About About Contact Contact

Is Facebook facing facts..?

Almost since its inception I’ve been tracking Facebook behaviour, initially for business reasons and latterly in trying to understand if there are actually any limits on the amount of personal information that people will give away in exchange for apparently free access to digital services.  So far I haven’t seen any constraints because it appears that Facebook users naively believe the network is all about, and of course limited to, their friends.  (We won’t define “friends” here).  In cold reality Facebook isn’t a social system, it is an advertising network which produces 88% of the company’s income.  This is just like Google which isn’t merely a search engine, it’s a highly efficient advertising network which gets over 90% of it income from advertising.  To an old Internet geek like me, nothing is what it appears to be in the realm of Web services which people frequently use like those simple and now apparently essential communication services like email, search, mapping and digitally talking to friends.  There is, as I’ve said before, no such thing as a free lunch - if you are not paying money for a service then you are the lunch.  Facebook is now beloved by advertisers and their advertising agencies because it apparently meets the ideal requirements of providing simple measurable metrics, whilst also providing global reach (a super-sized audience) as well as frequency (a high proportion of people using the network every day).  As a consequence Facebook’s advertising revenue is on a roll, up 57% on last year, and it is predicted to come in somewhere north of $6 billion at the year-end (2013).  Interestingly, I reckon that there is scope for Facebook, or somebody else with a global internet reach, to provide a social network service costing around $25 year.  Facebook is raking in a healthy profit with average revenue per user of $1.74 per month.  Think about it.  Would you and your family and friends be willing to pay this sort of price for a private, secure and advertising-free social network?  I would and, who knows, this could possibly be a way for Microsoft, usually so tardy at seeing and utilising trends, to finally get an edge by pioneering an advertising-free, genuinely social network. But back to Facebook.  The company is dependent on gigantic numbers of people using their network regularly and frequently.  The larger the user numbers, the higher Facebook’s advertising revenue.  But to me Facebook’s quoted user numbers have always been dubious.  Sometime before the publicity about Facebook shares being publicly traded, I was questioning how they actually measured their visitors.  This was because, long before then, in my own qualitative research I had come across “Facebook fatigue,” although Facebook’s own numbers seemed to deny that this was happening.  I wasn’t alone; in the UK early in 2008 Nielsen online data was also showing that enthusiasm for Facebook was waning.  To be fair, trying to accurately measure the number of people using Facebook has always been difficult – even for Facebook.  However, in my mind there is no question that in the last three years the number and frequency of people using Facebook in the developed world, especially young teenagers, is declining.  Now that is even beginning to show up in Facebook’s own data.  As the company explains: “…models suggest that usage by U.S. teens overall continues to be stable, but that there has recently been a decline in DAU (Daily Active Users) among younger U.S. teens.”  Advertisers are usually behind the curve on trends, and I am not casting aspersions on Facebook’s business model, just their figures.  Is it not an ironic paradox that in the developed world as increasing amounts of money is being spent on advertising products and services on Facebook, there is a simultaneous decline in the number of younger users, the very audience that advertisers especially want to reach?  Unfortunately it appears that Facebook doesn’t have much control over this situation.  If you run any sort of business with a total reliance on “big data” and simplistic metrics, this state of affairs illustrates some key limitations.  But this decline is not reflected in the Facebook data used on the chart above, far from it.  Facebook is stating firmly that any decline, especially amongst teenagers, is severely over-stated.  Really?  I’m not so sure.  In the Internet world, what grows quickly can decline equally quickly, and the young social herd of previously heavy users are moving on.  Fledgling teenagers don’t want to be associated publicly with their Mums, Dads, or Grandparents, do they? The data on the chart is taken from Facebook’s latest quarterly Securities and Exchange Commission (SEC) 10K filing, and what I’ve done is chart a comparison between daily and monthly users in North America.  America was where Facebook started and, if there is ever going to be a My Space or Friendster like decline, it is going to be there.  The company defines an active user:  “…as a registered Facebook user who logged in and visited Facebook through our website or a mobile device, or took an action to share content or activity with his or her Facebook friends or connections via a third-party website or app that is integrated with Facebook…”  Do note that users don’t actually have to have visited Facebook within one month to be counted, just possessing an app with Facebook integration that automatically checks the service, also qualifies.  With that in mind, you can appreciate that Facebook‘s daily figures will continue to look good because a high proportion of Americans own smartphones, (currently 62% according to comScore) and that number is likely to keep rising for some time.  