Yuma 4×4

Media and Communications

How Do We Measure an Audience?

How Do We Measure an Audience?


This FilmmakerIQ Lesson is proudly sponsored
by RØDE Microphones. Premium microphones and audio accessories for studio, live and
location recording. Hi, Welcome to FilmmakerIQ.com – I’m John
Hess and today we’ll set out on the task of answering the question – how do you measure
an audience? The most obvious way to measure an audience
is to count up all the tickets that were sold at the box office. Now for a stage play, this
is not exactly rocket science – you just add up the receipts of that night’s showing. But for a major motion picture release that
is playing in 4000 screens around the country or even around the world – this can get a
little bit more tricky. The basics are still the same – each movie theater would count
up the receipts and report to the distributor that day’s take. Now since the movie viewer
is the end of the sales line once he or she bought the ticket, we talk about the audience
in terms of dollars rather than tickets sold. Now with real time tracking systems installed
in most theaters, a distributor takes this gross revenue data from all the movie theaters
and adds them up to see how much a movie makes over say opening weekend. But wait a minute – you always hear about
which movie won the weekend at the box office on Sunday morning – before the weekend is
even over. Where do those numbers come from? Actually those are estimates. Each studio
has a particular formula for estimating the weekend haul based on historical data – comparing
the movie that’s in theater right now to a similar one that was released a similar time
of year with a similar advertising budget. If we know Friday and Saturday’s take we can
make a fairly accurate estimate of what Sunday’s take will be. Then on Monday afternoon the actuals numbers
are totaled and printed in trade journals. So why go through the trouble? One reason
was to account for some of the movie theaters that didn’t have real time tracking. Previously
only 90% of American and Canadian theaters were tracked by Rentrak and Nielsen EDI. But
in 2009, Rentrak buys up Nielson EDI becoming the sole provider of real time (or at least
near real time) box office receipts and now boasts coverage of 99% of all theaters in
the U.S. and Canada. These results are purchased by the studios
and yet they’re still releasing box office predictions before the weekend is over. Why?
Well the answer may be buzz. Having the figures for the weekend box office gives Monday morning
news programs something to talk about which in term makes it into water cooler conversations
and you just generated some free word of mouth advertising for the film. But while we’re on the subject of Box Office,
let’s try to figure out what some of these industry terms mean. Now this is a topic that
can get extreme dense and we will probably do a video on budgetary numbers at some point
but I want to give you just a taste so when you see these numbers there’s at least a little
context. And because there’s a lot of money at stake, take these percentages and averages
with a grain of salt because there’s a lot of creative accounting going on. When the numbers are announced, they are gross
numbers – that is the total amount of revenue generated before any expenses are subtracted. The first in line to get a cut of the gross
is the exhibitor – the movie theater itself. Historically in negotiating with a distribution
company, a theater would determine it’s base costs of running the theater called the nut.
Then a sliding scale was negotiated so that the distributor gets the majority share on
opening weekend – upwards of 90% of the gross falling 10% every weekend. This created an
incentive for the studios to drive audiences to see a film on opening weekend while incentivising
exhibitors to hold movies in theaters as long as possible as they would make a bigger chunk
of the ticket price on older movies. But with the shortening of the release window
and competition between Video on Demand and Streaming, recently exhibitors have moved
more toward a split based on national box office results – with a straight up split
between 48% and 63% of the box office – the better the sales the more the distributor
takes. The result is movies go in and out of theaters quicker and pay off faster. 80%
of the box office revenue for a film is generated in the first 2 weeks of release. Because theaters are only making half on the
movie ticket price, the real money in the theater business comes from the rather large
markup on concession stand items. Popcorn and soda cost mere pennies to make but the
theater gets to keep 100% of the profit. So when you look at how much money a movie
grosses – the first thing you have to do is cut that in half. That’s how much the distributor
gets. But wait, we’re not close to net profit yet. According to Baseline Intelligence, a research
firm for the entertainment industry, the six major studios spent an average of 37 million
dollars per picture for print and advertising in the year 2009. Now that’s only average
– the major tentpole films can spend upwards of 85 million dollars. The Print side – the P of the industry abbreviation
P&A – covers the cost of creating actual film prints for distribution. 35mm film prints
can range from $1,000 to $3,000 a piece depending on length of the film and quality of the print.
For a wide distribution of say 4,000 theaters that can really add up fast even with quantity
discounts that the studio can muster. Which is why, to the chagrin of Quentin Tarantino
and Christopher Nolan, studios are pushing so hard for digital projection. According
to the MPAA, traditional projects made up 81% of the US market in 2009 but four years
later in 2013, that number is down to just 7.5%. The fast move to digital almost mirrors
that of sound conversion in speed although digital’s drive was cost savings and sound
was a new dimension in storytelling. Still the Print only makes up about 10% of
