8 ms·

Facebook Advertising is Fool's Gold

80 points by msomers 14y ago

helipad 14y ago
I wonder what would happen if Facebook ran a scheme like AdSense. Advertising on the Facebook site itself seems not to be working, but I could foresee a situation where website owners allow Facebook to display ads on their site for a fee per click just like AdSense. Facebook already has a javascript presence on many, many websites already.
- shinchan 14y ago
  Wow. Its like you just read my mind! I was about to say the same thing. Indeed, especially since there are a lots of apps out there implementing Facebook login, especially mobile apps, games and apps alike, could this be a solution to their mobile strategy? (this is not a rhetorical question.) Whats stopping them from implementing this?
- AznHisoka 14y ago
  Adsense makes up a minority of Google's ad profits. The billion dollar business is in Adwords, not Adsense.
  - helipad 14y ago
    Indeed - I've no idea what the revenues would be - was just an idle thought. Right now the problem is getting people to believe in Facebook's revenue to get some confidence back in the market. Not saying it's is a silver bullet by any means.
  - PelCasandra 14y ago
    Minority? It almost half of all Google ad profits.
    - mda 14y ago
      It is more like Adwords: 70% Adsense: 30% http://investor.google.com/earnings/2012/Q1_google_earnings.html http://investor.google.com/earnings/2012/Q1_google_earnings....
  - earl 14y ago
    sadly, no. Adsense is a $10B/year business. Even ex tac of 30%, that's still a big business. From wikipedia [1]: In Q1 2011, Google earned US$2.43 billion ($9.71 billion annualized), or 28% of total revenue, through Google AdSense [1] http://en.wikipedia.org/wiki/AdSense http://en.wikipedia.org/wiki/AdSense
- brandnewlow 14y ago
  I used to think this as well, but there's a problem. What if FB ads are not as good a monetization solution for publishers as Google ads? This seems quite possible. Facebook advertising is much much cheaper than Google advertising on a per-click and per-impression basis. With Google ads, advertisers are used to paying as much as $20/click in some areas. I've never heard of a FB advertiser paying more than $2/click. Likewise, CPMs on FB are in the single digit pennies quite often. On the web, banner ads are around $0.75 CPMs for decent sites. Facebook might have looked at this and might believe that if they launched a network like this, the payouts just wouldn't be able to compete with what Google's able to give publishers.
- mootothemax 14y ago
  I wonder what would happen if Facebook ran a scheme like AdSense. Maybe that is/was one of the aims behind the Like button? Facebook could use that connection to tailor the adverts displayed to the user (along with the site's content!), provided the user is logged into Facebook at the time. Admittedly, I think doing so would kick up a bit of a stink ;)
- mkr-hn 14y ago
  I've never had good results from ticking the box for Google's display network in AdWords campaigns. The culprit is poor implementation by webmasters, not the amount or detail of data available. I don't see why a Facebook display network would do any better.
- alain94040 14y ago
  They'll do it as soon as they are confident that they can pay websites more per click than AdSense. It's a winner-take-all market. But AdSense is very vulnerable: intent (the search string), which is the one reason Google advertising works, is absent from AdSense. As an example, I once wrote a blog post with a title like "startup financing: take the elevator, not the stairs". Google placed ads on that post for Thyssen elevators. It had just no clue that the topic was startups, not actual elevators. Not impressed.
  - earl 14y ago
    Google uses search activity from the past several hours to help match ads in adsense for content[1]. Hiding those from other ad networks is, imo, one of the best reasons they should be sued for antitrust. We've recently started to expand the use of the query words in referral URLs to a few hours so we can so we can continue to deliver more relevant ads. The technical way that we're doing this is by associating the relevant query words in the referral URL with the existing advertising cookie on the user's browser. After a short period of time (a few hours) the query words are no longer used for the purposes of matching ads. [1] http://adsense.blogspot.com/2010/02/better-contextual-matching.html http://adsense.blogspot.com/2010/02/better-contextual-matchi...
    - maxerickson 14y ago
      That article doesn't discuss the details of the implementation, but it could very well be the Adsense js that is tagging the cookie with the referral data. Any ad network can implement that.
