UsageWatch.org

For the past few months, I've been collecting evidence about adoption of home networking, and in particular installed base and attitudes. The evidence I found indicates that home networking is progressing extremely slowly, and doesn't seem to be the natural next step after connecting the home to the Internet with a broadband access.

Two analysts (Diffusion Group and In-Stat) estimate the number of networked homes worldwide to be in the 40M range (respectively 35M in 2004; 50M estimated in 2005). And make forecasts of 162M in 2010 and 200M in 2009, invoking:

growth fuelled by broadband service providers who are beginning to push combined modem networking solutions (RGWs)

In general however, home network penetration remains relatively low even in countries with high broadband penetration such as South Korea (2%), Germany (2%), UK (10%), Japan (16%), US (17%). The US number is corroborated by Forrester Getting home networking into second gear (5-01-05), which speaks of 19% of online households with home network.

Attitudes and motivations related to networking the home are analysed, for the US, in a Harris Interactive Survey (October 2004) of online households (n=699). In the sample, 20% of households have a home network, the main reason being to share a broadband connection, followed by a printer and files.  Those who do not own one consider that they have no need for one. Also price, poor knowledge of the gear, uncertainty about their ability to set the network up are among the reasons invoked more frequently.

A third piece of evidence to understand adoption of home networking comes from Forrester Trends Laptops and home networks transform behaviour (27-05-05), which finds that households with home networks and laptops have a significantly more intense Internet life (spend more time online (+66% than average household), and do more of all online activities), use the Internet from everywhere in the house, and watch less TV.

Forrester surveyed 12.155 NorthAmerican households in December 2004.

I was trying to put together some sort of unified view of Broadband Internet adoption in Europe, along the same lines of what I just did for the US, when I read Elizabeth Lloyd's enthusiastic article on Hong Kong's Digital Potential and turned my attention to Asia-Pacific instead.

There is a second reason to that. I'm having a lot of trouble with gathering statistics on the European Information Society. Contrary to the US, little of this information is available online. Few reports can be downloaded for free. Most of them have to be ordered, sometimes only by fax; have to be paid for in advance with no preview available. A good example is the process designed to order EITO's European Information Technology Observatory 2004. It is enraging!

Hong Kong
Hong Kong has a total population of 6.8 M people. Nielsen//Netratings indicates that they are 4.9 M to use the Internet, that is almost three out of four Hong Kong residents (72.5%). In the summer 2003, 66% of Hong Kong households had a Broadband connection.

Since 2000, when Internet users were 2.3 M, adoption rate has been very high to 113,7%. Elizabeth Lloyd invokes Hong Kong's early embracing of the Internet in 1991, high-living standards, relatively low access prices, good infrastructure, in terms of both PC penetration and broadband offer. ITU places Hong Kong world second behind the Republic of Korea in Broadband Internet; and world third for second-generation mobile penetration.   

Asia Pacific
Looking at the Asia-Pacific region more broadly, the ITU Asia-pacific Telecommunication Indicators 2004 report indicates a 38% increase in the number of Internet users from 2000 to 2003, for a total Internet population of 255 M. Broadband penetration is also very high, four of the top ten broadband-connected countries are in the region: the Republic of Korea (23.3%), Hong Kong (18%), Taiwan (13.4) and Japan (11.7%). Singapore is very close also with a penetration rate of 10.1%.
Among the factors invoked to explain strong adoption are:

• a favourable regulatory environment
• the emergence of regional equipment manufacturers
• urban demographics

Quoting the lead author of the report, Eric Nelson

The role of governments has also been critical in helping the rollout of broadband. Governments have taken steps such as becoming pre-eminent adopters of the technology themselves, stimulated the development of adequate national backbone networks, created incentives for the establishment of competition, interacted closely with the private sector and given subsidies and other incentives to extend coverage into rural areas to reach new groups.

Nevertheless, let's not forget that within the region very large disparities exist between the developed urban and the less developed areas.

In a previous post, I reviewed research on Broadband adoption in the US to arrive at an overall understanding of the Internet adoption situation as of end 2004.

Some explanations of this situation are offered in Michael Powell: Broadband suffers from Innovator's dilemma, an interview of FCC Chairman with Larry Lessig.

