Hybrid data extends beyond Home and Work PCs, and as a result of these measurement enhancements and the additional sources measured, metrics including Unique Audience, which Nielsen uses to rank the top web brands, witnessed changes in data for July. Therefore July data can not be trended, but moving forward Total Internet Audience data can be trended with previous months’ Total Internet Audience data.

Hybrid measurement combines Nielsen’s online panel, a people-based representative sample employed to measure consumer’s Internet behavior using Home and Work computers, with tag-based data from websites to account for Internet use from any source. Thanks to this hybrid approach, Nielsen’s Total Internet Audience metric includes web browsing activity from all devices, including mobile devices, tablets, secondary PCs and access points outside of home and work locations.

Overall 213 million Americans were active on the Internet in July 2011 from all sources included in hybrid measurement. Internet access through home and work PCs continued to grow to 249 million individuals in the U.S. during July 2011.

Traditional TV
Overall TV viewership increased 22 minutes per month per person over last year, remaining the dominant source of video content for all demographics. In addition, Nielsen data shows that consumers are willing to pay for high-quality TV content, with broadcast-only homes less than a tenth of U.S. TV households.

Mobile Video
Though still accounting for just a handful of hours per month, mobile video viewing continues to see marked gains, increasing 41 percent over last year and more than 100 percent since 2009.

Timeshifted TV
Timeshifted TV continues to grow, both in the penetration of DVR devices in the home and the time spent.

Internet Video
Internet video streaming also saw increases in time spent; this behavior is the highest among a younger and diverse subset of the population.

More details are available to download in the complete Cross-Platform Report.

Emerging Traditional TV/In-home Internet Streaming Trend
Until the fall of 2010, Nielsen data consistently indicated that the heaviest media consumers are so across all platforms. A subset of consumers from television and Internet homes has now emerged that defies that notion, with the lightest traditional television users streaming significantly more Internet video, and the heaviest streamers under-indexing for traditional TV viewership.

This behavior is led by those ages 18-34.  The group of consumers exhibiting this behavior is significant but small. More than a third of the TV/Internet population is not streaming, whereas less than 1% are not watching TV.

For more in-depth information on overall viewing behavior—by ethnicity, gender and age—as well as emerging trends and device and delivery penetration, download Cross-Platform Report.

UPDATE – 12/18/09: This article on the A2/M2™ Three Screen Report and related documents have been updated to include data from the full time period of the quarterly report, resulting in increased numbers for in-home, and time-shifted TV viewing. The previous report did not include viewing from the first week of the new Television Season.

Nielsen’s latest A2/M2™ Three Screen Report – a quarterly analysis from our Anytime Anywhere Media Measurement™  initiative – shows considerable year over year growth in terms of time spent for DVR (up 22.5%) and online video (up 34.9%) in Q3 2009. Given the consistent spike in usage among the three screens of television, Internet and mobile, consumers are clearly adding video platforms to their schedule, rather than replacing them.

“Americans today have an insatiable appetite for not only content, but also choice,” says Nic Covey, director of cross-platform insights at Nielsen. “Across all age groups, we see consumers adding the Internet and mobile devices to their media diet — consuming media anytime and anywhere possible.”

What We Watch Weekly
For the first time this quarter, Nielsen reports how much time the average American spends in a typical week with TV, Internet and mobile devices. Without a doubt, consumers of all ages spend the majority of their video time (nearly 99%) in front of the television, while DVR and online video are becoming more widely used.

• In 3Q09, the average American watched more than 31 hours of TV per week, with 32 minutes spent in playback mode with their DVR.
• In addition, each week the average consumer spent a little more than 4 hours on the Internet and 22 minutes watching online video.
• The average consumer spent 3 minutes watching mobile video each week.

The TV and Internet figures in this report are calculated using Nielsen’s National TV and Internet panels, which are measured electronically and reported on a regular basis. The Mobile phone figures are collected by Nielsen via a quarterly survey and give a firsthand look at how early adopters self-report their usage of mobile video.

Other Highlights of the Report

• Almost 99% of video content watched in America is still done on traditional television
• DVR and Online Video continue to show solid growth – up 22.5% and 34.9% respectively in time spent from Third Quarter 2008
• Teens continue to watch mobile video the most, at just over 7 hours per month, though mobile video is not just a young medium, as mobile users Adults 45-54 report viewing nearly 3 hours of video on their mobile phones

• Download Nielsen’s A2/M2™ Three Screen Report

Nielsen’s sample represents 98% of the population in the 51 markets thanks to its address-based sampling (ABS), which uses randomly selected home addresses rather than telephone numbers. As a result, ABS targets cell-phone only (CPO) homes and landline homes with unlisted numbers. This CPO group, which has previously been left out of all other U.S. radio audience measurements, skews toward younger, tech-savvy consumers and accounts for 15% of households in the sample.