      - earl 14y ago
        Perhaps I wasn't clear. Google is hiding referer data, particularly the query string, for a growing number of searchers. Today it's anyone logged into google, or about 60% of traffic to my blog. Tomorrow, they may well do it for anyone. Right now for the logged in users, and potentially in the future for anyone, the only ad network that can still see the search terms in the referer string is adsense, because Google search can pass them to adsense internally. No other ad network can see them. This is obviously incredibly valuable for adsense. And it should trigger an antitrust violation because they're using their monopoly / near monopoly in search to juice their offsite context and display ad networks.
mikecane 14y ago
So basically this is saying that Facebook needs a search engine to be effective at advertising. Is DuckDuckGo up for sale?
- suking 14y ago
  LoL - Fb could build that pos in 2 hours.
- 3lit3H4ck3r 14y ago
  Why would they buy DDG when they (FB) already have Bing? Oh...yeh. Is DDG for sale? :)
amitvaria 14y ago
But consider Facebook's "friction-less sharing". Facebook now knows you just listened to a song, read and article, took a trip, had a baby, got married.... Now think of the targeting you can get there.
- antr 14y ago
  All that sounds very nice, but that is what FB has all been about. This kind of non-quantitiave arguments won't validate a $100bn valuation.
- sp332 14y ago
  The targeting goes all the way down. Gabriel Weinberg (who runs DuckDuckGo) targeted an ad to one person! http://www.gabrielweinberg.com/blog/2010/05/a-fb-ad-targeted-at-one-person-my-wife.html http://www.gabrielweinberg.com/blog/2010/05/a-fb-ad-targeted...
- tomelders 14y ago
  You sell cars. I listen to Donovan, The Beatles, The Stone Roses, Mozart, Au, REM, Dj Shadow, Handsome Boy Modelling School, Chopin, Ludovico Einaudi and Daft Punk (to name but a few). I've liked pages for five star greek hotels, the burger place on the corner, the guy who sells awesome falafel's on Hoxton market. You know I saw this video of a talking dog that I thought was totally awesome. Just this month, I read an article on the Greek debt crisis in the economist, an article on David Beckham in the Daily mail website, I shared a wikipedia page about Rommel, I liked an article about continuous integration with node.js and Jenkins. Now sell me a car. I actually like Volkswagens and Volvos, but you won't find that anywhere in my Facebook data.
  - cageface 14y ago
    I actually like Volkswagens and Volvos, but you won't find that anywhere in my Facebook data. I'd bet there's actually a pretty high correlation between many of the things you just listed and a preference for VWs and Volvos. It doesn't really take that many data points to start filling in the blanks. But I think this kind of targeting is going to be more useful in selling you a product you don't yet know about but are likely to want than it is a car that you are almost certainly already aware of.
tomelders 14y ago
For me, this is the strongest case yet for niche social networks being the future of social media. FB is too broad and demographics tell you little about the people you are speaking too unless you can inject some sort of context. If (via a Distributed Social Networking Protocol) people were subscribed to several social networks focused on their hobbies and interests, then I believe Social Media Advertising would be more viable, and more useful to the audience. I also think that when you look past all the fanfare and hyperbole surrounding social networking, what you're left with is really sophisticated bulletin boards and forums. I can see that from a marketing point of view, having one point of access to that many people might seem like a dream come true, but the numbers are suggesting that's not the case at all. Facebook is not the answer to an online advertisers dream. I don't think it ever will be.
- ajdecon 14y ago
  For me, this is the strongest case yet for niche social networks being the future of social media. FB is too broad and demographics tell you little about the people you are speaking too unless you can inject some sort of context. If (via a Distributed Social Networking Protocol) people were subscribed to several social networks focused on their hobbies and interests, then I believe Social Media Advertising would be more viable, and more useful to the audience. You don't need DSNP: niche social networks already exist. Ravelry and GoodReads, as examples, seem to be making a decent go of it. Would love to know how their advertising does.
  - tomelders 14y ago
    I think DSNP is essential. Imagine if you could only email someone with a GMail account if you had a Gmail account. Imagine if you could only check your Hotmail emails on the Hotmail website or through one of the official Hotmail Apps. That's how Social networking works right now. API's won't cut the mustard. We need a DSNP.
antidoh 14y ago
I am so glad that email didn't depend on advertising to eventually support itself "sometime in the future." Email is useful to me as email, and useful to my providers as a billed service. Yes, they're different things.
cletus 14y ago
I agree with this and have posted numerous comments here to that effect, particularly on the value of intent (with respect to the effectiveness of online advertising). Th attitude of "eyeballs not revenue" and finding that repeatable, scalable formula are almost a religion on HN and in startups. It's generally a strategy I approve of. Even dropping 20% since the IPO (currently trading at just over $30 as I write this; and bear in mind that even Groupon--at least initially--enjoyed a nice IPO bounce), the company is trading at a huge P/E ratio. I get why this is: it's speculative. FB is still viewed as a growth company and the speculators feel that there is huge unrealised monetization potential. I remain a skeptic regarding the value of "social" in advertising. The OP is right: all this data just means--maybe--a slightly higher CTR, at which point Facebook is just another display ad network and that doesn't justify their valuation. As an aside, IMHO Twitter is in this same "put up or shut up" boat. I don't believe Facebook is doomed (IMHO Twitter is). I just believe the value of the data silo they have is both overstated and transitory (at some point--one way or the other--Facebook won't be the gatekeeper to your profile and social graph). Facebook has done a lot of things right as a business (the Like button being foremost among those IMHO). Personally I believe their biggest mistake was spurning Apple: Apple wanted to use Facebook for their Ping boondoggle. Disclaimer: I work for Google in display advertising.
- RegEx 14y ago
  I'm confused over the significance of the P/E ratio. It seemed like a big deal when I was reading these stories about the FB IPO, but then I got to look at other ratios in tech. Amazon's is much higher than FB's 93.72 at 175.23 [0] [0]: http://www.google.com/finance?q=NASDAQ:AMZN&authuser=0 http://www.google.com/finance?q=NASDAQ:AMZN&authuser=0
  - tatsuke95 14y ago
    Oh, don't worry, the bears consider it out of whack and a good short as well. I'm not exactly sure what the deal is with Amazon, but their p/e was much more realistic in the not-so-distant past: http://ycharts.com/companies/AMZN/pe_ratio http://ycharts.com/companies/AMZN/pe_ratio
  - 14y ago
fakhrazeyev 14y ago
It does seem obvious. There is just a wall of hype hanging. And factor in the fact that it was reported that approximately 40 % of all social network accounts are created by spammers.
pbharrin 14y ago
Google's ads are better because you have intention when you are searching. That's not true when you are checking into Facebook to see what other people are doing. Are there other times when people use Facebook and they have some intent? Planning an event, or asking a question, those come to mind.
glesica 14y ago
Seems reasonable to me. Facebook is gathering what amounts to a static picture of each user then selling ads based on the pictures. There really isn't a time component in the data (though perhaps they can add one now that the service has existed for many years). On the other hand, Google has more or less assembled a bunch of archetypes based on behavior (people who search for X later purchase Y, etc.). Then they match queries to these archetypes. The time component is front-and-center here because the archetype match was prompted by a user action. A purchase is really a time-related event that transforms the user in some sense. By purchasing something you're changing "who" you are from the perspective of an advertiser (e.g. a Chevy driver could become an Audi driver). So Facebook has a great picture of "who" each user is at a the moment, but no real way of knowing "who" each user would like to become or when. Hence, perhaps, some of their difficulty in selling ads. Or not.
kevinlu310 14y ago
Facebook's advertising model is a simple classical displaying advertising model. It need be disrupted seriously. Nowadays, the most popular marketing model is displaying advertising model, which is actually an attention-driven marketing model. However, no matter how good they are context-awared and personalized, they are always guesswork, which can hardly achieve high relevancy. It is extremely difficult to guess users’ actual intention for most cases, even you may have a lot of information of them. For example, it is super difficult to guess what an individual is exactly looking to buy when he walks into a Trader Joe’s, no matter how much information you obtained on him, like his income, profession, age, education and etc. Users’ intentions are always dynamically changing from time to time. There are no general equations that can perfectly predict people’s intentions. As a result, no matter how perfectly a system can attract attentions, attention-driven marketing can never achieve high efficiency and accuracy. As a matter of fact, nowadays advertisement industry is the biggest bubble of the world. Every year, trillions of dollars completely wasted by irrelevant advertising unnecessarily. What is even worse is, the user experience is badly hurt by ads and spams. Such marketing system can be called as attention-driven marketing system. Everything is done to attract attentions. However, only a few attentions are really attracted, but majority of the attentions may not be finally turned into intentions. As a result, majority of the marketing costs are totally wasted. The future marketing systems will be built on intentions. Intention-driven commerce systems need let every user to express his/her intentions freely and help him/her achieve his/her intentions in highly optimized way by socialization and crowd-sourcing. The future world needs to be a world without displaying ads, but only pure relevant information that matches users' intentions dynamically and real-timely. This is the very objective of my current startup project as well.
