7 ms·

YouTube Myth Busting

14 points by alex_c 17y ago

alex_c 17y ago
Fairly short post, the interesting bit: Myth 3: Traffic, growth, and uploads are bad for YouTube's bottom line. There's been a lot of speculation lately about how much it costs to run YouTube. With revenue estimates ranging from $120 million to $500 million, and costs on an equally large spectrum, it seems people can pick any number to fit any theory they have about our business. The truth is that all our infrastructure is built from scratch, which means models that use standard industry pricing are too high when it comes to bandwidth and similar costs. We are at a point where growth is definitely good for our bottom line, not bad.
amadiver 17y ago
With awesome hits like "Worlds Apart", "CHIC.TV Models", and "Spin Cycle", it won't be long before I ditch the old cable box for YouTube!* *Just kidding YouTube. I love you, and I know you'll get better stuff than old "Starsky and Hutch" episodes any day now.
- pmorici 17y ago
  YouTubes current obstacle to getting better content is going to be Hulu which a large number of major content providers have a monetary interest in.
  - shrikant 17y ago
    YouTube can then look at taking content to international markets (as opposed to the U.S-only Hulu), which make up a fairly decent chunk of their viewership..
pj 17y ago
These myths are officially busted. No they aren't. Myths are busted with science and testing and experimentation. All this article did was contradict the myths. It did nothing to show the truth or falseness of the claims. If you are going to use something like the high quality show, Mythbusters, to model your contradictions, you should learn something about science from the show and actually prove whether or not your contradiction is true. This post didn't even come close to doing that.
- jerf 17y ago
  "It did nothing to show the truth or falseness of the claims." This is a primary source. If YouTube says that more traffic makes YouTube more money, nobody else is in a position to contradict them. I believe that lying about that in public could have legal repercussions.
  - sunir 17y ago
    I'm with the parent poster. The myths are not busted until some hard facts come out. There are artful ways to say true things that make the reader jump to conclusions that they intend, but are false. For instance, by claiming that they are in a position that growth is good, they make us think they are making profit per view, but that isn't what they said. They could have any number of non-profitable reasons to like growth, like strategic market share or building a huge content database. Further, their refutation that advertisers are not afraid of YouTube may lead some readers to conclude that 70% of Ad Age Top 100 Marketers are paying YouTube directly, whereas many (most?) are not. They are either using YouTube as part of a social media campaign or advertising within some user content hosted on YouTube. It's not clear at all what's happening on their balance sheet, so speculation will continue. To what end, I don't know, but after "What is Twitter's business model?", guessing how much money YouTube burns a year is the Valley's favourite party game. ;)
    - jerf 17y ago
      "I'm with the parent poster. The myths are not busted until some hard facts come out." If you adopt this position, then even "hard facts" won't satisfy you, since, again, they'd have to come from YouTube, a source you don't trust. To be honest, that's the position I'm in; I don't trust them on this point, though I don't particularly care, either. I quite carefully didn't say that the claims are true in their most obvious form, merely that this is as good as it is going to get for now, and in particular, that there is no applicable "science and testing and experimentation" to be done.
      - sunir 17y ago
philwelch 17y ago
"The world premiere of Joy Luck Club director Wayne Wang's film, "The Princess of Nebraska," was viewed 165,000+ times during the first 48 hours -- the equivalent of landing the 15th spot on Hollywood box office charts." Yeah, if movie theaters were free and instead of you going to them, they came to you. If there's an "equivalent" to 165,000 tickets on an opening weekend of a Hollywood movie, it's probably closer to 165,000,000 page views. And that's in how much revenue is generated--for the production company, a film playing in an actual cinema would rake in more than a YouTube release of equivalent revenue, because your local movie theater makes almost nothing on ticket sales while YouTube probably makes a significant percentage of ad revenue. It's not like YouTube can give all ad revenue to content creators and sell overpriced popcorn and soda, after all.
- TFrancis 17y ago
  I think that the benefits of price (free) and availability (theater comes to you) are only likely to increase movies watched via YouTube. Once a movie does have 165M page views in an opening weekend (and one will), which will be a better measure of movie performance: the box office or Google?
  - Retric 17y ago
    That depends on the movie, harry potter pulled in 139.7M over its Wednesday-to-Sunday debut period. I don't see free with advertising adding enough people to make close to that much money. However, a movie like clerks could be vary profitable online.
  - philwelch 17y ago
    The best measure will always be revenue. Can YouTube generate better revenue for film producers than movie theaters? Not even television in its heyday could do that.
app 17y ago
How is bandwidth pricing built from scratch? You're transferring bits like everyone else. Sure you might be doing it CHEAPER, but they didn't rewrite the rules to peering. These myths aren't "busted," they just refuted reported numbers with "that ain't true!"
- timdorr 17y ago
  It very well may be different, since Google bought a ton of dark fiber a few years back. They may be running their own network over to whomever wants to peer with them (Read: everyone!). They can get away with free bandwidth because a lot of providers will let them connect for free if they're able to bring the bits directly to their network, not via transit with some other provider that's costing both sides money.
jgamman 17y ago
most of the stuff doesn't play for me in NZ so we're still a ways from upload-once-play-anywhere
100k 17y ago
I like this one: "Myth 5: YouTube is only monetizing 3-5% of the site. This oft-cited statistic is old and wrong, and continues to raise much speculation. In our view, the percentage is far less important than the total number of monetized views..." They say it's old and wrong, but don't give the real percent, and spin in terms of "total number of monetized views". Based on this answer I'd guess the percent of monetized views isn't much higher than the "myth", otherwise they'd have a better spin.
jacquesm 17y ago
I've been moderated straight into the ground for this previously, but google does not nearly pay for bandwidth what you'd be quoted if you went shopping for bandwidth. At a 5G commit you're down to $4/Mbit, over that it gets quite a bit cheaper still. Youtube is consuming thousands of times that quantity. (the 5Gbit/sec will serve out 10K streams @600Kbps) Another good trick they pull is to cache videos that are requested frequently in or very near to the colocation offered by the larger ISPs.
- philjr 17y ago
  Yup, they peer at local internet exchanges (e.g. LINX in London - www.linx.net / INEX in Ireland - www.inex.ie), cache locally where they can (uk.youtube.com / ie.youtube.com are what you get redirected to periodically over here) and I would imagine that this drastically reduces their "internet" bandwidth consumption.
  - andyking 17y ago
    I use the O2 ISP and I often notice that if I stream a popular video from YouTube (or anything from the BBC iPlayer) my browser will tell me it's "transferring data from" an IP address which is in my ISP's own range. It must be cheaper for them to cache popular content within the ISP system and send it out that way.
jimmybot 17y ago
What I'm curious about is why they bothered to do the myth busting. What's the motivation to do this? If everyone thinks even the mighty Google can't get YouTube to break even, then no one else will try to compete with them in video, and when they get around to figuring out how to make money, they will own the entire whole short video market. Meanwhile, Google overall remains a desirable place to work in the eyes of many engineers. And Youtube advertisers care about how effective their ads are relative to what it's costing them, not whether or not Google is making money from Youtube. There may be some concern over the long-term availability of Youtube as an advertising channel if it's not making money, but do you need to think about that when you need to run an ad campaign right now? So I mean, I don't really believe all the spin either, but why tell anyone else otherwise? They don't certainly aren't out to raise funding either...