Other radio insights within the 51 markets measured by Nielsen include:

• The 12+ listener on average tunes to 2.5 stations each week. CPO listeners 12+ listen to 2.8 stations per week
• Persons 12+ have a total Average Quarter Hour (AQH) rating of 15.7 and listen for 21 hours and 57 minutes per week
• African-Americans and Hispanics in the 51 markets tune in more than the average population at 26.5 hours and 25 hours per week respectively. These growing demographics also have higher quarter hour ratings (18.9% and 18.2%) than the average population.
• Radio listeners age 18- 34, which account for 27% of the measured population, have a total AQH rating of 15.6%, which is in line with the entire market.

Download the full press release announcing new ratings and insights.

The television industry is about to sail into a vast ocean of granular insights on TV viewing.  The ocean we refer to is, of course, Set Top Box (STB) data.  Other industries have also entered their own uncharted waters, with access to multiple sources of more granular information – manufacturers have shipment data and store data, magazines have circulation figures and sales data, the Internet has log files.  In most of these cases, the industry players have found a way to navigate through huge, sometimes disparate, sets of data to make more informed business decisions. And where they have not, confusion has ensued.

We in the television business can learn from the experiences of other industries and add value and deeper insights to our currency measurements by incorporating complementary data, like STB data.   Let’s take a deeper dive into STB data to understand what insights STB can and cannot provide.

The Promise

STB data has significant potential.  The sheer size and granularity of STB data enables a much deeper dive into audience behavior than possible with audience samples used for currency measurement.  We can provide insights into networks that currently are not reported, due to a combination of smaller audience and sample size, enabling what is known as ‘Long Tail Reporting’.  We can start looking at smaller and smaller geographies, demographics and time periods; second-by-second viewing analysis anyone?   Another big benefit of STB data is, with access to purchase data or some other data set, the ability to measure ROI on advertising campaigns with much greater fidelity than currently possible.   Not only does the size allow for the deeper analysis, it also improves stability and reduces standard error in the analysis.   As TV gets more advanced and interactive, viewers will engage in new behaviors.  STB data will be an invaluable source of insights into these behaviors.

All of this is well documented and discussed at almost every conference where STB data is reviewed. There are whitepapers, power-point presentations and articles galore written about all this.

Nielsen is extremely excited about extending the insights it currently provides the industry.  We view STB data as an extremely valuable source of information that, once adjusted for gaps, can provide deeper insights on TV viewing, advertising delivery and effectiveness.   We bring to this endeavor our expertise in processing and integrating multiple data sets to create consistent and meaningful insights for our clients.  We already do this for our Consumer Packaged Goods clients, our Internet clients and our Mobile clients.

The Perils

Discussions around the usage of STB data are not merely academic discussions.   A currency decision based on pure STB data would have unwarranted financial implications for ad buyers and sellers.  The truth is that viewers of different viewing platforms (over-the-air, cable, satellite, telco) watch different shows.  Winners and losers would be inappropriately determined if ratings were based only on cable or satellite data.  Ratings can only be based on an inclusive and representative source of audience estimates that takes into account viewing across all viewing platforms within households.

Let’s take a look at some data.

There are significant swings in ratings for networks and popular shows when we compare viewership from the Nielsen National People Meter (NPM) sample to viewership from various cut-back NPM samples representing Wired Digital Cable-only homes or Satellite only homes.  Some examples from the ‘08-’09 season:

• Cable networks would do much better in Digital Cable-only homes – as we would expect – with some networks getting a lift of 20+% in audiences.
• The Fox broadcast network would do 4% better with Digital Cable-only homes, while the CBS broadcast network would lose 6% of its audience.
• American Idol would do 12% better with Digital Cable-only homes, but 7% better with Satellite homes.
• Ratings for Desperate Housewives would be 12% higher with Digital Cable-only homes, but 6.5% lower with Satellite only homes. That is a swing of 18.5% for a single show.
• The Mentalist gets slightly lower ratings with either Digital Cable-only or Satellite only homes.

If all these types of changes were aggregated, the financial impact would be profound, with hundreds of millions of dollars shifting hands.   For example, at the June 24 Advertising Research Foundation conference on audience measurement, Alan Wurtzel, NBC’s President of Research and Media Development, reported that NBC had asked multiple set top box providers to generate ratings for the final episode of Heroes.  According to Mr. Wurtzel, “We gave them the same request.  What we got back were different answers.”  The difference in ratings from the same data source was six percent, which translated into a variance of $400,000 in ad sales for a single episode.

