Blog Reaction #1
Write a 200-250-word reaction (the equivalent to one page) to the readings and discuss where audience research is now and where you think it may be heading, especially in light of audience fragmentation.
Can audience researchers maintain the status quo with their techniques? Why or why not?
As we move forward, what type of information might be more valuable to media producers? Is it enough to just measure eyes and ears, or should audience researchers be measuring more?
If so, what else should they examine? Can this additional examination, if warranted, be monetized to help sell audiences to advertisers? How so? Which is more important? Exposure research or engagement research? Why? Which is more valuable financially?
Post your Blog Reaction as a comment to this post.
Also, I'll be sure to address some of the points in the blog posts in VoiceThreads.
Jack
Can audience researchers maintain the status quo with their techniques? Why or why not?
As we move forward, what type of information might be more valuable to media producers? Is it enough to just measure eyes and ears, or should audience researchers be measuring more?
If so, what else should they examine? Can this additional examination, if warranted, be monetized to help sell audiences to advertisers? How so? Which is more important? Exposure research or engagement research? Why? Which is more valuable financially?
Post your Blog Reaction as a comment to this post.
Also, I'll be sure to address some of the points in the blog posts in VoiceThreads.
Jack
Hey everybody, I didn’t initially see the blog reaction prompt before I started writing this, and as a result ended up being much more long-winded than I planned. My apologies – I’ll cut down in the future!
ReplyDelete-Scott
Prior to these readings, when I thought about audience ratings, I only thought of them from the quantitative standpoint, and always believed that kind of measurement was shortsighted and does not capture all of the important information. So for starters, I was interested to read more about the increasing measurement technologies and theories based on audience engagement, appreciation, and retention.
Napoli writes at various points in the article about the fragmentation of audiences and the various options they have to consume content in today’s World. (Options which are more prevalent and easier to use in 2014 than a few years back when article was written) He talks about how accurate and reliable ratings are now more difficult to attain, although he later discusses new technologies measuring engagement, such as interactive set-top boxes, ad clicks online, and discussion boards. The Digital Audience Ratings (DAR) that Variety is now publishing seems like it could solve this problem to an extent. I would wonder if it also may be an easy way for companies aggregating data to get around privacy concerns, as many social media sites that DAR is looking at data from – Facebook, Twitter, Google+, etc - require some level of access to personal information.
In terms of the various ratings methodologies presented, a number of questions came to mind which I’ll list out a bit here:
- With regard to set-top box measurement, how does it account for a person who powers off their TV, but leaves the set-top box on? (I do that) This would be similar to the logging on/off issue discussed in relation to the reliability of measurement technologies.
- Napoli discusses a key principle of media consumption, historically, based on audience availability. I wonder if with the ever-increasing access to content, if this maxim will change. For instance, many professional environments have television and “cyber-loafing” is an increasing epidemic that employers are trying reduce, but perhaps it’s a boon for content providers?
- Just because I’m still relatively naïve about ratings data, concerning Napoli’s noting of similarities across old and new media platforms, I would wonder if there’s an effective way to measure, for example, how many Top Chef viewers also view its online companion, Last Chance Kitchen? Or is it just assumed that only viewers of the show would view the online content? On this wavelength, I wondered in Variety’s insights about Big Hero 6 if they actually know that debuting the latest clip during the ABC special directly led to over a million views on Youtube, or if that’s an assumption. Guessing it’s an educated guess based on when it was watched.
Hi, Scott-- It makes sense to reply to each section of your reaction, so that's what I'll do. I agree that the measuring devices of today are superior to what we were using just a few years ago. One big issue that we have-- and that I'll suspect we'll always have-- is that data are messy. Nielsen, for instance, uses a combination of methods to collect its data-- random telephone surveys, "Nielsen Families" complete with set-top boxes and color-coded buttons to press, and qualitative 'media diaries.' They then take all this disparate info and combine it into one weekly report that they sell to companies. To answer your questions, the set-top boxes only keep track of you watching IF you're a Nielsen Media Family and you press down the button assigned to you. You wouldn't be measured if you left the set-top on, but the TV off. Now, is it possible that someone presses the button, forgets about it, and gets measured as watching TV shows? Yup. Happens every day. Nielsen, however, is interested in generalities and patterns, not necessarily specific viewing patterns. Even though media companies tend to treat the ratings as "hard numbers," Nielsen tells everyone in their reports that they come with a margin of error (usually +/- 3%). This means that in many cases we don't know the true #1 show because the ratings are so close. They're just estimates. I love your final question. I think, depending on one's age, many viewers sometimes watch TV on a television, and sometimes watch it on another type of screen (iPad, laptop, smartphone, etc.). Older people (50+) probably still watch most, if not all, of their TV on television sets. Right now, there's nothing that exists that ties together who watches Top Chef and its online companion. We collect the data for both, but we don't know what the overlap is. Most audience research is syndicated (everything is standard), but what you're suggested would require custom audience research. This is done by every media company, BUT the results are proprietary, so they're not released to the public. It's quite possible that the studio that produces Top Chef has commissioned such a report, but we just don't know. Finally, yes, it was an educated guess regarding Big Hero 6. They were hoping it would generate buzz, but it's difficult to predict.
DeleteGreat points Scott!
DeleteThe current capabilities of matching audience behavior across devices (set-top box/viewership to online behavior or even purchase attribution) vary based on the accessibility of household data. When we work with Comcast for example, we're able to match television consumption with online behavior and actions. We often employ technology that tracked and matched mobile devices on a household level (ie who within the household owns and uses which mobile device), and how are they consuming said media at home versus in the office, depending on which IP address theyre accessing, this data is often used in real-time to predict whether or not they're more likely to respond to an ad from Toyota or an ad from Verizon. But I'm sure e networks are al
Published too soon... As I was saying, I'm sure the networks are also using this data to make short and long-term decisions around audience optimization, engagement tactics and programming. I'm sure Erin would have great insights here!
DeleteHi Jack,
DeleteQuestion about the Nielsen set top box, I always thought that there is a sort of timer on the device. For instance, if I log in and decide to watch a Top Chef marathon all day, wouldn't I have to after an hour or whatever, re-login? It seems not having some control over my friend leaving her box on BET all day to support me would be really messy data collection.
Hi Angela,
DeleteMy family was a 'Nielsen family' back in the early 80s and we always left ridiculous shows running at all hours just to mess with the data collection. There wasn't any such thing as logging in and such. I can't attest to whether that is different today or not.
I look forward to meeting you guys in October!
Susan
In Webster’s Audience Research Overview, he writes a bit about the domination of syndicated research. I am curious as qualitative measures become more popular, if companies will continue to increase the amount of custom research conducted. Throughout the piece, Webster repeatedly mentions the quantitative research dominance due, certainly in part, because of its existing data and its larger sample sizes. Immediately, I began to think about some shows I’ve been a particular fan of, and their respective lack of success with quantitative ratings. The Atlantic published an article in 2011 (link) which focuses on the critical acclaim, and ratings failures of Friday Night Lights. The ratings were in fact so low that DirecTV essentially had to rescue it from being cancelled midway through the series, as noted in the article. HBO’s The Wire was in a similar category, heralded by critics, but not leading in ratings. I started to wonder if The Wire were created ten years later if it would’ve had more success – Breaking Bad which is regularly discussed in the same breath as The Wire saw its ratings nearly double its ratings from the Premiere’s in 2012 to 2013. (link) That number is still pretty small in the grand scheme of things, especially given its critical acclaim.