Three factors are invoked to explain comparatively weak US Broadband adoption (in 2000, N°1/2 in broadband penetration; in 2004, n° 13):

1. legacy, prior success in dial-up Internet access has made the switch to broadband more difficult
2. geography, the way the Amercan territory is configured and inhabited makes wiring everybody extremely hard
3. multi-provider, the US are experimenting with a multiplicity of Broadband delivery technologies: DSL, cable, wireless, WiMax, Satellite; and operators

Is Broadband adoption rate the main issue though? Both the Pew Internet Project and the Online Nation Report show that the number of households with no form of Internet connection is still pretty high and, among dial-up households, many are those that say they are just not interested in upgrading.  Availability of Broadband acces may serve those who are already part of the Information Society. But may not be the key to engage more people into the Information Society.

Whereas the large majority of US households have adopted the PC - they were 87.6% to have done so in October 2003 - they are about one in two to have an Internet connection (55%) and about one in five to have a Broadband one (19.9%). And if Broadband adoption is on the rise, overall Internet adoption seems to have reached a point of equilibrium. This means that a little less than one American household in two doesn't have an Internet connection, and that many American households are content with a Dial-up connection (60%).

From adoption data, the picture that emerges is one of a multilayered American Information Society. If on the one hand, the PC has confirmed its place in the American household, Internet has integrated only about half of them, and Broadband Internet a fifth of them. This article synthesizes research on Broadband Internet adoption: adoption rates, demographics, geography, motivations and evolving behaviours.

Sources
The latest report A Nation Online: Entering the Broadband age from the US Department of Commerce, Economics and Statistics Administration, National Telecommunications and Information Administration, dated September 2004 was made public recently. Based on surveys carried out about a year ago, in October 2003, as part of the US Census Bureau's Current Population Survey of 57.000 households, the report focuses on Internet adoption and use. Together with the Pew Internet Project Data Memo Broadband 2004, published April 2004, and a Nielsen//Netratings report on Broadband adoption published in September 2004, these reports lead to a good understanding of the extent, distribution and motivations behind Broadband adoption in the US.  Unfortunately, I haven't yet come across qualitative research to push this level of understanding further.

Internet Households
Looking at Households, first, PC adoption rate has stabilized at 87.6%, with 61.8% of these PC households connected to the Internet. In the past 25 months, Internet adoption rate has been feeble at 4%. By comparison, it had increased 9% between August 2000 and September 2001. The percentage of Internet Households has stabilised at around 55%, corresponding to a figure of 62 M households, a figure lower than the estimation of about 70 M I made in the US Broadband Adoption 1.0 post. The percentage of Broadband households is slightly lower, at 19.9%, equivalent to approximately 22 M households, compared to my estimation of between 25 to 30 M households.

Internet Users Population
Looking at individuals, US Internet users correspond to 58.7% of the total population, approx. 170 M people. They access the Internet from different locations:

• At home, using a Dial-up connection, 51.8% (corresponding to about 88 M)

• At home, using a Broadband connection, 33.9% (about 56 M, about halfway in the 48 - 65 range I estimated)

• Elsewhere, they have no Internet at home, 14.2% (about 26 M)

What about the 41.3% who do not use the Internet?
Whereas the majority of non-Internet users doesn't have Internet at home (75.3%), some (24.7%) have an Internet connection at home and people around them using it. In most cases, the connection is Dial-up (17.7%), but there is also a sizeable number of Broadband connections (7.0%). These people just seem not to be interested in Internet.

Demographics
Who's most likely to have Broadband at home?
It is more likely that the household be:

• Asian American & Pacific Islands (34.2%) or White (25.7%)

• with a family income of 75.000$ and above (45.4%) or between 50.000$ and 75.000$ (27.9%)

• with education beyond Bachelor's Degree (38%), Bachelor's Degree (34.9%) or some College (23.7%).

This demographics corresponds well to that described by the Pew Internet Project which found that more than half college educated people aged 35 and younger have a Broadband connection at home (52%); almost half of Households with income of 75.000$ or above (47%) and of College graduates (40%). Also people who have been online for 10 years or more are more likely to have a Broadband home connection (56%).