To look at it in a broader perspective, we looked at ratings and ad spending across multiple platforms, using the Nielsen’s NPM and Monitor Plus services.  Our estimate of the inappropriate result would be as follows:

• If C3 ratings estimates were based upon viewing only from Digital Cable homes, it would have cost the broadcast networks approx. $340 million in ad revenue so far this season. If ratings estimates were based only on Satellite homes, the broadcast networks would have 4% lower ratings and would be $730 million poorer.
• Cable networks, as we would expect, would benefit from using Digital Cable homes only – to the tune of $2.5 billion in additional ad revenue. If Satellite homes only were the basis for rating, it would have generated $600 million in additional ad revenue for cable networks.

These differences are a direct result of different viewing patterns between STB homes and non-STB homes.  And that is all before one factors in the various gaps associated with STB data.

The facts and myths of Set Top Box data

Over the last few years, we have been analyzing STB data from multiple sources and would like to share some of our insights with you.

1. Larger samples are not always better than smaller samples: Usually when comparing a larger sample to a smaller sample, increased size provides a more stable estimate and a lower the standard error around the estimate.  HOWEVER, this is true if the smaller and larger data sets being compared are of comparable quality – in terms of the completeness in representing the population and the accuracy of the data.  A high quality smaller sample will provide more accurate information than a larger sample with systematic biases.

2. STB data is NOT Census Data: STB data is simply not available from all TV households.  11% of US homes have no cable, satellite or telco service.  They continue to have their entertainment and information needs adequately met by free, digital, over-the-air TV (obviously without set top boxes).  Another 19% have analog cable with no set top boxes. Altogether, non-STB households account for about a quarter of all TV viewing.  We also know that in homes with access to viewing through STB, full TV viewing from the home would not be captured in the reported STB data.  There is on average one TV set in such STB homes that currently do not have an attached set-top box, and therefore, such non-STB viewing in such a home would not be reported.  This non-STB viewing in STB homes account for nine percent of total TV viewing.

3. Homes that view television thru cable, satellite or telco are different from other homes:  While representing a very large data set and covering a major portion of US households, using STB data to deduce the overall viewing can be misleading because the viewing in these households is different from other HHs.

• STB homes have more TV sets – 2.8 sets per HH vs. 2.4 for the average home
• STB homes watch more TV – approximately nine hours a day vs. 8 ½ for the average home or approximately six hours for broadcast only homes.
• STB homes are larger HHs and make more money
• STB homes do more time-shifting

As mentioned above, even in homes that have cable, on average one set per HH does not have a set top box.   Viewing on these non-STB sets is different.  Kids networks (Disney, Adult Swim, Cartoon, Nick) have higher viewing levels on non-STB sets than on STB sets, suggesting that while the main TV set in the HH may have a digital set top box, the TV in the kids room may not.

Data Gaps

Now let’s get to the STB data gaps.  STB data represents tuning from the Set Top Box, not the TV.   STB boxes are frequently kept on, even when the TV set may be off (consider your own habits at home in this regard).  We are finding that about 10% of boxes never get turned off for over a month.  About 30% of boxes stay on for 24 hours on any given day.  This varies from system to system and from box to box – which creates another interesting challenge in terms of harmonizing and standardizing the STB.  All of this before we get to trying to figure out who is actually watching TV and being exposed to a program or a commercial message – STB data does not tell us who is watching.

Industry players are trying to address these issues by developing sophisticated models to account for these gaps.  The question then becomes – what information is informing the models to take into account the gaps?  How do we find out what the STB data may be missing?   How do we know what we don’t know?   And how do we validate the models to determine their accuracy?  In the world of finance, didn’t Wall Street hire a lot of smart mathematicians and statisticians to develop models for them?  How much did we lose due to the overconfidence and overreliance on those models that failed to incorporate the dynamics of the real world?

The Solution

There is a way to complete the data set and harness the full potential of STB data in order to learn more and make better programming and advertising decisions.  Nielsen’s NPM sample tracks TV viewing for approximately 50,000 people across all platforms – cable, non-cable, satellite, new sets, old sets, sets in living rooms and bedrooms, in basements and kitchens.  Nielsen’s NPM sample is the currency that informs the US TV industry and is the gold standard for TV audience measurement.  The NPM data set allows us to create the modeling between panel and STB data – enabling accurate persons’ level viewing and extending it to a much larger data set for granular analysis and stable estimates.  The model was developed and presented at the ARF conference in New York in 2007.