ReplyDeleteSo based on Philip Napoli’s discussion, it sounded to me like more qualitative measurements could lead to programming decisions that have more to do with quality of content than number of viewers, but I’m not sure that’s true. When I came across this article from Ad Week, titled “David Simon: HBO Would Pass on The Wire if I Pitched it Today,” I became a little concerned about the future of quality content. It also befuddles me to think about shows which all of my colleagues seem to talk about, and I see lots of online message boards about, but do not get high ratings. Perhaps my peers and my demographic is simply in the minority in terms of the content I watch?
I think you hit on a great fear across the board and that is the BEST programming tends to be something that breaks the mold. If we rely strictly on data and patterns of what has worked in the past, then we'll never get anything original. This might explain, for instance, why most of the critically acclaimed shows can be found on cable and not broadcast TV. In broadcast television, there's a race to be second. No one wants to be first, because it's too risky. Cable, out of necessity, had to take risks. It's paid off. Today, the top talent (directors, writers, actors) flock to cable television. Because of the inroads cable has made, the broadcast nets have changed course and are now trying-- thus far, unsuccessfully-- to follow their model of buying entire seasons and telling one story over several seasons rather than having self-contained episodes. We'll see how it goes. The good news is that there will always be a market for original TV (I hope) as long as we rely on revenue streams in addition to advertising. For instance, the broadcast nets must make sure their programming attracts a wide audience without offending advertisers. Cable nets like HBO and Showtime don't have to worry about advertisers because they get all of their revenue from subscription fees. Cable nets like AMC and FX have to worry about advertising, but not to the extent that the broadcast nets do because they get a significant portion of their revenue from subscription fees. Now if we ever reach the point where those fees go away and we rely exclusively on advertising, then, yeah, we're going to see a major decline in the quality of programming (in my opinion).
DeleteI'm really enjoying this conversation but the links to the articles that you referenced aren't showing up. Scott, would it be possible to post them as I'd like to read those articles. And to chime in - I echo both of your sentiments about the best storytelling being on cable (or digital subscription platforms) because those platforms celebrate complex programming and are able to offer creative control with less advertiser concerns. And to add to the list of brilliant but cancelled shows, let's not forget [my all time favorite program] Arrested Development which was cancelled by Fox in 2006 and then re-born on Netflix.
DeleteHey Erin - sorry, I did notice the links did appear, but was lazy in getting them up, here they are:
Deletehttp://www.theatlantic.com/entertainment/archive/2011/04/friday-night-lights-how-a-low-rated-drama-launched-the-next-a-list/237381/
http://www.rollingstone.com/movies/news/breaking-bad-premiere-shatters-former-ratings-record-20130812
http://www.adweek.com/news/television/wires-david-simon-hbo-would-have-killed-my-show-today-157252
Arrested Development reminds me about in the early 2000s when social media was still very new and nothing like it was today, Family Guy was cancelled, then restarted years later based on audience support and engagement.
Couple of questions: You talk about what might happen if subscription fees went away and cable became more dependent upon advertisers, is it plausible that broadcast tv may one day rely less on exposure ratings to determine what stays on tv versus what gets cancelled? In other words, if engagement is connected to critical acclaim, do you see a time when that will matter more than actual number of eyeballs (week to week)?
DeleteAngela - This is such a great question for discussion.
DeleteIt seems like the highest quality programing often comes from networks that are not beholden to advertisers at all - such as critically claimed shows that are on HBO and Netflix. I hope these don't go away soon since more creative freedom is afforded to programing without advertising constraints. I'm often surprised when I see a traditional network show winning awards these day since they work in a far more regulated environment.
To react more specifically to some of the questions posed in the prompt, it is my belief that researchers will need to continue to adjust and improve their techniques. I’ve discussed already some specific examples I see where ratings data seems very difficult to measure – as even if set-top box data is tracked, those boxes could be left on, or could be in public settings where measuring exactly what is watched, and by whom becomes even more difficult. I’m curious about DAR’s methodology with this as well – Variety is posting straight numbers, but does DAR also break up their ratings based on demographics? It would seem that any digital audience measurement tool would skew towards a younger audience. Certainly DAR has considerable value, but in my opinion, may also have some tweaks to be made.
ReplyDeleteI am very torn about the qualitative vs. quantitative discussion, as I’d like to say that qualitative measures, including engagement with and appreciation of content must be emphasized in the future, but I also know that as it currently stands, numbers drive business, and I don’t foresee that changing, at least not for networks that rely on advertising. Perhaps premium channels that do not show ads, and rely more on subscription fees, are in a position to account for qualitative measures, but as the business moves online, and more and more numbers become available about audience exposure, it becomes difficult for producers not to pay attention. As David Simon said in the article I’ve linked to, the numbers and data have just become more and more important to the point that they become a necessity for shows to continue airing.
Going forward I might like to see a ratings system created that could quantify all viewers against, or alongside, engaged viewers. While Nielsen and DAR ratings together may provide a picture that captures some of this, I think to effectively quantify engagement, especially in today’s World, it might require techniques which are incredibly intrusive. Perhaps a deal could be struck where users could be offered some sort of discount, maybe as low as $1/month off their cable/satellite bill to install an app on there phone/tablet/etc which could track all mentions or views of anything they’re doing that relates to media. This would be complicated and would likely require Nielsen, networks, and advertisers to help defray that cost, but the added information could be invaluable. Exposure research may seem more valuable because of the sheer numbers, but perhaps engaged viewers are also more likely to be affected by the content, or even advertisements they see. Engaged viewership would also eliminate consumption of content that is somewhat passive (watching TV while reading) or very passive (TV on in the other room).
Interestingly, Google already tried the method-- with minimal success-- of offering discounts to customers who would be willing to have their TVs, laptops and wifi devices tabbed to keep track of TV and web site ratings. Consumers balked presumably because they don't like the idea of one company keeping an eye on all of their screen media use. Google was serious about getting into the ratings game, both TV and online, and they sufficiently scared Nielsen that Nielsen approached them and joined forces. Today, they work together to provide ratings in the online world... Well done.
DeleteAs a member of the Indian television audience, I have always wondered as to how the television ratings and audience analysis games are played! The readings gave a good insight into the fact that there is a lot more to it than just about what the viewer likes! As Jack mentioned, audience research is audience research anywhere in the world. There are the quantitative and qualitative methods. There are surveys and focus groups and set top boxes and regulatory authorities. But I want to talk about how Indian audiences react to media, and how audience research could benefit from going deeper into the viewers’ minds.
ReplyDeleteOne of the huge changes that have come about in the last decade or so in the media scenario in India is that more and more people actually have access to technology and media. From the big metros to remote, rural parts of the country. This brought about an interesting issue. The kind of programming appreciated by the two groups is very different and involves very different societal and cultural situations. That has had a great impact on both the kind of programming and also audience ratings. Middle class family melodramas became huge for a while with top notch ratings. So it was rather fascinating to note that while the access to ‘modern’ technology and media was growing, what is believed to be ‘regressive’ television was very successful. Was it because the cultural background of who the viewers were was now different or inherently this was the kind of programs that viewers (especially women) preferred in India? Most of my friends would scoff at these shows when we got together and called them old fashioned; but every single one of them knew exactly what was happening in the show! One of the factors that I felt was responsible for this kind of programming was the fact that one huge media conglomerate was churning out very similar shows for various channels. So I wondered if it was the lack of real alternatives. My theory was somewhat substantiated when reality shows came into the picture. Suddenly they were the in thing! No one apparently watched Big Boss (Indian version of Big Brother) but everyone knew who got eliminated last week. Dance shows and other talent shows replaced fiction series. Audience fragmentation is working in interesting ways in India.