Geography
Where is Broadband adoption rate higher?
In terms of geographical distribution, Broadband adoption rates are higher in the West (23.8%) and Northeast (23.1%) than in the South (17.3%) and Midwest (17.8%) of the US. Nielsen//Netratings gives further support to this finding. Among the Top 10 Local Markets in terms of Internet users who connect via Broadband are:

• San Diego (N° 1, 69.6%)

• New York (N° 4, 66.8%)

• Seattle (N° 7, 63.0%)

• San Francisco (N° 8, 63.0%)

• Los Angeles (N°9 61.6%)

• Boston (N°10, 61.4%).

Incidentally, four of these cities are also among the ten cities with the highest Creativity Rank, according to Richard Florida's Creative Class analyses. 

Where is Broadband adoption rate lower?
People living in rural areas are less likely to have home Broadband. Broadband adoption is lower in rural areas (24.7%) than in urban areas (40.4%). The Pew Internet Project data are even more dramatic: only 10% of rural Americans access the Internet using high-speed connections, about one-third the rate for non-rural Americans. The Project observes that, apart from the availability of such an access, rural Americans have demographics characteristics that make it less likely for them to have Broadband connections even when available: they are older, less educated and less affluent.

Looking at awareness of available Broadband infrastructure, the "A Nation Online" report finds that respondents say that:

• They live in an area in which Broadband is available (77%)

• They do not live in an area where Broadband is available (8%), this was the case for a higher percentage of rural people (27%)

• They don't know (15%), this was the case for a higher percentage of rural people (35%).

Online behaviour differences between Broadband and Dial-up Internet users
Daily Internet use is more frequent among those who have a Broadband connection at home (66.1%) than among those with Dial-up service (51.1%). This finding is confirmed by the Pew Internet Project: Broadband users are more likely to go online on the typical day (69%) than Dial-up users (51%).

Broadband users tend to engage in more online activities:

• Purchase products or service, + 15% (49.2 among Dial-up users - 64.3 among Broadband users)

• Bank online, + 14.9% (23.8 among Dial-up users - 38.7 among Broadband users)

• Listen to radio, watches TV and Movies, +13.6% (17.3 among Dial-up users - 30.9 among Broadband users)

• Get news, weather, sport, +11.8% (64.4 among Dial-up users - 76.2 among Broadband users)

• Search for info on Health services and practices, +8.5% (33.4 among Dial-up users - 41.9 among Broadband users)

• Search for product or service, +7.6% (75.7 among Dial-up users - 83.3 among Broadband users)

• Playing games, +6% (among Dial-up users 37.1 - 43.1 among Broadband users)

On the other hand, Communication activities are very intense and vary little, + 4.1% (among Dial-up users 88.9 - 93.0 among Broadband users).

To assess behavioural differences, the Pew Internet Project uses a different criterion; the activities tried and carried out in a typical day, from a reference list of 18 activities, and also find a relative increase in intensity and frequency of Internet use:

• Dial-up users on average have tried 7 of these activities and on a typical day have engaged in 3 of them

• Broadband users on average have tried 9 of these activities and do 4 of them on a typical day.

Overall, more people engaged in these online activities in 2003 than in 2001:

• Bank online, + 10.4% (from 17.4 to 27.8)

• Purchase products or service, + 8% (from 44.1 to 52.1)

• Search info about Government services or agencies, + 5.6% (from 30.1 to 35.7)

• Search for info on Health services and practices, + 5.5% (from 34.1 to 41.6)

Motivations
Why do people subscribe a home Broadband connection?
The Pew Internet Project argues that people adopt Broadband if they value the time they spend online. They want to make the best of it, feel efficient and enjoy this time. It is then not surprising then that as they have a Broadband connection, they are more active online and spend more time.

The most frequent reasons to be given to explain the choice of Broadband are:

• Connection too slow or frustrating: 36%

• Download files faster: 21%

• Job-related tasks: 10%

• "Always on" connection: 7%

• Use phone and Internet at the same time: 7%

Why don't people upgrade from a Dial-up to a Broadband home connection?
A majority of Dial-up Internet users are content with this connection (60%), whereas 40% would like to upgrade. For the former, the main reasons are lack of interest and cost:

• They don't need nor are interested: 44.1%

• Too expensive: 38.9%

• Not available is cited by one Dial-up user in ten (9.8%) on average, but by more than one in five (22.1%) Dial-up users in rural areas.

Why do people choose not to connect at all to the Internet?
Lack of interest is the most frequent reason given to explain non-adoption of Internet, followed by cost and equipment issues.