The key steps to creating audience estimates using STB data are:

1.       Set on/off. Determine gaps between tuning records and TV set on/off from Nielsen’s NPM sample to inform the STB data. This ensures that the overall tuning levels in the STB data are accurate.  The industry is referring to this as ‘cap and edit’ rules.

2.       Viewers. In a very simplistic example, if 40% of the viewers to a show are men 18-34 making between $100K-$150K from Nielsen’s NPM sample, then that probability, demo and income level can be assigned to tuning records coming from the STB data.

3.       Viewing for environments not covered by the STB data. This would include TV sets within cable homes that don’t send data back (which tend to show higher levels of viewing to kids’ networks) and over-the-air only homes (which constitute about 10% of the total homes and by definition, have much higher viewing to national broadcast and local programming).  Again, the NPM panel data would provide the basis for modeling these gaps from STB data.

As NBC’s Alan Wurtzel said, “We’re learning that there is less and less there, there.  This is really hard stuff.”   We agree and we can help with the solution.  Nielsen has the relevant assets to inform the STB discussion and extend the insights on television viewing.  We look forward to working with the industry to take advantage of the promise and avoid the perils of set top box data.

This year about 54% of homes in the U.S. had more than 3 or more television sets, 28% had 2 television sets and only 18% had 1 television set.

Other Key Stats

There are 114.5 million TV homes in the U.S. in 2009

• 38% of U.S. TV homes have digital cable.
• 88% have a DVD player, while VCR fell to 72%.
• 82% of homes have more than 1 television set.
• 11% of U.S. TV homes only have the capability to receive TV reception “over the air”. These homes have neither cable nor ADS.

Download Nielsen’s Television Audience Report.

“There is incredible demand for more accurate, reliable and high-quality Web audience measurement. Nielsen’s new panel is a huge leap forward in measurement to address that need,” said John Burbank, CEO of Nielsen’s online division. “Nielsen continues to strive to provide clients and the industry with the best, most insightful research possible. This enhanced platform enables us to do just that. One of the most exciting things about this new methodology is by including Nielsen’s TV/Internet convergence panel for online audience reporting, we are laying the foundation for better future cross-platform measurement.”

John Neitzel, Group Research Manager, for Microsoft Advertising noted: “We are excited to see Nielsen take this leap forward in the evolution of online measurement. In combination with increasing the number of sites, the new audience methodology will help ensure that a truly representative panel of the Internet is being measured, which can help improve online advertising effectiveness.”

For more on Nielsen’s expanded panel and methodology, read the complete media release.

Recently, there’s been a fair amount of buzz about supposedly new and revolutionary ways to measure audiences online. Just this week, a press release from a competitor promised that their new “hybrid” methodology (panel + census) will provide “a comprehensive accounting of the complete digital media universe.”

The Nielsen Company has long believed that using panel and  census data – the best of both worlds – is a great way to measure the Web, especially in the increasingly layered, three-screen world. In fact, we were the first to commercialize the hybrid approach in markets around the globe and in the U.S. in 2007 with our VideoCensus product.

To learn how VideoCensus provided a new measurement standard, click here.

Experience and integrity in research have never been more important.

New forms of audience measurement need to be created openly and transparently, with the industry. Nielsen’s hybrid systems have been built in conjunction with local industry bodies and the market itself. We worked tirelessly to educate the U.S. market in advance of our VideoCensus launch and it is the reason why we are pursuing Media Ratings Council (MRC) accreditation for our new system to be launched later this year.

Server data is complex and its analysis is complicated, to say the least. Any company that specializes in server data has a deep appreciation of this fact. Analyzing this data is a skill set that can’t be obtained overnight. Nielsen’s decade of experience with server measurement in the market is essential in making a hybrid methodology work.

In about a month, Nielsen will launch a new approach to panel measurement that will deliver the most reliable portrait of Web audiences to advertisers and publishers to date. This isn’t because of the sheer size of the panel, it’s because of the quality. The panel will provide an incredibly stable foundation for other exciting Nielsen audience measurement.

Like the foundation of a home, the panel component of a hybrid measurement system is vital, as is the ability to understand and reconcile results with server data. With a solid foundation, you can live in a home for years. Start with a shaky foundation, and the home isn’t worth the monthly mortgage payment.

Know your foundation. Build your business on responsible research.