Another big change that has taken place in the last few years is the foray of extremely successful and popular movie stars into television. Often on the show Kaun Banega Crorepati (Who Wants to be a millionaire), contestants are more excited about meeting the host Amitabh Bachchan, who is a huge movie star, than even the winnings! But then when the same show was hosted for a season by Shahrukh Khan, arguably one of the biggest movie stars India has had, the ratings took a sharp dip. So do the viewers like celebrities on TV or do they not? Was the audience’s interest in the concept waning? Probably not because when Amitabh was back as host, the show really took off again!
While Bollywood is one of the largest movie industries in the world with big stars, of late popular TV stars are part of medium budget movies which go on to become big successes. Popular FM RJs are big stars in successful reality shows The same news on different language channels of the same network is viewed very differently. So audience response is probably not just about the content but also its context in their own cultural setting.
I think what the audience wants is ever changing and the audience research methods and media producers need to be able to adapt rather than stick to predetermined testing methods. I am new to audience research in popular media so I am not sure what the method is to do so but I would love to learn and find out.
Hi Kalyani,
DeleteYour post explains a bit about the content offerings on Indian TV but can you share anything about how the shows are evaluated? Is there a primary research company which tracks ratings for the industry similar to Nielsen in the US? Do you know what methods they use to obtain their data (surveys, diaries, set top boxes, etc) and how this relates to the advertising system?
Regarding your comment about "regressive television" being successful, I think that is what is happening in the US as well and it relates to Jack and Scott's earlier post about high-quality content being cancelled due to low ratings. For the life of me I can't understand how "The Big Bang Theory" is appealing to viewers (apologies if any of my fellow classmates are fans) but it's the #1 comedy on television :)
Erin
Thanks for the response Erin. I did want to get into audience metrics on my post and realized the post was getting rather long! The television broadcast industry has gone through a number of phases in the last few years and consequently audience measurement has undergone changes too. Up until the early 90s, we only had one television channel. The govt. run Doordarshan. During the days of the single channel Doordarshan monopoly, DART (Doordarshan Audience Research Team) was the only metric available. It used the notebook method of recordkeeping across 33 cities across India. DART continues to provide this information independent of the Private agencies
DeleteIn 1997, a joint industry body appointed TAM (Television Audience Measurement) as the official record keeper of audience metrics. It was backed by Nielsen. With the help of investment from non-resident Indians, aMAP (Audience Measurement Analytics Limited) was set up in 2004. It specialized in overnight ratings. BARC (Broadcast Audience Research Council) was set up in India in 2012 with the specific purpose of designing, commissioning, supervising and owning India's Television Audience Measurement System. BARC is promoted by three apex Joint Industry Bodies - Indian Broadcasting Foundation (IBF), Indian Society of Advertisers (ISA) and Advertising Agencies Association of India (AAAI), representing the three key stakeholders in Television Audience Measurement - Broadcasters, Advertisers and Advertising & Media Agencies respectively
http://www.barcindia.co.in/news.html#news49 talks about the direction that India is taking in terms of audience research and analysis. It is an interview with Partho Dasgupta, CEO of BARC.
Hey Kalyani,
DeleteI find the Kaun Banega Crorepati case to be interesting and perplexing - is it possible that the audiences have an infatuation with Amitabh Bachchan? Is there maybe a buzz about him that causes audiences to watch the show and want to meet him? I don't know for sure, but did ratings dip for The Price is Right when Drew Carey took over for Bob Barker?
The regressive television debate is an interesting one - my feeling is that media producers, as a way to generate more and more money from advertisers, will push and push and push the same kind of content while the exposure ratings are high, and will only stop when it clearly nosedives. In that sense, I don't have much confidence in broadcast networks to be innovative. This goes along, Jack, with your initial response to my comments. I think broadcast networks are also too consumed by the primetime time slots, and trying to beat each other out. Perhaps one could attempt to launch a new and innovative show with an ad campaign that encouraged viewers to set their DVRs and catch it later. The advertisers would hate it, but it could symbolize an embrace of the new ways people consume content. Maybe they could even scrap commercial breaks and try a model where the show is "brought to you by (insert product)"
As for the Big Bang Theory, nothing frustrated me more at the Emmys than seeing Jim Parsons beat out actors I find to be much funnier than him.
This comment has been removed by the author.
DeleteThis comment has been removed by the author.
DeleteKalyani --
DeleteInteresting analysis above and thank you for sharing on Indian TV, such a nice window into another marketplace. Wondering if you could share a bit more about the "Who wants to be a millionaire?" example you shared above. I looked at both Khan's and Bachchan's wiki pages, while Khan is certainly younger and Bachchan's career is longer; they did seem to have similar vibes and resumes.
How long was Khan granted to host the show before the switch back to Bachchan? Can you share a little more about the two actors? Do you believe that Bachchan is a better fit?
To tie this back to Scott's comment, when "The Price is Right" switched from Barker to Carey ratings did dip, there was a nearly 15% drop in 2008 when Carey took over http://insidetv.ew.com/2008/03/05/carey-versus-ba/
Since the show is such a staple of TV gameshow popular culture in this country with some time and tenacity Carey not only got ratings back but UP.... in 2014 the show celebrated its 8000 episode and continually holds the #1 spot for day time TV http://pittsburgh.cbslocal.com/2014/04/07/price-is-right-airs-8000th-episode/
The reason I ask how long a shot Khan got at hosting Millionaire is a wonder in relationship to whether or not his success could have happened over time?
Thanks all for a fun and engaging conversation!
-Sammy
Excellent job, Kalyani--
DeleteTelevision, as a mass medium, provides a type of status conferral to those who are ever present. This might help to explain the host difference. For example, in the U.S., talk show hosts, over years, have become much bigger stars than their 'guests'-- who are supposed to be the stars. This is in stark contrast to someone who might interview stars for a different medium (like magazines, for instance). In that case, the interviewer remains largely unknown to the masses. In TV-land, however, the talk show hosts are the big stars-- the reason to watch. If we apply it to game shows, programs such as Wheel of Fortune without Pat Sajak or Jeopardy without Alex Trebek are simply not the same-- even if you replaced them with movie stars. Until a viewing audience 'confers' the status to the next host (which doesn't happen immediately), one should expect the show to suffer. Even as Drew Carey has recovered The Price is Right's ratings, I assure you that if CBS announced Bob Barker's return to the show, the ratings would skyrocket-- at least initially-- because of the status he has from being on television for so long.
Kalyani-when you were living there, were you aware of DART or was there little reason to pay attention to ratings and who was leading the primetime time slots, because you didn't have choice? Also, Primetime is the most important daypart for programmers & advertisers here, is that the same in India? Or do media execs and advertisers focus on audience patterns at other times?
DeleteA few other thoughts, the celebrity as a host doesn't always work here either. Think Rosie on Own two years ago or Queen Latifa's show, Rikki Lake's second stab at it. Sometimes I feel like what people want to watch is cyclical and the tv talk format is less important to folks now. And the window people are given to succeed is unfair because network expectation is based on how talk shows rated in the past.
Scott, 24 was super innovative, no? But as it relates to audience, do you think the audience responds to innovation or interesting story lines? The studios probably think those live events (used on ER and Grey's Anatomy) or musicals are innovative, I think they are distracting stunts. I don't believe they actually rate higher than their taped episodes.