• Don't need/not Interested (41.6%)

• Too Expensive (22.9)

• No or Inadequate Computer available (22.5)

The Digital Future Report 2004 also adresses this issue. The reasons given for not being online are:

• either no computer at all or inadequate computer, (40% on the increase with respect to 2002 and 2003)
• lack of interest (24%, stable)
• lack of Internet basic knowledge (18%, slight increase)
• concern about costs (10%, stable)
• too time consuming (8% on the increase)

When asked about the likelyhood they will go online the following year, the majority considers that they will not do so (61.2%). 

Why did people discontinue Internet service?
Cost and equipment issues are given as reasons for interrupting Internet service.

• No or inadequate computer available: 27.5%

• Too expensive: 27.2%

• Don't need/not interested: 18.4%

The Digital Future Report 2004 finds similar motivations in a slighlty different order: no computer is the most cited (and growing), no interest is the second (but descreasing) and concern with costs the third. Some respondents indicate also time consumption and insufficient use as further reasons for stopping to go online.

 

One of the first things to know when trying to understand the social construction of the networked homes is how many households have set up a connection to the Internet, and among them how many have chosen to have a potentially always-on high-speed connection. The presence of such a connection is a clear indication that the household is engaged in the Information society as an active node of network.

However, even what looks like a simple question, e.g. what is the number of Internet homes as of November 2004, is in fact quite difficult to answer. To simplify things, I looked first at the US situation only.

How many US homes are connected to the Internet? And with which type of connection?

The number I came up with is that nearly 70 M US households are connected to the Internet; and somewhere between 25 and 30 M have Broadband access.

Here is how I came to that estimation. I used two kinds of data:

• Survey data, the number of households with active Internet access, further decomposed in narrow or broadband connections

• Business reports, the number of ISP subscribers to broadband connections. Notice that these numbers refer both to residential and small business subscriptions, and to Cable and DSL connections.

According to the US Census, in 2003 there were 111.3 M households in the US. At the end of the same year, market analyst eMarketer put the total number of online household at 68.3 M (eMarketer, November 2004) corresponding to an adoption rate of circa 61%. Of the 68.3 M Internet households, 24.8 M (36%) had a broadband connection. Similar estimations come from JD Power and associates' Internet Service Provider Residential Customer Satisfaction Study (TM). The study indicates that household Internet penetration is stabilizing at 66%, corresponding to a yearly growth of 2%. Broadband connections account for 39% of all Internet connections, what represents a 13% increase in a year.

Looking at the number of subscribers to broadband connections, the FCC indicates that there were 26 M broadband lines in service at the end of December 2003 (FCC, June 2004). This number however includes homes and small businesses. More recent estimates, that again don't distinguish between households and businesses, put the number of broadband subscribers at about 30 M. Quoting LRG eMarketer reports that the twenty largest cable and DSL providers in the US, accounting for about 95% of the market, have 30.9 million broadband Internet subscribers (LRG, 10th November); while the DSL Forum indicates that the total number of subscribers is 29.186 M (DSL Forum, 30th June).

The same fluctuation characterizes the statistics about the number of home Internet users who have broadband connections. They vary between 48 M and 65M. The Digital Future report and Nielsen//NetRatings converge to put the number of broadband home users at between 64 M (Nielsen//NetRatings, August 2004) and 65 M (Digital Future Report, December 2003). The Pew Internet Project makes a lower estimation at 48 M broadband adult home users. (Pew Internet Project, March 2004). Assuming that on average there are two Internet users per household, the estimation fluctuates again in a similar range of between 24 and 32 M households.

Where: US
When: December 2003 - September 2004

It's been now more than twenty years that "ordinary" people have had Personal Computers at home, and about eight years that they have connected them to the Internet. A very significant portion of leisure time has been spent gaming on the PC; writing and distributing text; looking at, and editing, digital pictures and videos; listening to stored or streamed music and radio; exchanging music and videos; planning travels; learning; practising hobbies. Also a large part of the social and civic life has gradually relied on the connected PC. People keep in touch with family, friends and community by e-mail, IM or web blogs and sites. Political participation, relationship with the administration, financial and insurance management have also been moved to the Internet. Over the years, more digital technologies have been connected to the home PC: printers, scanners, more PCs, wired networks, digital cameras, digital music recorders and players, DVD. And the digital technologization of the home is still very much in progress: more recent entries are DVRs, wireless networks, VoIP and Media stations.

