ADV 483 Quiz 2What do these measures tell us about the audience?When we measure audiences, what do we want to know?● As an advertiser and publisher/media owner○ Want to know audience size - how many people reached and level of activity○ Audience profileReach: the total number or percentage of users that a media item reaches in a givenduration● The most basic measure of how many people reached● For a TV channel: people who watch a channel at least once (for at least a minute)○ In a given week - weekly reach○ In a given month - monthly reach● For a website: people who visit the site (any page) at least once○ In a month - monthly unique visitors (or reach)○ In a week - weekly unique visitors (or reach)○ On a day - daily unique visitors (or reach)● Each person is counted only once regardless of how often/long theyvisit/watch/read based on an acceptable definition of what counts aswatched/visited/read○ TV channel: 1 min○ Website: 1 page viewed for at least 3 secondsComScore● Does Comscore provide reach?○ Yes, a website reach is it’s unique users expressed as a percentage of totalinternet users○ Since comscore reports data on a monthly basis, the reach reported byComscore for a website is its monthly reach9/29● Combined reach = Outlet 1 + Outlet 2○ Ex: Champaign DMA■ 10,000 people watched monday night kickoff (pre-game show)■ 15,000 people watched monday night football games■ 5,000 of these people watched both■ What is combined reach for these programs?■ Combined reach = Reach #1 + Reach #2 - (overlap)■ 10,000+15,000-5,000=20,000● Comscore also reports another measure: average daily visitors● Average daily visitors = average of the daily reach (daily unique visitors) of thewebsite over the month○ Total unique visitors not equal to just daily unique visitors*30● Youtube videos○ Does the # of views of a youtube video equal its reach?■ No people watch more than once○ Is the reach of youtube equal to the sum of the reach of all youtube videos?■ No people watch more than one video“Gross” Measures of Audience● Measures that capture the total level of audience activity (regardless of whether theycome from the same or multiple users)○ Views, minutes, visits, box office● Map measures to measurement methods○ Can gross measures be estimated from a server centric approach? Yes● Server centric methods can estimate most gross measures but for measuring reach(unique users) one needs user-centric methodsGross Measures in TV● Rating: the percentage of individuals (or households) that watch a TVchannel/program/a segment for its entire duration● Ratings = number of people or households watching/total tv viewing universeCalculating a rating● Imagine these 10 people watching tv at a given point in time● Rating of channel 2 (3/10)*100 = 30%● Rating of channel 3● Share: viewers to a channel/total number watching tv (PUT)*100● PUT = persons using television● Share of Ch 2 = 3/6 = 50% (higher than rating)● Programs have different durations not everyone may watch the whole program● Ex: Lady Gaga Superbowl halftime show viewed by 117.5 million people, 6 millionmore than game watched itselfRating is an “average”● Assume a 60 minute program has 2 segments○ Segment 1 of 30 minutes had a rating of 10○ Segment 2 of 30 mins had a rating of 20○ Rating of whole program: 10+20=15■ Remember the segments are now of unequal duration■ (10*20 + 20*40)/60 = 16.67● Instead of simple average you take a weighted averageRatings vary by markets● In Philadelphia, the Super Bowl had a rating of 50● Chicago: rating of 40○ Philly has 10 mil tv viewers○ Chicago has 20 mil tv viewers● Combined rating: total viewers of program/(total tv viewers)*100○ Overall rating = weighted average of rating in Philly (rating 50, 10 millionhomes) and Chicago (rating 40,20 million homes)○ (50*10 + 40*20)/ (30) = 43.33● Ex: Oscar Live Stream on ABC○ First segment 30 minutes - 70 million viewers○ Second segment 60 mins - 50 million viewers○ Third segment 120 mins - 60 million viewers● Calculate average viewers per minute○ 30 (segment 1)*70 (viewers) + 60*50 + 120*60 / 30 + 60 + 120 (totalminutes)○ 2100+3000+7200/210 = 58.57 million viewers on average○ Universe was 300 million (given)○ Rating - 58.57/300 = .195 or 19.5%○ Rating is always expressed as a percentage but without the percentage sign● Rating is same as impressions○ Rating is essentially the same thing as an impression■ Ratings are for programs/videos/larger segments■ Impressions are for ads○ Both are gross measures○ Reach on the other hand is a cumulative (net) measure - we don’t countduplicates again● Common online measures○ Total visits○ Total minutes viewed○ Total pages viewed10/6Audience FragmentationHow do people allocate their time to available media options?How do advertisers respond?● Media strategy: mass audience vs. nicheTransition of a population from few large audiences to many smaller audiences● Homogeneous -> heterogeneousDivision of attention across platforms and outlets● Autonomy: ability to consumer media content any time and anywhereAudience Concentration● The audience still heavily consumes a few dominant outlets or productsCan you have both audience concentration and fragmentation at the same time?● Yes, you can have people watching at the same time on different platforms● Ex: Squid GameThe Long Tail - main ideaTheory: online channels will fatten the long tail● Power Law Demand Curve: one quantity varies as an exponential power of the other● Distribution: physical vs digital○ Blockbuster vs Netflix● Biggest money in the smallest salesTheory (Anderson)● Physical vs virtual retail○ Shelf space, cost● Many good niche products are available from virtual retailers● Many people prefer those niche products● People can find the best niche products through recommendation systems● The long tail will get longer and fatter, the head (not dominated by hits) smaller● Media culture will be richer and more fragmented● Evidence (Elberse)○ The long tail is getting longer, not fatter○ A decreasing number of popular offerings (hits) tend to dominateconsumption○ Hits still dominate among those who consumer niche products○ Niche products are consumed by heavy users○ Hits are better liked, on average, than niche products■ Double jeopardy● Lower market share brands also have