the P&A – most of the rest of the money is spent on advertising. And this is where it
can get really complicated if not even a little “creative” with the numbers. In an increasingly
media saturated world – you really have to try to create near a instant brand awareness
of a film in just a matter of a couple weeks and the only way to do this is with a well
crafted advertising strategy involving, television, social media and even bill board advertising.
How that money is spent is where studios can inflate costs – for one many studios are part
of larger media conglomerates. Paramount is a subsidiary of Viacom which owns CBS and
a host of cable channels. So buying air time for a Paramount movie from CBS is almost money
shell game with really no limit on the price Creative Accounting aside, According to Larry
Gerbrandt who wrote in the Hollywood Reporter, Accounting that an average of 55% of domestic
sales go into the exhibitor’s pockets – the average 2009 movie release (again average)
had to do gross $186 million dollars in the domestic box office in order to recoup the
production and P&A costs, and that’s not counting the percentages taken up by deals struck by
producers, stars and production companies. Now remember these are all averages but $186
million domestically is not something that the vast majority of movies are able to accomplish. But don’t shed a tear for the studios just
yet – there are several other windows for studios to earn money – first through international
box office receipts which can sometimes dwarf domestic, then through DVD/Bluray, streaming,
Video On Demand, merchandising, and of course licensing for television. I find it rather ironic that when Television
first hit the entertainment scene back in the 1950s, the movie studios hunkered down
to do battle against this device that kept people at home away from the theater. And
then only a few years later, TV became savior of the movie business as a way to generate
even more revenue. TV signals are broadcast through the air from
a broadcasting tower. Anyone with an antennae (and now a digital decoder) can pick up the
signal and watch. So how do we determine how big a TV audience is? Well that question plagued Arthur C. Nielsen.
In 1923, Nielsen started his company AC Nielsen with the purpose of selling engineering performance
surveys. Soon his company was in the business of creating market research reports – In 1932
Nielsen began to track food and drug transactions giving brands a way of determining their share
of the whole market. The term, market share – your piece of the industry – was coined
at the Nielsen company. In 1942 after the purchase of Audimeter, Nielsen
begins to track the popularity of radio station programs using a sample of 1,000 homes. In
1950, Nielsen begins this service with audiences.Television Now a question you might have initially is
how can a sample size of only 1,000 or so homes generate an accurate report of a population
of say 10 million TV sets? Well the answer to that is statistics. Obviously the larger
your sample size, the more accurate your results will be. If we poll the entire population,
we will get a perfect breakdown – in theory. But polling an entire population isn’t plausible
for cost and time reasons so we have to take a sample. Without getting too deep in the
math, if your sample is truly random and free from bias then you can calculate the margin
of error and it turns out you really don’t need that big a sample to get the margin of
error down to about 1-2% But that is if your sample is truly random-
and that’s not necessarily easy task o accomplish. So to track Television viewing habits, Nielsen
relied on two methods of reporting. The first were diaries. These were paper logs of what
you watched and how long you watched over a period of a week but as you can imagine,
diaries are prone to error – people forgetting to input data or even downright cheating and
falsely reporting their viewing habits. The second method Nielsen used were set meters.
These were devices installed on the TVs of Nielsen families that recorded the use of
the television and would telephone home base with the results every night. Set meters could
record minute by minute data to track the viewing habits of a household without error.
But the meter didn’t know who if anyone was watching, only that the TV was on. In the
1980s Nielson added People Meters which could now track the viewing habits of different
individuals in a household. In the 1990s the Portable People Meter came out which listens
for radio, television and cable audio to report back – the latest generation uses wireless
cell phone signal. Unfortunately in a terrible missed opportunity, they made them black so
we will never get the pleasure of saying One Eyed One Horn Portable Purple People Meter When you see Nielsen ratings for a national
show they are broken down into three figures – rating point, share and viewers watching.
In 2013 there were an estimated 115.6 million television households in the United States
– called the universe – a national rating point is one percent of that – or 1.16 million
households. To determine the rating point we divide the number of viewers called impressions
by the universe – in other words a rating point is the average percentage of people
watching of all possible TVs. The second part is the share. That is the percentage of televisions
that were IN USE use during the time we’re trying to measure, the share will always be
higher than the rating. Viewers watching is an estimate based on how many people are watching
per household – this can vary from program to program – imagine the difference of viewers
per household of the latest guilty pleasure reality show versus an event light sports
game like the Superbowl. Here’s a real world example. Lets saw we have
a show that’s airing at 8PM on CBS. It’s national rating is 2.8 with a 9 share with an estimated
10.64 million viewers. That means 2.8% of all televisions are tuned that program which