However, if we were to listen to industry press releases, technology analysts, journalists commentators, and innovators, we would be led to believe that most homes are networked or soon will (in the most developed countries) and that residents of these homes are passive consumers who use the technologies that are marketed to them in the way designers meant them to be used. In reality, the process of «technically» networking the home and of «socially» inhabiting it is largely more complex and highly intertwined. 

One the threads I'm follolling is the analysis of innovation in the domestic sphere. My objective is to qualify what "inhabiting networked homes" means as of November 2004. I approach the issue from an adoption perspective, trying to answer the questions "How many homes are networked homes?" and "How networked are they?" proposing a taxonomy of networked homes. Whenever research exists, I complement the adoption analysis with a usage perspective that tries to characterise the way people actually use the technologies. The underlying assumption is that adoption and usage best reflect the meaning that these innovations have acquired for people.

What could an operational definition of what a "networked home" is be?
A networked home is a household who has adopted the Internet as one of the mediators of many of its everyday activities. In terms of equipment, its basic configuration consists of at least a PC and an Internet connection. More advanced configurations include a broadband Internet connection and a home network, potentially wireless, connecting two or more PCs. In terms of usage, the networked home mediates typical household activities, such as fostering social relations; research; planning and preparing trips or purchases; managing finances and administration; leisure time alone and together. The level of equipment and the type of usage patterns are highly correlated, with greater intensity of Internet use in household with broadband Internet access.

Further support for estimating overall DVR adoption rate at around 4% of US households comes from Nielsen via ITfacts. More interestingly, however, Nielsen, quoted by MediaPost notes that "in some markets - Las Vegas, San Diego, Los Angeles, San Francisco, Austin and other "Western "markets - it's pushing 10%". These markets incidentally correspond pretty much to high density Creative Class areas. Empowering technologies, like DVRs, may couple particularly well with the Creative Class lifestyle.

Where: US
When: November 2004 (public statement)

Sean Michel Kerner's TV 2.0: HDTV, Program Guide and DVR Outlook mentions results from a Jupiter Research survey of domestic TV and Music digital equipement that sets DVR penetration at 4% of US households. This compares to 55% DVD players and 22% Home theater systems penetration. Intentions to buy confirm this trend. There are more respondents who intend to purchase a DVD Player (10%) than a DVR (4%).

Where: US
When: September 2004 (published)

According to data published by InfoTrends Research Group/CAP Ventures, this year more than one out of two US Internet households (59%) have digital cameras, a growth of 14% over 2003. This represents 42 M of the 72 M US households that have an Internet connection. The survey also found that 26% of these households, equivalent to around 10 M, have a wireless home network. How all of that technology is brought together remains to be understood.

Where: US
When: 2004

Thanks to Sean Carlton Advertising is more than just TV commercials for pointing out two pieces of customer behaviour research that enrich our understanding of DVR usage. Research from MPG confirms reports that the very large majority of DVR users (90%) skipp commercials on recorded programs. More surprisingly, they just do the same when they watch live TV (84%). How do they do that? Lyra's DVR Love: a survey of DVR users found that DVR are used for a delayed viewing of the program - the delay being of about half an hour - and commercials are fast-forwarded.

Where: US
When: May 2004 (publication)

We know that the number of households with PCs plus Internet connection is very large. But what about houselholds who have gone one step forward in developing their digital environment and have connected PCs, peripherals, Internet access and other digital devices to create home networks? Wi-Fi Moves In estimates the number of home wireless networks at 8.7 million (around 8%) US households in 2004. Forrester's Consumer technographics Q4 2003 set a similar number of networked home at 8 million households, including both wirelined and wireless networks. Adoption rate was found to be lower in Europe with 6 million networked homes, corresponding to less than 4% of European households.

The demographic profile of the users of wireless home networks indicates that they are young (51% between 18 and 34 years of age ; 76% when we include the 35-44 age group), affluent (80% earn between 60K$ and above 100K$). Internet is generally accessed from the living room (79%); the bedrooms (40%) and the kitchen (22%).

Once the network is up, the majority (77%) expresses the wish to connect more devices, and in particular the portable ones, to update and share contents; stream music (29%) and TV (25%). They also express the intention to buy new digital home entertainment products in the coming 12 months.

Where: US, and partly Europe
When: July 2004