Kalyani -- it is interesting that big movie stars in India have moved over to television. I wonder how many big film stars are going to jump on the Netflix bandwagon here in the U.S.? If you're going to make a foray into TV, I'd want it to be a pretty sure-fire bet. Because their algorithms are so stellar, Netflix is a great predictor of which of their shows will be successful. I found that 70% of their shows will be renewed for a second season, where as regular network television has only a 35% chance of survival. If I were a big star, I'd want to hitch on to a Netflix pilot, like Matthew McConaughey, Woody Harrelson and Colin Farrell in HBO's "True Detective" versus, say, (sad to say) Robin Williams in last year's "The Crazy Ones" on CBS.
DeleteThis comment has been removed by the author.
DeleteErin - Completely agree with your thoughts on Big Bang Theory. There was an article a couple years ago in the NY Daily News that mentions the show really took off once people discovered the reruns through syndication. Jack pointed out, "Television, as a mass medium, provides a type of status conferral to those who are ever present." Perhaps the reason it's successful is directly related to the sheer amount of exposure the show gets and not a real measure of quality.
Deletehttp://www.nydailynews.com/entertainment/tv-movies/cbs-big-bang-theory-reaches-tv-plateau-article-1.1204629
Kalyani - Side note: I LOVE Shahrukh Khan and am continually impressed by his perfectly groomed eyebrows.
Christina I am so impressed! How do you know Shahrukh Khan? :)
DeleteThanks Sammy and Susan for the comments. Interesting questions and observations! I feel Indian audience is extremely emotionally invested in their stars :) When Amitabh Bachchan had a horrific accident a few years ago, millions of people flocked temples and prayed hard and held vigil outside his hospital. I wonder then if it is a case of the personality being bigger than the product.
DeleteAs for more stars moving to television, it could probably be because the reach of the medium has obviously multiplied and there is BIG money in TV, almost as much (if not more) as in films.
These readings were a helpful refresher on different types of research, ways of analyzing data, critiques of current methods and an exciting look ahead at new ways of tracking and representing tv consumption. I particularly enjoyed reading about audience measurement as it relates to Nielsen since they’re the dominant ratings source in my industry. Many years ago in college I learned about the in-home methods that were used to measure consumption by “Nielsen families” but was surprised to learn that these antiquated techniques are still in practice given the technical capabilities now available to us. I can’t imagine a written paper diary log being an accurate method or one that is representative of younger viewers (who would likely prefer to use technology). As Napoli notes “effective sampling often is integral to accurate and reliable ratings” and I struggle to see Nielsen’s system providing an effective sample given the fracturing of the tv landscape in both content offerings and platform choices.
ReplyDeleteListen First’s (DAR) system provides an interesting enhancement to the ratings equation. Audiences who are more engaged are obviously more likely to tune-in, act as an advocate for your show, watch another show on your network, buy merchandise, etc so there is value in augmenting our perceptions of ratings to include this data but, at this stage, it’s likely more useful for networks and media producers than it is appealing to advertisers. The financial model still rewards eyeballs for the time being. Each year as advertisers media budgets reallocate to account for new content and platform options, I predict traditional TV’s slice growing smaller and as a result, ratings will have less influence. For the last 3 years I’ve attended the Digital Content Newfronts which are essentially advertising upfronts for digital portals (Hulu, Yahoo, YouTube, Vice, Maker Studios, etc). It’s interesting to see these new content distributors emerge and offer advertising alternatives to TV spots.
For digital content producers, engagement contributes more to views (which are the primary measure for streaming sites and essentially ‘ratings’) in this space than on linear tv. On YouTube, the site’s algorithm actually awards priority search ranking to videos whose channel regularly uploads new content, whose videos have a high flow through rate via playlists and most importantly whose audiences are actively engaged via commenting, likes, or shares. Understandably these are very different ecosystems but I find it interesting to compare the two models.
Very true Erin.
DeleteWhile there are different ecosystems, don't forget we're still talking about the same audience... at least on a macro level.
I find it interesting/telling when some advertisers still place a lot of focus on traditional costs per Mille, and some place a much higher value on more current metrics such as "cost per completed view" (in online video) or "effective CPM", cost per action and so on. Typically I find the difference is most compelling between brand advertisers and DR advertisers (direct response), where a brand advertiser might be someone like Lexus who want to focus on getting their message to as many households within their target demo as possible, and a DR advertiser might be Netflix who needs to measure the success of an ad based on the number if new subscriptions that can be accurately attributed to it.
You prompted my thinking down a rabbit hole regarding the future of audience engagement and content providers....
DeleteYour comment regarding data usefulness to content producers, rather than advertisers, may be moot in a few more years as the lines between "content providers" and "advertisers" becomes increasingly blurred. Consolidation of media conglomerates is becoming merged with consolidation of industrial / product production, and as we move towards becoming an increasingly data-driven and ideas-driven economy I can't help but imagine a scenario in which the sole focus of any content production is also the explicit consumption by an audience of not only that product (show, movie, music, or other), but the content *brand* in it's entirety. If that is the case, Robert's idea that the audience remains largely the same at the macro level remains even more true. Content producers will begin to focus more on creating their ideal audience's behaviors, associations, and allegiances, and less on the inherent differences they present at first consumption, as the idea that viewers are already brand-aware at first content contact becomes increasingly true.
Well done, Erin. The nut that everyone is trying to crack is how to monetize 'engagement.' What's it worth? The media industry-- and as Tim mentioned in a voicethread comment, it parallels other industries, too-- seems to be in agreement that engagement is worth something, and that engagement + exposure is worth more than just exposure, but how much more? Advertisers want to know, too. Robert's assessment regarding DR advertisers and brand advertisers is spot on. If we speak in terms of the social psychology of how advertising works, there's a difference between the central route and the peripheral route. Television 'mostly' operates via the peripheral route-- as long as the brands get their message out and are well known, mission accomplished. DR advertisers, however, wants to see a correlation between their ad and the # of hits to the web site, orders, calls to the number, etc.
DeleteSarah hit on something where we may all be going in the future regarding programming and that is the complete loss of the line between the program and the advertisement. In other words, the program IS the commercial and the commercial IS the program. We're not talking simple product placement, of course, but, rather, subtle product integration where certain products are presented positively within a compelling story. Product integration already brings in billions of dollars a year in revenue for programs. What's the difference? When Mitch and Cam from 'Modern Family' are shown driving their Toyota Prius, it's product placement. When the storyline centers around the characters going shopping for the new Prius, it's product integration. From the same show, the character Jay drives an Audi (product placement), and in one episode the Dunphy family does everything it can to acquire an iPad for Phil (product integration). We shouldn't at all be surprised to see this trend continue and it's not too far-fetched to imagine a case where product integration can provide enough revenue to cover production costs, provide a profit to the producers, and deliver increased sales to advertisers.
You guys all make such great points but I'm a bit horrified by that blur and that many won't even see it coming. The program is the commercial and the commercial is the program -- yikes. I'm hoping that there is also another trend in the making. Because we can know so much about who is watching, I'd love to see programming created for very engaged niche audiences that can still monetarily make sense for users, advertisers and platforms.
DeleteSusan to the rescue! I like your framing better, that's for sure. Selfishly, I'd love to live in a world where because I was interested in science, physics, and astronomy, there could be more shows for 'niche audiences' like me - with, sigh, the advertising to support it - because I have a wide impact range, or because I'm a classic loyal customer once I have found a show or product I love.
DeleteI think we are currently only getting a piece of the pie where audience attention and ingestion is concerned. To directly answer our first prompt Jack NO, the status quo will no longer cut it as a means of audience research and measurement. I think Listen First’s ideology is getting closer to where we might want to for the future of decision making for advertisers and providers. We don’t have the whole picture where audience participation or consumption is concerned at this point.