represents 9% of all televisions turned on at that hour. Now let’s compare that same
channel at say 10PM which has a rating of 3.4 and a 10 share but only an estimated 9.29
million viewers. The number of viewers is down but the share is higher because perhaps
more people turned off the TV and went to bed. Of those that are still watching TV – now
10% are tuned in – up from 9% at 8PMl. But this only applies to viewers watching
live – what about time shifting watching on a DVR? Well for this Nielsen adds three more
streams of data – there’s Live, Live + Same Day, and Live + 7 Days. Services like TiVO
have deals with Nielsen to provide this data to be included in their measurements. So now what are Sweeps? The term dates back
to 1954. Well remember the paper diaries? Nielsen just didn’t have the manpower to log
all the paper diaries across the nation at the same time – currently they send roughly
2 million of them in 210 television markets. So what they did was to sample viewing habits
for one week four times a year but stagger their sampling by territory starting with
the east coast and the sweeping west over a period of 4 weeks.. So Sweeps is really
a national 4 week affair that occurs in February, May, July, and November. Well if these were
the months being sampled, broadcasters quickly realized that it was to their benefit to get
the highest ratings in these few months before the diaries had to be turned in. Now why go to all the trouble to measure the
size of a TV audience? The answer is money – advertising rates. The higher the rating,
the more money the network can ask for advertising on a particular show. Each 30 minute block
of television contains 8 minutes of commercials – 6 of those minutes are dedicated to national
advertisers – their price is locked in with the national ratings and share metrics. The
remaining 2 minutes are dedicated to local market advertising and those rates are determined
during sweeps week. Now, as I have grown accustomed to saying,
this is just the tip of the iceberg when it comes to television metrics. There’s data
on commercial viewership and ratings can be broken down along geographic and demographic
lines. But the TV landscape is changing and changing
fast as more and more media is now being consumed through the internet. Computers are quite simply amazing things
for communications. All those problems we had with counting tickets at the box office
or logging paper diaries at Nielsen are completely avoided with computers who can tireless sit
and count views 24/7. Online video on demand services like Netflix, Hulu and Amazon can
track your profile and know exactly what you watch, how long you watch it and when you
watch, the ads you click on et cetera et cetera. The information age isn’t just about information
being at your fingertips, its about using the information we gather about the you and
what we do with it. But since you are paying for those legal streaming sites, there’s not
really a lot of incentive on your part to inflate your viewcount. But that’s different when it comes to a monetized
platform like YouTube. YouTube’s parent company, Google, has been in a battle with black hat
SEO and spammers for a long time. Google wants to make a better experience for the web user
(while selling advertising to you) while the black hat SEO and spammers want to generate
as many click throughs for quick financial gain. On YouTube – views are like currency and they
really become currency through ad dollars – so inflated view counts mean wasted advertising
dollars and unhappy advertisers. Generally tight lipped about what exactly constitutes
a view, YouTube has gone on record as defining a view as a video request generated by an
actual viewer who got what they wanted and had a good user experience. So videos that
are set to autoplay on a website will not count because no user actively requested it. Now things get a little more complicated.
When you upload a video to YouTube, copies of it are cached in servers all around the
world. That way a person in say India who requests your video can contact the local
Google server rather than having to call up a server that’s half way around the globe. Now much like the Box office we talked about
earlier, periodically each local Google server will call up tje central view count server
and report in how many people watched the video. The central view count server adds
up all the views and displays it on the site. That is until it hits 300. Currently there are 100 hours of video being
uploaded to YouTube every minute and that number keeps going up. Now honestly the vast
majority of those videos will not do more than 100 views or so. Even if monetization
is enabled, these channels won’t make a penny through advertising so you don’t really have
to use precise view counting on those videos. But once the view count hits 300, YouTube
begins to take notice and applies a stricter algorithm to verify that indeed those views
are coming from legitimate sources – that they are user requested and not tricked into
clicking on it.. And that’s why you will often see a video hit 301+ and sit there for a few
hours while the server goes back and verifies the views. Views are still being counted during
this process but not shown publicly. Eventually the server catches up and the views are accounted
using this strictly counting algorithm, From Box Office, to Television Metrics, and
finally Internet tracking data we’ve just touched on the science of measuring audience.
People develop full careers just on this stuff alone and these statistics have the power
to move millions of dollars in determining what gets greenlit and what doesn’t. That’s
the power of the audience, now that the tools of media creation are getting better and cheaper,
having an audience is becoming more crucial. But in the end, they still just want to watch
and experience something great. I’m John Hess and I’ll see you at FilmmakerIQ.com