Facebook itself admits that user growth and advertising revenue will primarily be driven by people using mobile devices, and many investors choose to believe that the company has finally cracked the secret of mobile advertising.  So there it is: regardless of what I say, Facebook advertising revenue is highly likely to continue to increase, but I have an instinctive feeling that advertising and small screens do not, and never will, mix well. There are two difficulties with Facebook’s current optimism:  Firstly, interruptive advertising on small, intimate devices doesn’t produce positive brand equity for advertisers.  People of all ages become utterly task-focussed when they use a small screen and easily get irritated and ultimately annoyed by any disturbance, so any brand equity produced is likely to be more negative than positive.   Secondly, the metrics produced by Facebook have what in statistical terms is called a wide degree of disturbance in the data.  This is particularly true for mobile devices with apps which continually and automatically check Facebook every day, classifying the person with that phone in Facebook terms, as a daily active user.  Put politely and simply: this means there is a wide difference between reality and the measured data.  I admit that, to some extent, Facebook acknowledges a degree of this in their 10K SEC filing, (this is after all a legal duty), but they underplay the actual magnitude.  Facebook has a history, before and during its public offering, of providing wildly different figures about mobile device usage.  What at first sight appear to be accurate figures about Facebook usage can in reality be very misleading.  The number of people using, or not using, Facebook via a smartphone with a Facebook-enabled-app is growing, so the figure for so-called daily active users is likely to continue to be inaccurate.  On even closer inspection there are a lot of estimates used in Facebook’s production of usage metrics.  Facebook estimates that the number of people in 2013 who have duplicate Facebook accounts is between 4.3% and 7.9% worldwide.  Then there are the non-human accounts for pets or businesses - Facebook’s figure for these is between 0.8% and 2.1% globally.  And remember Facebook’s vaguely classified “undesirable accounts” that are estimated to be somewhere between 0.4% and 1.2% worldwide.  Adding up these discrepancies means the margin of error in user activity, (using Facebook’s own figures, remember) is between 5.5% and 11.2%.  This could easily be masking a real decline in usage.  And we mustn’t forget all the children under 13 years old who aren’t supposed to be using Facebook that has been measured by comScore at 1.5%.  To these numbers we probably have to add some sort of figure for the “unknown unknowns,” so precisely described by Mr Rumsfeld, yet laughed about at the time For example, Facebook has admitted to over-counting American activity in 2012 by 3% as there were problems with their sorting of location data.  I’m afraid I’m also suspicious about exactly how exact Facebook’s robot or multiple account detection really is.  By way of proof I’ve taken the liberty of including the going rate for brands to buy Facebook “Likes” on the chart.  If you buy in quantities of at least 100,000, a price of less than one cent per “Like,” implies that there must be many multiple “ghost” Facebook accounts which can be triggered by automated script activity, that Facebook has yet to detect.  When you add up all these factors there could easily be a 10% to 25% inaccuracy in Facebook’s own metrics.  And I reckon it’s probably much higher in the daily active data because of the growing number of automatic log-ins from mobile devices. In an attempt to check out what is really happening on Facebook I’ve turned to an old, tried and tested, market research technique.  What does the data show when you ask a reasonable number of people about their Facebook behaviour?  GlobalWebIndex does exactly that and their latest figures show a 16% decline in American teenage Facebook activity from the fourth quarter 2012 to the third quarter in 2013, with some other countries showing a much steeper decline.  I think there are two fundamental reasons for this – the first is, as I’ve already mentioned, that mobile devices with their small screens are only good for simple tasks and that doesn’t include trawling through Facebook timelines.  The second and more fundamental reason is that any kind of social networking eats time, and people are becoming more aware of just how much of their valuable time Facebook activity is taking.  When Pew surveyed Americans on their social networking activity they found that 61% had taken Facebook “vacations” of varying lengths, with 20% doing so permanently.  Pew went on to ask people who had taken any sort of break from Facebook, why they had done so?  The largest group (21%) said that their “Facebook vacation” was a result of being too busy, or not having time to spend on the site.  And that is something that Facebook, and its ever increasing amount of advertising, will never be able to fix.  It also goes some way to explain why the young, trendy social herd is moving on to social services like SnapChat.  For those of you who haven’t yet come across it SnapChat is a fast and easy to use application which creates pictures and stories that can be quickly shared with a smaller, more personal set of friends.  SnapChat is far better designed for use in the mobile world than Facebook which probably explains why Facebook recently offered to buy SnapChat for $3 billion or more in cash.  It may also be the reason why SnapChat’s CEO Evan Speigel turned the offer down. December 2013  
Click here to download the PowerPoint chart: Click here to download the PowerPoint chart:
...with analysis & insight...