- alain94040 14y ago
  How do you promote events in your non-bubble view of the world? An event can be in person (imagine TechCrunch Disrupt for instance), or a new TV series launching... I can imagine very well how I would target people who like the series 24 if I'm launching a new thriller. Or readers of Hacker News for a tech conference. Etc...
  - kevinlu310 14y ago
mag487 14y ago
Not trying to defend the viability of FB ads here, but this analysis seems a little shallow. Falling within a certain demographic may not "make" you click an ad, but the same could be said of performing a search on Google. If circumstance Y makes one want to buy a product, then as long as falling within demographic X is correlated with being in Y, targeting X could be a good idea. It is (obviously) the reason why toy commercials air during children's shows and not during basketball games.
ma2rten 14y ago
You are mistaken to think that the search-engine show-people-what-they-were-searching-for-anyway kind of advertisement is the only kind of advertisement that can work on the internet. I used to work for Hyves, which was until a year ago or something the biggest social network in the Netherlands. Hyves lost to Facebook in terms of number of users, but something that they did do right was modernization. Hyves had an in-house sales team and studio, which sold custom viral campaigns to cooperate advertisers. These campaigns could usually go together with a TV campaign and/or other media. For some products search engine advertisement is the right kind of advertisement (parrot secrets, maybe cars, I don't know), for others not such much (food stuffs for example).
CookWithMe 14y ago
I agree that Google can target alot better IF you have a product that solves a problem people search for. IF. > What researchers discovered was that people bought milkshakes as a breakfast replacement because it was entertaining during a long, boring commute, and would keep them full until lunch. How do you advertise for them on Google? If they search for "breakfast replacement for long, boring commute"? For anyone who can match their product to search queries, Google is the way to go, agreed. But not all products can, because people are not actively looking for solving a problem/need that your product solves. If I look at my current facebook ads, these are not things I actively search for, e.g. drinks, shavers, parties, facebook games, ... If I had to spend marketing $ on these, I'd probably choose facebook ads over AdWords as well.
redwood 14y ago
There are two primary types of ads 1) ones that give you what you want 2) ones that drive demand Google dominates (1) and I assume Facebook is trying to dominate (2) which includes a lot more display advertising (and TV, and billboards). Facebook probably thinks a lot of demand-driving advertising can be accomplished through viral online marketing through FB platforms and they are likely trying to exploit these channels with services for paying customers. Imagine paying Facebook to simulate something being trendy among people's trend-setting friends. It'll be a fine line but this must be what they hope to do. Agree the $100B is too high, but certainly they're easily a profitable company, long-term.
- billybob 14y ago
  > 1) ones that give you what you want > 2) ones that drive demand A big part of Google's success is that #1 is much easier to prove than #2. "People bought our product more because they had become familiar with it over the last month via newspaper/radio/Facebook ads" is a hard-to-prove claim. "We got X clicks from Google ads leading directly to Y purchases" is much easier.
kelvin0 14y ago
Ever since FB has gone public, I've noticed so much 'negative' stories/comments, and I really wonder why these are only coming out now... FB has been around for a long time and nothing has changed in its marketing strategies since it seems. Personally, I don't have a FB account, or own any stock but this targeted 'bashing' seems useless and has the smell of targeted anti-FB propaganda. Why don't we focus on the real issue being people rights to privacy and the shady practices of FB? More focus on the real issues please, do not get distracted by the 'anti-hype' of the week.
- tatsuke95 14y ago
  >"Why don't we focus on the real issue being people rights to privacy and the shady practices of FB?" These things are all connected. The public is being sold the idea that this is a $100BB company, one of the biggest in the world. Yet when some of us look at it (yourself, even) we see a great tool that doesn't justify the price as a business: as cletus says at the top of the thread, Facebook may just be another ad network that depends on its massive, engaged user base. Yet FB has problems with how it actually treats those users, through privacy issues and shady practices..