65 thoughts on “How Do We Measure an Audience?

  1. Do you have any idea of how much help you are providing to aspiring film makers such as me. 🙂 Wonderful lesson John. How often do your videos come out? Once a month? And what all online resources do you have excepting your YouTube channel and website? Thank you. 🙂

  2. If what you said at the end is true – about youtube's counting method – then why would they add the autoplay function?

  3. 1) Surprised to see a video as FilmakerIQ hasn't updated their site in a while.
    2) There's some kind of distortion that makes John's beard look like a pencil animation.
    3) The other measure that could be discussed is the test audience.

  4. Great video! Your channel is simple amazing. May I make a video suggestion theme? Maybe you could make some video talking about digital VS film, and you know… telling about the whole history of the digital cinema, 4K, 8K, and so on..

    And I don't know if it's something that you have in mind.. But I think, definitely, many people would enjoy.

  5. Very informative video.
    Youtube actually looks so hypocritic. At 19:28 John says that a view is considered a view by Youtube if a person got what they wanted and had a good experience, so autoplay videos don't count. Right? But lo and behold, Youtube is introducing their autoplay feature which plays videos people don't want and which doesn't provide a good experience. It's just…. ugh.

  6. Love the videos
    Was this shot on an A7s? The noise in the darks leads me to think its 8bit slog2 shot at "correct exposure"
    Also couldn't help but notice a boom shadow

  7. The question, however, is whether or not measuring the mere amount of views is a good way to determine the value of an online video for advertisers.

    So for example, I'm both subscribed to Filmmaker IQ and FailArmy: Your videos make roughly 10k-50k views each (neglecting occasional viral ones), while FailArmy videos make about 2m-5m on average.

    But a FailArmy video just doesn't matter to me as much as one of yours. 5 minutes after watching I have already forgotten about what happened. After all fails are not something you extensively discuss with others either. They are mere distraction.

    Your videos on the other hand stay in my head for much longer. They are interesting and helpful and inspiring and that's why I truly care about them. They ultimately enrich my life in a way.

    So I think it's fair to say that views very much differ in quality. Shouldn't that be measured in a way, too? Because as an advertiser I'd rather be interested in whether the audience is truly engaged or just simply distracting themselves than in the view count alone.

  8. IMHO. Theaters relying on popcorn and soda revenues is a flawed business strategy undermining the entire industry, and it will eventually put the movie industry in comatose state. Selling pop corntwice the price of the ticket is stupid. It should be the other way around.

    Given the fact that "most" theaters are located inside shopping malls, movie tickets should on the contrary bring along some benefits (like discounts) for the moviegoers in other stores and food court etc. People love discounts and this I believe will perpetuate the business.

  9. I love these videos, keep up the good work! Could I suggest a video on animation or animated features? I'm curious on how they affect the film industry and the role they play.

  10. Thanks so for such a comprehensive answer to a question I have asked when I see these seemingly inflated figures.

  11. Filmmaker IQ always provides informative and entertaining videos! It would be cool to see a behind-the-scenes of an episode. 🙂

  12. Great video again John!!  I liked the final part… Youtube is really amazing for filmmakers when you have a lot of information about your videos. I'm sure tv will dissapear in the future  and users'll join some playlist or youtube channel. In fact, here ads are more direct and effective.

  13. John, I'm sure you hear this all the time, but your videos are amazing!  How about a "behind the scenes" video at some point?  We can learn how to properly prep a script, rehearse, perform, and edit.  That would be an awesome experience for us, your audience!  Thank you for all you do!!!  🙂

  14. At 14:30 the rating for the 10PM show goes up. If fewer TVs in the Universe are on, shouldn't the rating be lower? The share is higher because of those fewer TVs, a higher percentage is tuned to the show. Am I misunderstanding the rating as a percentage of the Universe?

  15. Hi, love your videos! Especially the science and history of colour in film. I think a video about the History and Science of the Film Dissolve would be amazing!

  16. Box Office Numbers are in… But what do they mean? Check out our lesson on how to measure an audience:

  17. Love your videos. Best channel. I wanted to request you if you could do a small video on the film conversion process. You know, like from film to digital. I ve heard that a lot of directors still shoot on film and then convert it to digital. Why not just shoot it digital without going through all the hassle?

  18. I have to wonder if the desire for high resolution data on viewership will one day spell the end to scheduled television outside of live feeds such as news and sports.   the DVR already has millions or even tens of millions not watching shows in their original time slots.   But with VOD a company like Comcast has very precise data on what shows are watched.

  19. Please make a video on budgets!! I see net gross and pruduction budget numbers but have no ideas whether the production company has made money. After being splited half for exibitors, is the gross again cut in half for distributors?
    And whose money (distributor's or producer's) is used for the movie production?

  20. Do modern cable systems make Neilson obsolete? I would imagine its trivial for Comcast to know what I am watching. And due to how HDMI works the box knows when the TV set is actually powered on which would prevent false positives from boxes left on.

  21. Your content is some of the best I've seen on Youtube, I've learned so much about films and the science behind them in the time I've been subscribed to your channel.

  22. There's some interesting stories around PPMs…. To make sure they are actually used and not left home with the radio on (remember ppl are paid to carry these), they detect movement. Apparently someone once attached a bunch of PPMs to a ceiling fan in a room with a radio on.

  23. What I guess I don't understand is why the number of tickets are never reported. If actual ticket numbers were widely reported success of a film could be understood for decades regardless of inflation of ticket prices. Star Wars and Snow White would still be in the top ten most successful, and there would be a better understanding of the impact of those films for society. Surely now, if not in the 20th century, electronic ticket numbers could be easily counted.