Home Home Archive: Free PowerPoint download Free PowerPoint download
Click image to enlarge
Click here to download the PowerPoint chart: Click here to download the PowerPoint chart: Click here to download the PowerPoint chart: Click here to download the PowerPoint chart: Click here to download the PowerPoint chart: Click here to download the PowerPoint chart: Click to return to page
Archive Archive
About
Contact
Home
Click to return to page

Is Facebook facing facts..?

Almost since its inception I’ve been tracking Facebook behaviour, initially for business reasons and latterly in trying to understand if there are actually any limits on the amount of personal information that people will give away in exchange for apparently free access to digital services.  So far I haven’t seen any constraints because it appears that Facebook users naively believe the network is all about, and of course limited to, their friends.  (We won’t define “friends” here).  In cold reality Facebook isn’t a social system, it is an advertising network which produces 88% of the company’s income.  This is just like Google which isn’t merely a search engine, it’s a highly efficient advertising network which gets over 90% of it income from advertising.  To an old Internet geek like me, nothing is what it appears to be in the realm of Web services which people frequently use like those simple and now apparently essential communication services like email, search, mapping and digitally talking to friends.  There is, as I’ve said before, no such thing as a free lunch - if you are not paying money for a service then you are the lunch.  Facebook is now beloved by advertisers and their advertising agencies because it apparently meets the ideal requirements of providing simple measurable metrics, whilst also providing global reach (a super-sized audience) as well as frequency (a high proportion of people using the network every day).  As a consequence Facebook’s advertising revenue is on a roll, up 57% on last year, and it is predicted to come in somewhere north of $6 billion at the year-end (2013).  Interestingly, I reckon that there is scope for Facebook, or somebody else with a global internet reach, to provide a social network service costing around $25 year.  Facebook is raking in a healthy profit with average revenue per user of $1.74 per month.  Think about it.  Would you and your family and friends be willing to pay this sort of price for a private, secure and advertising-free social network?  I would and, who knows, this could possibly be a way for Microsoft, usually so tardy at seeing and utilising trends, to finally get an edge by pioneering an advertising- free, genuinely social network. But back to Facebook.  The company is dependent on gigantic numbers of people using their network regularly and frequently.  The larger the user numbers, the higher Facebook’s advertising revenue.  But to me Facebook’s quoted user numbers have always been dubious.  Sometime before the publicity about Facebook shares being publicly traded, I was questioning how they actually measured their visitors This was because, long before then, in my own qualitative research I had come across “Facebook fatigue,” although Facebook’s own numbers seemed to deny that this was happening.  I wasn’t alone; in the UK early in 2008 Nielsen online data was also showing that enthusiasm for Facebook was waning.  To be fair, trying to accurately measure the number of people using Facebook has always been difficult – even for Facebook.  However, in my mind there is no question that in the last three years the number and frequency of people using Facebook in the developed world, especially young teenagers, is declining.  Now that is even beginning to show up in Facebook’s own data.  As the company explains: “…models suggest that usage by U.S. teens overall continues to be stable, but that there has recently been a decline in DAU (Daily Active Users) among younger U.S. teens.”  Advertisers are usually behind the curve on trends, and I am not casting aspersions on Facebook’s business model, just their figures.  Is it not an ironic paradox that in the developed world as increasing amounts of money is being spent on advertising products and services on Facebook, there is a simultaneous decline in the number of younger users, the very audience that advertisers especially want to reach?  Unfortunately it appears that Facebook doesn’t have much control over this situation.  If you run any sort of business with a total reliance on “big data” and simplistic metrics, this state of affairs illustrates some key limitations.  But this decline is not reflected in the Facebook data used on the chart above, far from it.  Facebook is stating firmly that any decline, especially amongst teenagers, is severely over-stated.  Really?  I’m not so sure.  In the Internet world, what grows quickly can decline equally quickly, and the young social herd of previously heavy users are moving on.  Fledgling teenagers don’t want to be associated publicly with their Mums, Dads, or Grandparents, do they? The data on the chart is taken from Facebook’s latest quarterly Securities and Exchange Commission (SEC) 10K filing, and what I’ve done is chart a comparison between daily and monthly users in North America.  America was where Facebook started and, if there is ever going to be a My Space or Friendster like decline, it is going to be there.  The company defines an active user:  “…as a registered Facebook user who logged in and visited Facebook through our website or a mobile device, or took an action to share content or activity with his or her Facebook friends or connections via a third-party website or app that is integrated with Facebook…”  Do note that users don’t actually have to have visited Facebook within one month to be counted, just possessing an app with Facebook integration that automatically checks the service, also qualifies.  