ReplyDeleteIn the future I would be both excited and terrified to see a ‘full scope’ measurement of audience behavior. To stretch our minds for a moment I implore you to imagine an environment where we could truly measure your interaction with content. If we could have metadata from device usage and actual attention to content would be more able to better designate advertising dollars? For example, if we could understand what you were doing while you were watching (right now a forensic crime show is playing in the background as I type this, am I engaged in the show? Not really. Should that matter to the network providing the show? Do I still count as a viewer since it is on in the room I am currently in? Traitionally the answer is yes, am I valuable consumer, No.) would it matter that I am using a laptop in conjunction with watching the show?
In our reading Webster remarks, “…in-depth → interviews and →observation have been used by both theorists and practitioners to understand the social uses of television, how people adapt to the introduction of digital video recorders (DVRs), and concurrent media use throughout the day. ” What I am about to assert is in the same vein but an incredible push to the envelope.
What I am asserting is sort of a combination of ‘second screen experience’ as well as a better intelligence of consumer behavior. What if my TV could know I was on my laptop? Although seemingly a little intrusive it could be interesting to have a device detection intelligence from my TV to understand when I am looking away from the TV and when I am looking at it (like a Microsoft Kinect device) could we use this as a means of collecting more accurate engagement from our sample size audiences? If we could use this sort of ability to not only understand when we were looking at the content emitting from TV’s but also understand what the consumer is looking at if not looking at the TV could be incredibly powerful. We could understand why the consumer is looking away or disinterested; are they working on something totally separate such as homework? VS. Is the consumer actually a super fan researching interesting tidbits about the show they are watching while they are watching? I think there could be incredible value and an opportunity for monetization for both audience research firms but also for content providers to its advertisers if they were armed with richer information.
I’m pretty sure that if the esoteric science mechanism of how audience research actually works was clearer and more obvious to providers and advertisers there would be even more attention around measurement devices that are more accurate. If the failings of the current systems were clearer more would (and should) be having the conversation we are and asking the important questions such as “How important and accurate is the information rating systems are providing?” along with “What experience and implications do these numbers capture and what does it explain about our consumer’s behavior?” I think it time to create new benchmarks for the data we collect and make decisions on.
As I’ve asserted before it is difficult to say which is more important research engagement vs. exposure. I think we currently use these parameters because they are what we are familiar with. If we could blend the constructs slightly to capture a combination of both of these sort of researches ‘expo-engagement’ research by not only measuring the time we actually watch the content (exposure) but also the time we engage with the content we could get truly accurate profiles of ingestion.
Reminder, I think this is both SCARY and interesting! What do you think? :)
DeleteJust adding this note so i can turn 'notify me' on... this will send you email updates when others reply to your posts.
DeleteSammy,
DeleteI think you're right on in your assertion that determining actual interaction would be the best measurement, but I also have the somewhat cynical view that things will not change until the ad dollars want it to change. That being said, my assumption is that viewers that are interacting the most with their content are also most likely to be affected by the advertisements they see. I think it would make sense for some custom research to be conducted with one group of viewers that, perhaps, are watching a show with laptops in hand, and a discussion board already set up for them to interact with each other, and another group that watches the same show with laptops set up that do not have any discussion board (but do have access to everything else we typically do on our computers/tablets/phones while watching TV). If there's a correlation between active engagement, and product interest/purchasing, maybe media producers would start to turn their attention to content that is most engaging, as oppose to what's viewed the most.
Excellent questions, Sammy. The technology already exists to do this as we could rather easily, with consumer permission, install eye-tracking devices in homes (http://www.tobii.com/) that could register when one was paying attention to one screen vs. another.
DeleteI, too, would be horrified and excited about the prospect. Obviously, having this kind of insight into human behavior would aid immensely in understanding market trends for both viewing and purchasing, BUT, on a philosophical level, I'd probably be opposed to voluntarily providing this kind of information.
I'm under no false illusions regarding the amount of data that already exists on my own behavior (Wegman's grocery store cards, credit card data, 'points' cards, web tracking, cell phone data, Netflix, cable viewing, etc.) and one could argue that the battle over privacy of this type is over, I still believe individuals would opt out of any type of technology where they felt they were being tracked 24/7.
Hey Sammy, I use to work for a company back in 2000 called Spiderdance and we produced two-screen interactive television. It was figured at the time that 70% of people were fiddling with their laptops while watching network TV. Our platform allowed people to login, and then Spiderdance's technology would sync with the timecode of the show they were watching and then we could chart everything about how that person 'played' along. And because they were logged in, we knew from past behavior what they were interested in and could incorporate that into the programming. I'm really surprised that two-screen engagement never caught on more fully.
DeleteI think it’s clear that standard ratings systems in film and TV are in need of an update. We don’t follow the same sort of viewing habits, we don’t use the same devices and we don’t interact with our media in the same ways. In that respect, DAR is certainly a step in the right direction.
ReplyDeleteBut when I look at the DAR from the position of an advertiser, it’s not clear that this represents much of an improvement. The Simpsons marathon may have created a ton of buzz for FXX, but if the audience consisted of a small group of dedicated Simpsons fans like me, just looking to relive some of the best old moments (of which there were many!), it’s hard to see the added value to the advertisers that the elevated rankings would suggest—short of any ads related to the content itself (i.e. the ad for the Simpsons/Family Guy mashup that FXX ran throughout the marathon was certainly effective). The other example, as Scott raised before, is that of a negative engagement. If the masses take to social media to criticize a release, that would create minimal or negative brand association for advertisers—which is also not reflected in elevated DAR rankings.
So I think the DAR has room to evolve, and I suspect that it will have to start encompassing more qualitative information about the type of engagement, taken from the audience itself. The problem here is that while public feedback is easier than ever to obtain, reliable and nuanced public feedback—the kind which could help qualify the nature of engagement—is increasingly rare (due in part to the YELPification of public discourse, by which everything becomes either the ‘best/worst thing ever’).
I’d argue, therefore, that qualitative data might need to be gleaned automatically from the quantitative set. Our conversation turned to Google briefly and I wanted to expand on that because I think Google’s approach to analytics holds the key to understand where all of this might lead. The value Google was able to bring to websites was not just showing how many people came to visit the site, but where they came from, what they did when they were there, where they stayed, and what they did after they left. In effect, Google found unique ways to quantify both exposure and engagement, on multiple devices, and—most important—track that over time. That gives deeper insight into not just audience exposure or engagement, but audience behavior, which is the holy grail for marketers: being able to understand how viewers tend to react when presented with certain information.
Sammy raised a lot of good questions about what a system like this might look like when applied to TV content, how it might be tracked, and how that might be monetized. I’m interested in exploring some of these things further. But, as a final thought (and something Erin touched upon in her post), I’m struck by how ‘antiquated’ our tools for audience research are, given the technology we have at our disposal. We might consider some of the Google-inspired methods for audience research groundbreaking for TV, but they’ve been commonplace in Internet world for a decade or more.
Thanks Tim --
DeleteAs a fellow Simpsons lover, I too was incredibly happy to see a 'simpsanity' marathon on FXX. It is my opinion that something like the clout for a nostalgia driven marathon is very different that regulatory measurements. While it's critical to have the ability to calculate the rates for one-time instances (like a live concert or the superbowl) and be able to appropriately monetize the ad space, i think the opportunity to gather consistent trends and better understand behavior is even more important.
Also agree that tapping intelligence similar to google's clearly superior tracking might be the right path to realize my semi-sci-fi theorizing as well as get cleaner data. :)
AUdience research and tools
ReplyDeleteNote - I'm designating numerical sections to my blog reactions just to make it easier to keep track of any ensuing comments or conversations.
1) I very much enjoyed these readings. Mr Websters article provides a succinct view on some different facets of audience research, although it does seem rather one-dimensional. And his opinions seem limited to just one channel/platform (television). As a result, my reaction to this article in particular was somewhat pedestrian. But first off, I will ask that we take a much broader definition of what an audience is - not just an audience of a TV program, but of all media (web, social, mobile, online video, paid, live, On-demand, and so on). I think this makes it a bit more interesting (and realistic), and allows us to consider the full 360 degree audience profile rather than just the portion related to television ratings.
2) I also question his basic definition of quantitative and qualitative research because (as one example) he cites theatre-testing as a form of quantitative research. I've always thought that qualitative means that I can count the results and it and quantify them against my expectations or goals. Quantitative research, in my mind, enables statistical analysis that leads to insights, leanings and/or interpretations of what such data represents. Yet Webster discussed quantitative methods rather generally as surveys, such as asking theater-goers a series of qualifying questions. This strikes me as a little vague and possibly confusing quantitative with qualitative. I'm curious to know what everyone things about the fundamental difference between these methods of research. Especially you Jack, do you think I'm misunderstanding the meaning of qualitative versus quantitative?
In response to (1) check out my post above i am totally on the same page! I think we are only looking at a piece of the picture.
DeleteHi, Robert--
DeleteYou are correct that quantitative research relies heavily, though not exclusively, on inferential statistics to generalize from a sample to the overall audience in an effort to predict (or gain insight, leanings, interpretation). The other part of quantitative research simply provides us with descriptive information-- usually in the form of percentages, charts, graphs, standard deviations, etc. There is no 'statistical testing' like there is in inferential quantitative research.
There are only a few ways that quantitative researchers collect data (this applies to both the physical and social sciences) and they include surveys, experiments, and content analyses (which is just really a survey of pre-existing sources rather than people). For example, earlier Tim referenced a study about 3D printing published by Appinions, a quantitative research firm headquartered here in Ithaca that uses systematic content analysis of millions of documents to present ideas as to who is most influencing an industry.
Qualitative research, by comparison, tends to allow participants to respond in an open-ended fashion. Methods include interviews, focus groups, participant-observation, field studies, etc. Almost all of the data (save for the demographic makeup of the participants) is based on words, so it's not possible to summarize the findings using numbers. Instead, qualitative researchers look for patterns.
I agree completely that we need to think of the audience as a whole and not just as part of one distribution method/platform and I think most media companies have moved in that direction. No one seems to rely exclusively on Nielsen ratings anymore, but, for better or worse, those ratings still matter more than any other metric adopted thus far.
This comment has been removed by the author.
DeleteProfessor Powers: Are research firms like Appinions really prepared to deal with the coming surge of 'big data' to more effectively deliver quantaitive research? Have you seen firms using correlated data from smart phones' GPS, user's location at time of consumption, length of consumption, and interface with supporting ad materials (i.e. downloading a TV soundtrack on iTunes)?
DeleteDAR takes us one step further in this direction, but only aggregating digital/social platforms' information. Yet our entire digital lives are becoming a platform for social engagement, and the boundaries of 'digital' wearing thin. Beyond DAR.... I'm curious whether firms are thinking about or trying to catch-up with the wealth of data available to them in more complex research methods, or if the capacity to do so is perhaps an entirely new field.
3) Now that I have that off my chest, I'd like to talk about the actual opinions set forth in this week's readings. As I mentioned earlier, the majority of ideas are limited to television, but the nature of audience research today is far more multi-dimensional. Napoli's ultimate notion that technology is "gradually damaging the foundations" of traditional audience markets is strangely protectionist of him. I would challenge that most people in the audience business (at least nowadays) would say that technological developments have greatly advanced the way we measure, evaluate and understand audience behavior. This in turn allows us to think about audience research very differently... Where Webster and Napoli seem to be focused on the end of the research funnel, asking simply "What do the ratings say?", today we
ReplyDelete4) Engagement is mentioned somewhat flippantly by Webster and Napoli as if it were an outlying form of research that garners little attention. I have to say that I find engagement metrics to be far more necessary that mere eyeballs. Viewership today means something very different than it did a few decades ago (or even just a few years ago). So a simple approach of measuring exposure (reach and scale) is significantly less meaningful than measuring depth of engagement, appreciation and responses.
4b) Let's just take television as an example of where audience research began, and what it means today; in the 1950s, 60s, 70s and (to some degree) 80s, the typical household would consume television as a one multi-generational family unit, huddled around one television set in the living room, all watching the same programming and sharing the experience. Fast forward to today and the typical household/audience is vastly different, consuming content on multiple televisions, while also engaging with social media, online video (don't make me list them again)... and doing so simultaneously on multiple devices. As a result, the qualitative and quantitative data that can be collected, individualized and matched against an audience's past and current behavior allows us to make macro and micro decisions, in real time, and even predict future engagement patterns. As an advertiser, this is where it gets interesting - being able to quantify, qualify and predict how highly a particular audience segment will index to perform a given action.
5) As think about the conclusion of my blog reaction, I'm compelled to consider where my own understand/approach towards audience research, and I think it's more about encouraging ourselves to think less about the tools (of which there are plenty) and more about the questions that audience research can answer... Therefore it's critical to first think about the various audience stakeholders (content producers, content providers, platform owners/publishers, advertisers), and then the subset of questions that each stakeholder typically prioritizes. From here we can start to think about which phase of the audience life cycle to focus our research on, and what techniques/tools to employ.
While it may not be a direct response, Rob, your post got me thinking - with the advancement in modes of consumption, one thing I find for myself is a more negative reaction to advertisements than I've had in the past, and I would think that's one piece of the puzzle that still needs to be solved by networks and advertisers together. Over the last 2-3 years I've found that when I watch a show live, let's use Top Chef as an example, I am actively frustrated that I cannot fast forward through commercial breaks and almost subconsciously have a negative association with the commercials I'm watching. So I would agree that targeting, and interpreting behavior is extremely important, but I would also ask if the long standing model of commercial breaks is in need of a change.
DeleteRobert, this ("critical to first think about the various audience stakeholders (content producers, content providers, platform owners/publishers, advertisers), and then the subset of questions that each stakeholder typically prioritizes") is PRECISELY what we're going to be doing when we craft/create the instrument intended for CEOs/innovators in the media world.
DeleteThanks guys!
DeleteAnd yes, Scott, relevance is a whole other topic when we talk about targeted or un-targeted ads (especially television, but similarly in all other channels). With the increase in demand (pun intended) for on-demand content, the availability and scale of different ad formats is also increasing, making it easier to skip to content or even re-watch a relevant ad.
ReplyDeleteThese articles were really interesting. Full disclosure, I have a love, hate relationship with audience analysis. I have been on the losing hand as a producer and also as a viewer; suffering pain and anguish (hyperbole) when critically acclaimed programming is ripped from the lineup in favor of another reality show!
While I will admit that the status quo is not a suitable approach to the changing tides in content consumption, I am conflicted. I think a hybrid of what we have, plus DAR and something yet to be developed is most likely the best way to gauge what viewers are watching, how, why and who is doing the watching. I think the way the old (literally and figuratively) Nielsen box collects its data is useful; each member of the household "signing" in may have its margin for error but I still think it is a fair indication of what people are watching. However, how people are watching, whether on their DVR, OnDemand or other is such a big piece of the pie that a way to measure that information needs to be included in order to provide a more whole interpretation. I think i remember years ago, when TiVo was the start of this evolution in consumption, analysis firms didn't even count those views in their data collection.