14 points by alex_c 17y ago

Fairly short post, the interesting bit: Myth 3: Traffic, growth, and uploads are bad for YouTube's bottom line. There's been a lot of speculation lately about how much it costs to run YouTube. With revenue estimates ranging from $120 million to $500 million, and costs on an equally large spectrum, it seems people can pick any number to fit any theory they have about our business. The truth is that all our infrastructure is built from scratch, which means models that use standard industry pricing are too high when it comes to bandwidth and similar costs. We are at a point where growth is definitely good for our bottom line, not bad.

amadiver 17y ago

With awesome hits like "Worlds Apart", "CHIC.TV Models", and "Spin Cycle", it won't be long before I ditch the old cable box for YouTube!* *Just kidding YouTube. I love you, and I know you'll get better stuff than old "Starsky and Hutch" episodes any day now.

pmorici 17y ago
YouTubes current obstacle to getting better content is going to be Hulu which a large number of major content providers have a monetary interest in.
- shrikant 17y ago
  YouTube can then look at taking content to international markets (as opposed to the U.S-only Hulu), which make up a fairly decent chunk of their viewership..

pj 17y ago

These myths are officially busted. No they aren't. Myths are busted with science and testing and experimentation. All this article did was contradict the myths. It did nothing to show the truth or falseness of the claims. If you are going to use something like the high quality show, Mythbusters, to model your contradictions, you should learn something about science from the show and actually prove whether or not your contradiction is true. This post didn't even come close to doing that.

jerf 17y ago
"It did nothing to show the truth or falseness of the claims." This is a primary source. If YouTube says that more traffic makes YouTube more money, nobody else is in a position to contradict them. I believe that lying about that in public could have legal repercussions.
- sunir 17y ago
  I'm with the parent poster. The myths are not busted until some hard facts come out. There are artful ways to say true things that make the reader jump to conclusions that they intend, but are false. For instance, by claiming that they are in a position that growth is good, they make us think they are making profit per view, but that isn't what they said. They could have any number of non-profitable reasons to like growth, like strategic market share or building a huge content database. Further, their refutation that advertisers are not afraid of YouTube may lead some readers to conclude that 70% of Ad Age Top 100 Marketers are paying YouTube directly, whereas many (most?) are not. They are either using YouTube as part of a social media campaign or advertising within some user content hosted on YouTube. It's not clear at all what's happening on their balance sheet, so speculation will continue. To what end, I don't know, but after "What is Twitter's business model?", guessing how much money YouTube burns a year is the Valley's favourite party game. ;)
  - jerf 17y ago
    "I'm with the parent poster. The myths are not busted until some hard facts come out." If you adopt this position, then even "hard facts" won't satisfy you, since, again, they'd have to come from YouTube, a source you don't trust. To be honest, that's the position I'm in; I don't trust them on this point, though I don't particularly care, either. I quite carefully didn't say that the claims are true in their most obvious form, merely that this is as good as it is going to get for now, and in particular, that there is no applicable "science and testing and experimentation" to be done.
    - sunir 17y ago