With that in mind, you can appreciate that Facebook‘s daily figures will continue to look good because a high proportion of Americans own smartphones, (currently 62% according to comScore) and that number is likely to keep rising for some time.  Facebook itself admits that user growth and advertising revenue will primarily be driven by people using mobile devices, and many investors choose to believe that the company has finally cracked the secret of mobile advertising.  So there it is: regardless of what I say, Facebook advertising revenue is highly likely to continue to increase, but I have an instinctive feeling that advertising and small screens do not, and never will, mix well. There are two difficulties with Facebook’s current optimism:  Firstly, interruptive advertising on small, intimate devices doesn’t produce positive brand equity for advertisers.  People of all ages become utterly task-focussed when they use a small screen and easily get irritated and ultimately annoyed by any disturbance, so any brand equity produced is likely to be more negative than positive.   Secondly, the metrics produced by Facebook have what in statistical terms is called a wide degree of disturbance in the data.  This is particularly true for mobile devices with apps which continually and automatically check Facebook every day, classifying the person with that phone in Facebook terms, as a daily active user.  Put politely and simply: this means there is a wide difference between reality and the measured data.  I admit that, to some extent, Facebook acknowledges a degree of this in their 10K SEC filing, (this is after all a legal duty), but they underplay the actual magnitude.  Facebook has a history, before and during its public offering, of providing wildly different figures about mobile device usage.  What at first sight appear to be accurate figures about Facebook usage can in reality be very misleading.  The number of people using, or not using, Facebook via a smartphone with a Facebook-enabled-app is growing, so the figure for so-called daily active users is likely to continue to be inaccurate.  On even closer inspection there are a lot of estimates used in Facebook’s production of usage metrics.  Facebook estimates that the number of people in 2013 who have duplicate Facebook accounts is between 4.3% and 7.9% worldwide.  Then there are the non-human accounts for pets or businesses - Facebook’s figure for these is between 0.8% and 2.1% globally.  And remember Facebook’s vaguely classified “undesirable accounts” that are estimated to be somewhere between 0.4% and 1.2% worldwide.  Adding up these discrepancies means the margin of error in user activity, (using Facebook’s own figures, remember) is between 5.5% and 11.2%.  This could easily be masking a real decline in usage.  And we mustn’t forget all the children under 13 years old who aren’t supposed to be using Facebook that has been measured by comScore at 1.5%.  To these numbers we probably have to add some sort of figure for the “unknown unknowns,” so precisely described by Mr Rumsfeld, yet laughed about at the time For example, Facebook has admitted to over-counting American activity in 2012 by 3% as there were problems with their sorting of location data.  I’m afraid I’m also suspicious about exactly how exact Facebook’s robot or multiple account detection really is.  By way of proof I’ve taken the liberty of including the going rate for brands to buy Facebook “Likes” on the chart.  If you buy in quantities of at least 100,000, a price of less than one cent per “Like,” implies that there must be many multiple “ghost” Facebook accounts which can be triggered by automated script activity, that Facebook has yet to detect.  When you add up all these factors there could easily be a 10% to 25% inaccuracy in Facebook’s own metrics.  And I reckon it’s probably much higher in the daily active data because of the growing number of automatic log- ins from mobile devices. In an attempt to check out what is really happening on Facebook I’ve turned to an old, tried and tested, market research technique.  What does the data show when you ask a reasonable number of people about their Facebook behaviour?  GlobalWebIndex does exactly that and their latest figures show a 16% decline in American teenage Facebook activity from the fourth quarter 2012 to the third quarter in 2013, with some other countries showing a much steeper decline.  I think there are two fundamental reasons for this – the first is, as I’ve already mentioned, that mobile devices with their small screens are only good for simple tasks and that doesn’t include trawling through Facebook timelines.  The second and more fundamental reason is that any kind of social networking eats time, and people are becoming more aware of just how much of their valuable time Facebook activity is taking.  When Pew surveyed Americans on their social networking activity they found that 61% had taken Facebook “vacations” of varying lengths, with 20% doing so permanently.  Pew went on to ask people who had taken any sort of break from Facebook, why they had done so?  The largest group (21%) said that their “Facebook vacation” was a result of being too busy, or not having time to spend on the site.  And that is something that Facebook, and its ever increasing amount of advertising, will never be able to fix.  It also goes some way to explain why the young, trendy social herd is moving on to social services like SnapChat.  For those of you who haven’t yet come across it SnapChat is a fast and easy to use application which creates pictures and stories that can be quickly shared with a smaller, more personal set of friends.  SnapChat is far better designed for use in the mobile world than Facebook which probably explains why Facebook recently offered to buy SnapChat for $3 billion or more in cash.  It may also be the reason why SnapChat’s CEO Evan Speigel turned the offer down. December 2013  
Click here to download the PowerPoint chart: Click here to download the PowerPoint chart: