The changes raise the question of whether channel placement has an effect on ratings.  Do stations gain an advantage from having a lower channel number, or are channel choices so diverse and complicated to begin with that channel placement doesn’t matter?

Radio stations have long understood the value of being in the “middle of the dial,” but it is less clear whether such an effect also exists in television.  Our initial findings are ambiguous–channel location seems to matter in certain instances but not in others.

Several networks have negotiated for more favorable places in the line-up.  For example, Discovery and Discovery Kids are moving from the programming guide’s figurative continental shelf of the 100’s into adjacent spots in the shallow waters of the low 20’s.

Other networks are making room for the next cohort of high-bidders.  CNN is vacating its spot for FX at the broadcast-like channel 10 and moving down to channel 78.

For the majority of Time Warner Cable subscribers who don’t follow cable trade press, the line-up shuffle will come as a surprise.  Those unprepared viewers looking for Adult Swim on the Cartoon Network and plugging the familiar two 2’s into their remote controls are likely to experience a disruption to their lean-back experience upon finding Discovery Kids on their screens.

If the accidental viewer find herself ensnared in Discovery Kids programming and takes in a commercial pod or two, the network may count this disruption as a happy accident.  And if the accidental viewer falls fast for the Discovery Networks’ brand and returns to its channels enough to be considered a loyal viewer, then the happy accident becomes part of the justification for working with the cable operator for a lower channel number.

The question is how many of these happy accidents can a network expect to gain with a lower channel number and to lose with a higher channel number.  Set top box data can help answer the question.

The figures below show how audiences with similar affinities for program genre tune in to the same program when the program appears in different places in the channel line-up.  Each figure identifies the Genre Segment analyzed, the network, the program, the date range, whether new or all episodes were considered and includes a table with the neighborhood head-end, the channel number, the average rating for the program at that channel number, and a box plot to illustrate the variability around the average rating.

For Bravo’s The Fashion Show, the effect of the lower channel number on ratings is unambiguous and positive.  The Reality Watchers, the subset of all viewers who have demonstrated a proclivity for unscripted television, tune in at different rates depending on where Bravo appears in the channel line-up.  In Malibu, where Bravo appears at channel 42, new episodes of The Fashion Show score an average rating of 4.6 among the Reality Watchers.  In Riverside, where Bravo appears at channel 135, the same episodes receive an average rating of 1.0 among the same audience.

For USA’s longer running and critically acclaimed Psych, the effect of channel line-up placement is harder to detect.  The Remote Detectives, the subset of viewers with a demonstrated preference for crime dramas, tune in to new episodes of Psych at almost the same rate in Monterrey Park where USA is located at channel 8 and in Riverside where USA is located at channel 60.

Finally, for CNN’s Larry King, lower channel numbers have a positive effect on ratings, except for in Malibu with both the highest channel number at 70 and ratings among the highest.  Fake New Followers, the subset of viewers with an affinity sarcastic news programming that carries over to the real news, tune in to Larry King at higher rates in Monterey Park where CNN is located at channel 10 than in Riverside and Longbeach where CNN is located at channels 48 and 61, respectively.

Channel surfing is essential to the television viewing experience, especially as the number of channels and the volume of programming on television explodes.  Following from the explosion of television content, viewers’ ability to discover new programs is impaired and their estimation of the opportunity costs of choosing a program is increased.

At Simulmedia, we think that certain viewers will spend less time surfing if they were better informed of program content and schedules.  We also think that, through segmenting viewers by their surfing habits, we might better identify the viewers that will surf less and tune in more after exposure to the right program promotions.  Scheduling promotions in spots that index highly for viewers likely to enter favorable surfing modes will encourage their substituting viewing time for surfing time and increase the ratings of the programs they choose.

To investigate surfing behavior further we tapped into data from Twitter, where dozens of people announce to the world that they are channel surfing every day.  Unlike traditional ratings or second-by-second set-top-box information, Twitter has rich qualitative information that reveals motivations behind viewers’ surfing.

A semantic analysis of the data revealed that the tweets could be clustered into 8 distinct channel surfing modes. Each of these modes represents a different state-of-mind, which in turn could correspond to receptivity to program promotion. The modes are listed below:

1. Anticipation - composed of people who tweeted about surfing as they waited for a program to start. For example, “channel surfing waiting until 3pm when ellen comes on!“.

2. Searching – composed of people who indicated that they were searching for something to watch. For example “Channel surfing . looking for some good cable tv“.

3. Multi-tasking – composed of people who were doing other things besides surfing and twittering. For example “Channel surfing and reading Roenick’s latest comments on how he thinks Babcock hates Cheli..“.

4. Split-attention - composed of people who distribute their attention between several programs at once. For example “Channel surfing between Idol and Lakers“.

5. About to sleep - composed of people who are just biding time by channel surfing before they fall asleep. For example “channel surfing, probably til I fall asleep.”.

6. Relaxing - composed of people who have indicated that they were in a relaxed state of mind and were channel surfing. For example “Plans fell through so chillaxing n channel surfing“. 

7. Bored & Awake - composed of insomniacs who are bored, and are surfing in the hope they fall asleep soon. For example “Channel surfing because I can’t sleep…..”

8. Converts - composed of people who found something to watch while surfing. For example, “Channel surfing and found last 15 mins of Penn State trouncing Stanford in the Championship game in women’s college#rugby on ESPNU“.

We’d expect viewers in different modes to respond differently to program promotion.  Of the eight surfing modes, Anticipation and Split-attention are likely the most difficult to sway with promotions.  Viewers in Anticipation mode surf to kill time before a scheduled program.  Viewers in the Split-attention mode are watching multiple channels at the same time.

On the other end of the discovery spectrum, we’d expect viewers in the Searching and Converting mode, roughly 30% of classified tweets, to be most likely to change their surfing time to watching time.  Directing promotions to viewers who are in Searching and Converting modes presents a untapped opportunity to increase ratings and also provide better media choices to these viewers.

Approaching loyalty through our personal television watching experience, we start with the programs to which we are loyal, the programs we eagerly await each week.  In light of our attitude toward those programs, we can’t help think that everybody that watches our programs is like us, that our programs’ audiences are as devoted as we are.

Extending to other programs, we think that all programs have a similarly loyal audience.  All these other programs that we’re not watching, they have their loyal audiences.  Other viewers are making dates to watch these programs and are eagerly awaiting their start times.

Dominant promotional strategy bolsters this perspective on television viewing.  Networks focus their marketing efforts on their loyal “core” audience – typically a demographic.  By concentrating their promotional arsenal in their own programming, networks endeavor to extend their core audience’s viewing to the next program.  If network’s loyal core audience is consistent and large, then the standard promotional strategy makes sense.

It turns out that we’re wrong.

Program loyalty is the exception, not the rule.  Only a minority of a program’s viewers is faithful and watches every installment.  Most of any program’s audience is just passing through.  They tune in to a program once or twice in the course of a season, and then go their merry way.  We touched on this subject in a previous post on primetime loyalty, and further analyses of individual programs have reinforced the pattern.

And if we hadn’t been looking at set top box-level viewing data, if we had analyzed per-program demographic composition in time series, we never would detected the sizeable churn in program episode viewing from week to week.  Looking at panel-based data, we would have seen the same or similarly demographically composed audience tuning in to each episode and failed to see that the larger part of a program’s ratings is attributable to transient viewers.

This makes more sense as we go beyond how we feel about our favorite programs and think through the implications of broader trends in television watching.  The fact is that we watch a lot of television.  The latest installment of Nielsen’s Three Screen Report has the average U.S. viewer watching a 151 hours of television in the home each month.  While we, personally, may watch less than the average, we recognize that we don’t dedicate all our viewing hours to programming to which we’re loyal.  We spend time surfing, sampling, and checking (and re-checking) programs we’ve heard about from friends and seen promoted.  In aggregate, our time on the hunt adds up and, from the perspective of ratings, accounts for a majority of attention to programming.

Our emotional approach to program loyalty now tempered by intellect and data, we can work though what this means for improving program promotion.  Immediately, we recognize that enticing one-time program viewers to commit another episode represents a significant ratings opportunity.  If we can find the one-time viewers that are receptive to watching again – whose attention is “for sale” – we can devise an effective promotion strategy.

At Simulmedia, we like to say “we have a segment for that”.

The fact that a program’s audience demonstrates observable differences in loyalty and in attentiveness (the volume of program minutes viewed) implies a program-specific segmentation.  We conceptualize the segmentation in the Loyalty-Attentiveness Matrix below.

a conceptual segmentation framework

This Loyalty-Attentiveness segmentation is unique for each program and independent of the Genre Segmentation we introduced last week.  Where we crafted the Genre Segmentation to explore notions of audience preference and their impact on what people watch, the Loyalty-Attentiveness Segmentation explores how people watch.

We find viewers tuning in to different numbers of a program season’s episodes.  In the matrix, the segments tuning in to more episodes find themselves in the right quadrants.  Collectively, these are the “Rooted” audience having demonstrated loyalty to the program.  The segments tuning in to fewer episode are placed in the left two quadrants.  These are “Rovers”, demonstrating less loyalty.

We also find viewers watching different volumes of program minutes.  In the matrix, the segments watching more minutes find themselves in the top two quadrants; segments watching less, in the lower two.

The Rovers with high attentiveness should be the focus of program promotion (the red box in the Matrix).  Though relatively less loyal, these are the viewers who have sampled the program with relative vigor.  Their higher degree of attentiveness to the program is a signal of their preferences.  Exposed to more program promotions, they may be more likely to return to the program for more episodes, helping the program’s ratings as they do so.

Segmentation is a useful technique in our work to understand people’s viewing choices.  We divide and subdivide the general viewing audience into segments based on data describing people’s attention to television programming.  Then, we analyze how our segments tune in to different networks’ offerings and their responsiveness to different promotions.   

The analysis sensitizes us to trends in tune-in and promotional responsiveness.  It inspires ideas on how we can improve our segmentation and the classification of attention data that underlies the segments. 

Ultimately, we hope segmentation can help explain changes in tune-in.  Given how different segments tune in to a certain program, we want to make recommendations on the best promotional messaging strategy.  Or, when a program grows more popular, we want to look at how different segments tuned in and answer the questions of how and why did ratings improve.

In the meantime, through our sandbox, we want to share our early exploration into people’s viewing choices and get feedback on our approaches. 

As a foundation for taking in our exploratory work, we present our initial Genre Segmentation.  Genre Segments are unions of audiences who tuned in to similarly classified programs.  We’ve assigned a name to each Genre Segment corresponding, sometimes humorously, to the type of programming they’re likely to enjoy and written a brief persona description to help cement the relationship. 

The size of each Genre Segment population is a function of the definition we’ve issued to determine whether a viewer belongs to that segment or not.  In the table below, the size of the segment is indicated by the percentage of the viewing universe that belongs to the segment.  Segments with more restrictive rules are smaller; less restrictive, larger. 

You’ll find Genre Segments in our analyses of Audience Attentiveness and our Network Audience Heat maps.