  24. wow, i`ve always wanted to know more about TV audience measurement, but now i know about Radio and theaters as well, thanks this was very informative.

  25. +filmmakerIQ I'm and eleven people ear old and has been getting really into filmmaking. Tried to find another channel to learn from besides film riot, DSLRguide, d4Darious (they are all great channels). You just earned a sub.

  26. I first tout that tv's and radio's dit had a builtin people meter builtin to anonimically tell the tv broadcaster the amount of watchers at wich time sothat a analogue/digital computer could messure out what you were watching based on the time,so for example channel A knows you watching football based on tome 17,00 o clock while channel B knows you were watching a commedy based on time 20.00 o clock etc,,, and radio channel A knows you were listening to that program based on time 15 o ck etc,,,you may think but how can you transfer a broadcast and a back signal from and back to those broadcasters without interferring eachother on 1 antenna or cable,simply by decoding the back signal in a high freq while the recieved signal is a low freq,but no,there is no fair estimation about it.

  27. But these days dont they just use a blue ray disk to give the theatures to play the damn movie because even with the case its cheaper, stronger then flim, easier to ship and store, and the 16 year old can probably press play at the right time to start the movie (i have no idea how movie theatures actually work besides the fact you go into a big room that gets dark and plays a great hd movie with great sound) since i immagine film is fragle and a bit hard to load into a projector

  28. Alao with tv ratings i would think that neslon ratings are out of date because if tv networks want to cash in they could probably work out a deal with cable providers disk network and comcast and such were the data of what each cable box is streaming and recording being sent to the tv networks and counted on a server i say this because well my tv shows me what current show is being played on a channel and can even show me a show recorded on another cable box in the house so i dont think it would be too hard

  29. I love your channel. I suggest everybody to watch all these videos multiple times.

    is there a chance you can make a video on the movie studios and the film libraries and mergers/acquisitions?

  30. i still finnd it sad tha the medium has become cheap for the average joe to produce. mainly because the the amount of content has double and the quality as gone down, and whatever it isn't the case now poeple have so much media to entertain themselves that they get bored quickly and switch to an other producers, that mean that the online market has become so saturated that no one can maintain a format and expect a career with it and people don't like when the "actor" of their media change the format, because they have identified a actor with a specific format. so they are bound to be forgotten eventually. just check Kwebbelkop he just play gta and scream. people eventually just get bored has what it is now call quick novelty of entertainment. i mean even the videogame industry cant keep up. they now have to hire for 300 m employees to devlop their game. and they, like movie producer only make their money out with microtransactions and dlc.

  31. i still finnd it sad tha the medium has become cheap for the average joe to produce. mainly because the the amount of content has double and the quality as gone down, and whatever it isn't the case now poeple have so much media to entertain themselves that they get bored quickly and switch to an other producers, that mean that the online market has become so saturated that no one can maintain a format and expect a career with it and people don't like when the "actor" of their media change the format, because they have identified a actor with a specific format. so they are bound to be forgotten eventually. just check Kwebbelkop he just play gta and scream. people eventually just get bored has what it is now call quick novelty of entertainment. i mean even the videogame industry cant keep up. they now have to hire for 300 m employees to devlop their game. and they, like movie producer only make their money out with microtransactions and dlc.

  32. I have been watching your videos for some months now. I am just now watching this one [2017!!], all your videos help me understand something. But this one has really caused me to look at youtube and how to use it in a whole new different light!!! Thanks and keep making these in-depth and educational videos. They help give us all a great respect for all aspects of film making!!!

  33. I always thought that sales figures should be given in ticket numbers not dollars because 1 person in 1939 is the same as 1 person in 2039, but $1 is not. But I understand why Hollywood doesn't want to do that because it would become too obvious that attendance has been essentially in decline for decades.

Leave comment

Your email address will not be published. Required fields are marked with *.