philwelch 17y ago

"The world premiere of Joy Luck Club director Wayne Wang's film, "The Princess of Nebraska," was viewed 165,000+ times during the first 48 hours -- the equivalent of landing the 15th spot on Hollywood box office charts." Yeah, if movie theaters were free and instead of you going to them, they came to you. If there's an "equivalent" to 165,000 tickets on an opening weekend of a Hollywood movie, it's probably closer to 165,000,000 page views. And that's in how much revenue is generated--for the production company, a film playing in an actual cinema would rake in more than a YouTube release of equivalent revenue, because your local movie theater makes almost nothing on ticket sales while YouTube probably makes a significant percentage of ad revenue. It's not like YouTube can give all ad revenue to content creators and sell overpriced popcorn and soda, after all.

TFrancis 17y ago
I think that the benefits of price (free) and availability (theater comes to you) are only likely to increase movies watched via YouTube. Once a movie does have 165M page views in an opening weekend (and one will), which will be a better measure of movie performance: the box office or Google?
- Retric 17y ago
  That depends on the movie, harry potter pulled in 139.7M over its Wednesday-to-Sunday debut period. I don't see free with advertising adding enough people to make close to that much money. However, a movie like clerks could be vary profitable online.
- philwelch 17y ago
  The best measure will always be revenue. Can YouTube generate better revenue for film producers than movie theaters? Not even television in its heyday could do that.

app 17y ago

How is bandwidth pricing built from scratch? You're transferring bits like everyone else. Sure you might be doing it CHEAPER, but they didn't rewrite the rules to peering. These myths aren't "busted," they just refuted reported numbers with "that ain't true!"

timdorr 17y ago
It very well may be different, since Google bought a ton of dark fiber a few years back. They may be running their own network over to whomever wants to peer with them (Read: everyone!). They can get away with free bandwidth because a lot of providers will let them connect for free if they're able to bring the bits directly to their network, not via transit with some other provider that's costing both sides money.

jgamman 17y ago

most of the stuff doesn't play for me in NZ so we're still a ways from upload-once-play-anywhere

100k 17y ago

I like this one: "Myth 5: YouTube is only monetizing 3-5% of the site. This oft-cited statistic is old and wrong, and continues to raise much speculation. In our view, the percentage is far less important than the total number of monetized views..." They say it's old and wrong, but don't give the real percent, and spin in terms of "total number of monetized views". Based on this answer I'd guess the percent of monetized views isn't much higher than the "myth", otherwise they'd have a better spin.

jacquesm 17y ago

I've been moderated straight into the ground for this previously, but google does not nearly pay for bandwidth what you'd be quoted if you went shopping for bandwidth. At a 5G commit you're down to $4/Mbit, over that it gets quite a bit cheaper still. Youtube is consuming thousands of times that quantity. (the 5Gbit/sec will serve out 10K streams @600Kbps) Another good trick they pull is to cache videos that are requested frequently in or very near to the colocation offered by the larger ISPs.

philjr 17y ago
Yup, they peer at local internet exchanges (e.g. LINX in London - www.linx.net / INEX in Ireland - www.inex.ie), cache locally where they can (uk.youtube.com / ie.youtube.com are what you get redirected to periodically over here) and I would imagine that this drastically reduces their "internet" bandwidth consumption.
- andyking 17y ago
  I use the O2 ISP and I often notice that if I stream a popular video from YouTube (or anything from the BBC iPlayer) my browser will tell me it's "transferring data from" an IP address which is in my ISP's own range. It must be cheaper for them to cache popular content within the ISP system and send it out that way.

jimmybot 17y ago

What I'm curious about is why they bothered to do the myth busting. What's the motivation to do this? If everyone thinks even the mighty Google can't get YouTube to break even, then no one else will try to compete with them in video, and when they get around to figuring out how to make money, they will own the entire whole short video market. Meanwhile, Google overall remains a desirable place to work in the eyes of many engineers. And Youtube advertisers care about how effective their ads are relative to what it's costing them, not whether or not Google is making money from Youtube. There may be some concern over the long-term availability of Youtube as an advertising channel if it's not making money, but do you need to think about that when you need to run an ad campaign right now? So I mean, I don't really believe all the spin either, but why tell anyone else otherwise? They don't certainly aren't out to raise funding either...