As a broadcast producer, I am obviously most concerned with eyeballs. Unlike Erin, it would be a rare occasion that anything I do, would generate advance buzz on FB or be trending on Twitter. I might even resent that my shows success and its value to the network is inextricably linked to conversations on Facebook. I want to use Facebook and Twitter to generate buzz, to promote, but I don't want it to count when ad execs are deciding how to spend their budgets.
The DAR approach is a start, but I have questions: Who makes up this sample? Does it matter what they are saying on FB? I mentioned on VoiceThread, what if I mentio on FB how awful Extant is and a thread began, does that constitute valuable engagement? Is it, "any buzz is good buzz"? Also, in looking over the DAR reports, I noticed that shows or networks that are traditionally watched by African Americans were underrepresented or altogether MIA. However, I read once that African Americans use Twitter and social media at a high rate, how can media institutions or advertisers trust that it is really representative sampling? Ultimately, right now, it seems like media will need to supplement this data with another, which is the exact opposite intention.
As a producer, I want to know what the viewer watches, when and how. If they flip away, I want to know when and where to. The sample should be required to participate in a multi-platform analysis? Whoever we collect linear data from, we collect digital data from too.
Finally, I want to be careful to not disrupt the experience. So whatever the evolution is, tracking when I look away (like my android does) or whether I am multi-tasking seems intrusive.
I chuckled aloud at your full disclosure, Angela. I think we all feel like that. (As an aside, I have come to accept that if I find a show to be of high quality, it will invariably get canceled because it doesn't attract a large enough audience-- Arrested Development, for example). Fortunately, thanks to cable nets creating original programming, the ratings don't matter so much so that programs like 'Breaking Bad' can stay in production despite relatively low numbers.
DeleteYou ask excellent questions about the DAR methodology. Unfortunately, they've kept their methods private, but I assume (and that's all it is) that they've taken into account 'negative buzz' and have incorporated that into their formulas to devise their numbers. Negative buzz is valuable information to media producers, but the assumption is that scoring high on the DAR scale is positive.
I agree with your assessment that, as a producer, you want to know what viewers watch, when, and how. I would add that it would also be quite valuable to know 'why' they watch (but we're not really good at measuring/answering that question-- at least not yet).
There's no question that engagement matters, but the real question is 'how much does it matter financially?'
Right now, at this moment, we don't know.
And that's the bottom line.
Well done.
Angela you promoted my thinking on this matter -thanks for your comments! You can read where my thoughts led me, below, but I also wanted to comment on your very last thought regarding 'disrupting the experience.'
DeleteDo you think that as we move towards increasingly multimedia experiences that tracking engagement will become, at some point, impossible? Will the number of factors at work become to great to correlate? I think at the very least there will be a steep learning curve in the next few years until we figure out the correct busineses, algorithms, and systems for Big Data (there was just a very interesting conference on the topic here at MIT last week), audience researchers may become lost in a sea of overwhelming information.
ReplyDeleteI sympathize with Angela when she describes the anguish of being a consumer on the wrong end of an ‘audience analysis,’ losing a show with high quality content to the whim of the masses! As a child I remember being vividly disappointed when the show Earth 2 on NBC was cancelled after just one season - too much science fiction, with an emphasis on ‘science’, said the low poll ratings experts. (It featured a boy in a solar powered wheelchair. Cool!) But it begged the question then as it does now - are we using ratings to drive ourselves towards a “weighted medium” where all shows or most content will follow a pleasant, generic middle, and lose out on the spectacular highs and lows, or lefts and rights, or ups and down, that the extreme content on either end of middle America has to offer?
The question is: What is the danger of following audience analysis too closely?
The answer that comes to mind: An incomplete picture of consumers preferences, which weights their actions today as predictive tomorrow and ad infinitum. Haven’t you ever been hooked on a new topic, or learned about a niche program, idea, or music, as a result of that content’s development over time?
Engagement, a topic little covered by both Napoli & Webster (as Robert also pointed out), can’t be dismissed or oversimplified. If my goal is to create lasting content that drivers viewership in the long-term, my audience will need to be engaged with more than a passing interest, or a single season of morbid curiosity. An overlay of engagement planning could mean the difference between being a Desperate Housewives and a Dr. Who… but I’m showing my consumption preferences here! Audience analysis will have to cover this more closely in the future.
To that point, I loved this quote from Webster: “Of theoretical interest are questions about what motivates media selections, what stimuli command attention, how media are used in everyday life, what meanings and/or pleasures are derived from use, and how people engage media as consumers and fans.” Syndicated audience research may become a thing of the past as the need for more specific variables and cross-reference in an audiences’ attachment, engagement, and viewing, are needed by increasingly complex media content producers.
From a financial perspective, engagement investment makes sense. As audiences begin to squeeze themselves out of traditional boundaries in an increasingly connected, digital world (i.e. when we can no longer assume that a middle aged married man in Ohio is necessarily interested in golf.. maybe he loves World of Warcraft!), it is the engagement of an audience rather than it’s demographics which will become valuable information. This in turn will inform producers’ ability to design content based on audience motivation, rather than audience attribute - a shift towards an age of user-diverse, or even user-driven content, in yet another field.
Ah, yes, Sarah, I, too, am horrified by a 'pleasant, generic middle.' For too long we've had programming that is only a quarter inch think but a mile wide. I can see, sadly enough, why the networks do it. "The Tonight Show with Jay Leno" had to serve many masters. Jay, believe it or not, is quite funny. (Even David Letterman use to say Jay was his favorite comedian.) But show execs couldn't target those that would most be engaged by his usual brand of humor -- much bluer as it ran to both jokes and politics. Because he had to appeal to everyone, Jay had to dumb down his monologue. He needed to serve the masses -- not offend anyone, and make certain to tell an equal number of jokes ribbing both Republicans and Democrats. (yawn) How nice if we can better know our constituents and target programming that engages directly to them.
DeleteOh, the good old days. When I started working at NBC in 2001, I heard a lot of grumbling from writers and production folks that ‘the suits’ were killing the essence of what constituted good TV. These newly hatched MBAs were diminishing the creativity with all their talk of numbers, they said. The suits would look at Q ratings and such, maybe how something was trending on HSX, scan Variety and the Hollywood Reporter and then watch how the shows were greeted at the ‘Upfronts’ and where advertising dollars were meted out to determine what shows or pilots would get a green light or be cancelled. That all seems pretty nebulous in comparison to how data driven we are today.
ReplyDeleteI believe what we’re doing now to track engagement online is pretty stellar but I still think we have a long way to go with film and TV. I’ve looked over Listen First Media’s offerings and I’m not convinced that they have the ‘end all, be all’ solution with measuring DAR versus exposure ratings either. In a posting back in February on their blog, they were trying to equate the buzz around “The Wolf of Wall Street” to future ticket sales and to whether that bump in buzz would see it get more Oscar votes. The film was definitely trending the week preceding the Academy Awards, but the film dipped -28.3% at the box office (according to Box Office Mojo) and won none of the five awards it was nominated for. Buzz in the case of that Leo/Jonah feature maybe meant fans were more engaged but it didn’t translate into a bigger bounty.