Nielsen’s VideoCensus provides timely statistics and insights into how consumers use video online.  This includes the size and demographic composition (age, gender, race, etc) of the viewing audience for each website, as well as important measurements like the total number of streams viewed and the time spent watching by the average viewer. All this information helps content providers and websites more effectively sell their assets and audience, while providing tools for advertisers trying to decide where to place an online video campaign.  In assembling this information, we use a number of different measurement technologies, each with its own advantages.

Why Nielsen uses samples to collect its research

• The only way to really report audience demographics (the age, gender, race and other characteristics of the person viewing video) is to actually measure what people – not computers – watch.  Because it is not possible to track the viewing of every user on every computer in the U.S., the best way to tell how many people are watching online video is to select a representative cross-section of the entire Internet population, monitor their viewing, and project the results to the population as a whole.  Just as a doctor only draws a small sample of blood to measure red and white blood cell counts, so too does Nielsen use samples (or panels) to measure Internet use.  Nielsen uses the same principles to measure TV ratings and consumer buying patterns.
• The key to accurate sample measurement is to create a panel in which every member of the population has an equal chance of being selected.  This ensures that the panel proportionately represents men and women, teens and adults, high and low income individuals, employees of large and small companies, and so on.  It’s also essential to represent both heavy and light users in proportion to the entire population.  If, for example, only heavy users were represented in our samples, our estimates would be too high and if only light users were selected our estimates would be too low.

How Nielsen’s VideoCensus panels are assembled

• Nielsen recruits its panelists through a number of methods, including email, online advertising, telephone calls and postal mailings.. Other Internet measurement companies rely exclusively on online recruitment, which skews the sample towards heavy users and typically overstates Internet activity.  Nielsen’s combination of recruitment tactics captures a broader spectrum of demographics and consumer behavior.
• The VideoCensus panel is composed of two separate samples:

– Nielsen recruits a very large sample through e-mail and online advertising.  The sample includes hundreds of thousands of households and allows for very granular measurement.  To make sure the information from this sample reflects the population as a whole, we adjust the data collected from this sample with a second, more representative Calibration Sample.

– The Calibration Sample is created by randomly identifying and actively recruiting panelists through their street addresses and phone numbers.  We are in direct contact with these panelists and pay them for their participation.  These are the same methodologies used to recruit TV ratings panelists. The “randomness” of this sample means it includes both heavy and light Internet users across all demographics, and is therefore more representative of the entire Internet universe. The combination of these two panels gives us the depth of a large sample balanced by the industry’s only truly random calibration sample.

How Nielsen collects Internet data from panels

• After panelists agree to participate in our panel, they install a Nielsen software “meter” on their computer, which enables Nielsen to measure their online and computer usage. When a panel member views a video, the meter communicates information about that activity to Nielsen. The panelist’s demographic information, the stream URL, and other site information are processed and reported within the VideoCensus system. (Note: Nielsen does not sell or publish user information.)

How Nielsen collect actual ‘Census’ viewing behavior

• Nielsen also directly measures the number of times a particular video is played. To improve the accuracy of this measurement, video networks or broadcast sites can embed a code or “tag” in their video players.  This process causes all viewers of content to send an anonymous ‘ping’ to Nielsen.  This so-called ‘census’ measurement makes it easy for Nielsen to identify and report the actual content or program being consumed, and allows us to report an actual count of times that content is played back.

Why Nielsen combines panel and website data

• Nielsen’s panel data provides the audience demographics of viewers, which is the industry’s ‘currency’ for media planning.  Census measurement provides the actual count of the times content is consumed and is typically the method used by online publishers and advertisers for selling and buying impressions.
• Nielsen combines the datasets and provides reporting that includes unique viewers, total streams, demographic composition, and time spent viewing.

Why Nielsen’s data differs from the clients’ internal server-based counts

• The most frequently debated number in online measurement is unique audience.  Website analytics systems (what we call ‘internal’ data) count cookies or Internet browsers, but not people.  These systems overstate audience for a number of reasons.
• If you visit a website from home and from work, generally that website counts you as two visitors.  If you regularly delete your cookies, each time you visit that website you’ll be counted as a new visitor.  Panel data collected by Nielsen, by contrast, measures actual people and projects their activity to a carefully enumerated Internet universe.

Why Nielsen’s numbers are usually lower than those from other measurement companies

• Because our randomly selected Calibration Panel includes both heavy and light users, our data is a more accurate reflection of the entire online universe.  Other measurement companies create their panels from people who answer online solicitation and who tend to be heavy users.  As a result, other measurement companies sometimes report numbers that are even higher than the internal server counts from the websites they are measuring.