studies are planned for the second half of 1999 and into 2000.
Stay tuned!
TV Viewing In Internet Homes
Nielsen Media Research - May 1999
Page 7