Alas, if only every TV show could be on Netflix, how lucky they would be. Netflix has 50 million worldwide streaming customers. Traditional television networks don’t have the data like they do which allows them to make better decisions and keep their customers happy. On a regular network, there is a 65% chance a new show is going to be cancelled. Giving a green light to a pilot is tricky because it’s pretty much all based on intuition. Netflix is fortunate to be an Internet company --with all users governed by an IP address and a login, they get to know their customers’ watching habits really well and make savvy picks in content accordingly (House of Cards, etc.).
Here’s a question, cable boxes are very intuitive and know our viewing habits, right? Comcast owns NBC. Why isn’t that information helping the Peacock network get a leg up when it comes to choosing good programming?
Whew, catching up now on the 52 comments, it's like reading a book just to get in on the conversation.
ReplyDeleteI’m not going to rehash all of the excellent opinions that you’ve all given. I’ve spattered some comments through them and feel like I can’t add any additional thoughtful information on the “status quo” prompt given. I’ll instead attempt to add to the conversation with talking about different biometrics that could be used in scale to accurately associate emotional, cognitive and physical metrics to the content.
I feel biometrics will become more commonplace as the devices become cheaper and more ubiquitous. The latest technology in the XBOX Kinect 2 shows how easy the implementation of this type of technology could be. It would avoid fatigue and user-error, while allowing for scalability and measuring engagement in an incredibly accurate way. The Kinect 2 can identify users automatically and track their movements and associate voices to them (up to six at a time). It can measure all body movements (including skeletal position and muscle tension), blink rate, body temperature and even heart rate.
All of these individual biometrics, uniquely assigned to an individual automatically, can easily be synced to the content for a to-the-second record of all of the reactions and engagement to the content.
Blink rate can be used to extrapolate lots of information, including fatigue, mental excitement or stimulation (especially with dopamine receptors), social interest (if they feel a social connection to a character), anticipation of an event (anything from a kiss to a scare), among many other associations (even schizophrenia!).
Heart rate can monitor excitement/relaxation, emotional stress and if tracked over a period of time, desensitization towards certain types of content (like violence or sex). This measurement is similar to galvanic skin response associations.
Eye tracking can be useful for a plethora of applications in media research. It’s one of the most reliable methods of understanding interest and usability (more accurate than subject recalling), it can measure fatigue, character preference, interest, and most importantly, if the subject is actually looking at the TV.
In addition to the biometrics, this technology can analyze the conversations being had around the content (including who is saying what and their approximate emotional states). It can also scan for any additional content being consumed (like if a podcast or music is on in the background). Lastly, it can identify logos within the scene, so for example, it can associate a can of Coke being drank or a sports jersey to a specific individual (the latter Microsoft has a patent on). And of course, the IP would be logged and can be linked to internet habits and social media accounts.
These technologies can take the exact, in-depth nature of academic studies and scale them easily while maintaining great accuracy. The hardware is itself fairly cheap (~$100), although the R&D and software/patent licensing would increase the costs.
Is this the audience researchers dream, a hybrid Orwellian/Huxleyian dystopia or something that would require too much hassle to implement correctly?
Here’s some of the articles I dug up for this (others I have read before, but don’t remember the links/books right now) :
http://www.psychologytoday.com/blog/beautiful-minds/201004/why-creative-folks-blink-lot
http://archpsyc.jamanetwork.com/article.aspx?articleid=209916
http://sfari.org/news-and-opinion/in-brief/2012/cognition-and-behavior-blinking-measures-social-interest
http://hfs.sagepub.com/content/36/2/285.short
http://www.sciencedirect.com/science/article/pii/S0022103106000825
http://www.researchgate.net/publication/26411583_Eye_tracking_and_its_application_in_usability_and_media_research/file/60b7d51b71597a5db5.pdf
http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=1297400&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D1297400
http://archpedi.jamanetwork.com/article.aspx?articleid=204790
And of course, the one thing I didn't mention with this technology, is that it easily allows you to measure what individuals are watching which programs at which times, if applied to the entire XBOX entertainment offering, where it's able to read what programing you have on through it's HDMI connection to the cable box. And if not, it can use audio analysis to identify the programming.
DeleteHi, Zack-- There's no doubt that biometrics will play a more important role in audience research in the near future. Even though many of the measurement techniques have been around for a while, they've just recently taken on a more significant role in audience research. As the devices become less intrusive (think wearable devices), I think we'll see more and more of this. The question will be whether or not this becomes something that consumers will accept (my guess is yes). As for Microsoft's foray into the entertainment industry, they have very quietly put together an entertainment center that has access to close to 50 million subscribers and their game playing/video watching preferences. Unlike other digital gaming platforms (like Steam's 60 million subscribers), Microsoft has set itself up to become a major media hub for American homes. It'll be interesting to watch.
DeleteBased on the challenges that Napoli outlines, It does not appear that audience researchers can continue to employ the same techniques and expect to garner an accurate representation of viewing habits. One solution to acquire more accurate television ratings is to remove the active participation of the viewer completely. Napoli points out that younger generations are more likely to complete surveys or participate electronically, while older generations are more comfortable with paper methods. This difference in preference often skews the demographics that are represented in audience research data. The key to generating accurate data over a large swath of varying demographics lies in reducing the barriers to compliance to zero. When a new customer subscribes to a tv package they should be asked if they want to contribute data about their viewing habits to audience research. If a viewer opts in, they fill out a short survey that includes demographic data and no longer required to complete any further steps to participate in the ratings or audience research system. A similar type of system could be used with internet activity as well. Napoli points out that more sophisticated ratings systems come at a higher cost. However, this method would be fairly inexpensive because it would not involve developing or distributing new equipment to participants.
ReplyDeleteIn the branded content realm, measuring views is not enough. According to a recent Mashable article, the average American adult spends 11 hours per day interacting with media. With this much media noise in our lives, it is nearly impossible to expect a user to recall information accurately or a desired take action after only one viewing. Consequently, viewcounts don't carry as much weight as they used to. In my experience, engagement is the factor by which many brands measure success. Whether or not an audience takes a desired action after they have viewed the content is the data that advertisers are looking for.
Napoli’s most recent sources are from 2008 and many of them are from the 90s. Although I couldn’t find the exact date, I’d imagine that this article was published no later than 2010. Audience researchers have made great strides in tracking the worth of engagement, even its possible economic value within the past few years. A 2012 study from PulsePoint says, “Companies that fully embrace social engagement are experiencing four times greater business impact than less engaged companies.” Although more accurate measurement tools have been developed, the platform on which consumers want to engage with brands is always shifting. The audience researcher’s challenge today lies in being able to predict where the audience will prefer to consume media next.
Sources:
http://mashable.com/2014/03/05/american-digital-media-hours/
http://www.pulsepointgroup.com/sites/default/files/SEE_Paper_4_27.pdf
Hi, Christina--
DeleteThere's no doubt that the proliferation of new devices has altered how we collect data on audience members. Grocery store cards keep track of our purchases; credit card 'points' programs; 'member' cards at retail outlets; Xbox 360 Kinect with its camera; Google keeping track of searches, email content, web content, etc.; snapchat gaining access to all info on the phone; Facebook's subtle policy changes regarding privacy, etc. None of this was much of a reality a decade ago. An issue to consider might be where the line is between
effective, in-depth audience research,' and 'invasion of privacy.' I would argue that audience researchers need to make sure that consumers are 1) aware of the data being collected; and 2) consent to such data being collected. It seems to me that our policymakers, who are always years behind the technology, will move in the direction of requiring explicit permission by consumers to any outfit collecting audience data. Excellent comments.