  - kelvin0 14y ago
    I don't think that anyone who did their homework would have bought FB stock as a 'sure thing for long time investment'. FB didn't fool anyone who wasn't too greedy for their own good. Obviously I'm not saying FB is not a hugely successful product with an immense user base, but obviously the fact that this IPO is not 'going as predicted' by it's underwriters was clear to me waaayyy before the stock hit the markets ...
alain94040 14y ago
I call survivor bias. Historically, the only player in online advertising is Google, which is 100%-focused on intent-based advertising. So the article, logically concludes that intent is the only way to advertise (despite decades of TV and radio ads). Some products do great with intent-based advertising. Other products (that you didn't even know you needed) need other forms of advertising.
colinshark 14y ago
This argument continues to come up, but I don't buy it- At least not without some data. My personal anecdote: Facebook ads are the only online Ads I click with any regularity. They are targeted to my interests, such as programming. They also hit me at a time when I am mostly bored and clicking around on anything that interests me. When I find myself on Google, I am on a mission for very specific information. Even if I am shopping, I do a research phase and a purchasing phase. Ads in the research phase don't woo me. I google straight to trusted sources. Once I'm in the purchase phase (shopping for best price), I am not wooed either because the ads are too broad. E.G, I want the D3100, not just any camera in general. I completely understand Google's strategy of "intent" based ads. I just think Facebook's user data is even stronger.
rajpaul 14y ago
If I create high quality, high traffic Facebook content, do I get any money from advertising?
torstesu 14y ago
Cached version for those getting the 509: http://tinyurl.com/c7ozldn http://tinyurl.com/c7ozldn
- xero_cuil 14y ago
  Thanks for this!
mbailey 14y ago
Was just reading the JTBD literature a few weeks ago. An obvious concept, but so oft ignored.

80 points by msomers 14y ago

I wonder what would happen if Facebook ran a scheme like AdSense. Advertising on the Facebook site itself seems not to be working, but I could foresee a situation where website owners allow Facebook to display ads on their site for a fee per click just like AdSense. Facebook already has a javascript presence on many, many websites already.

shinchan 14y ago
Wow. Its like you just read my mind! I was about to say the same thing. Indeed, especially since there are a lots of apps out there implementing Facebook login, especially mobile apps, games and apps alike, could this be a solution to their mobile strategy? (this is not a rhetorical question.) Whats stopping them from implementing this?
AznHisoka 14y ago
Adsense makes up a minority of Google's ad profits. The billion dollar business is in Adwords, not Adsense.
- helipad 14y ago
  Indeed - I've no idea what the revenues would be - was just an idle thought. Right now the problem is getting people to believe in Facebook's revenue to get some confidence back in the market. Not saying it's is a silver bullet by any means.
- PelCasandra 14y ago
  Minority? It almost half of all Google ad profits.
  - mda 14y ago
    It is more like Adwords: 70% Adsense: 30% http://investor.google.com/earnings/2012/Q1_google_earnings.html http://investor.google.com/earnings/2012/Q1_google_earnings....
- earl 14y ago
  sadly, no. Adsense is a $10B/year business. Even ex tac of 30%, that's still a big business. From wikipedia [1]: In Q1 2011, Google earned US$2.43 billion ($9.71 billion annualized), or 28% of total revenue, through Google AdSense [1] http://en.wikipedia.org/wiki/AdSense http://en.wikipedia.org/wiki/AdSense
brandnewlow 14y ago
I used to think this as well, but there's a problem. What if FB ads are not as good a monetization solution for publishers as Google ads? This seems quite possible. Facebook advertising is much much cheaper than Google advertising on a per-click and per-impression basis. With Google ads, advertisers are used to paying as much as $20/click in some areas. I've never heard of a FB advertiser paying more than $2/click. Likewise, CPMs on FB are in the single digit pennies quite often. On the web, banner ads are around $0.75 CPMs for decent sites. Facebook might have looked at this and might believe that if they launched a network like this, the payouts just wouldn't be able to compete with what Google's able to give publishers.
mootothemax 14y ago
I wonder what would happen if Facebook ran a scheme like AdSense. Maybe that is/was one of the aims behind the Like button? Facebook could use that connection to tailor the adverts displayed to the user (along with the site's content!), provided the user is logged into Facebook at the time. Admittedly, I think doing so would kick up a bit of a stink ;)
mkr-hn 14y ago
I've never had good results from ticking the box for Google's display network in AdWords campaigns. The culprit is poor implementation by webmasters, not the amount or detail of data available. I don't see why a Facebook display network would do any better.
alain94040 14y ago
They'll do it as soon as they are confident that they can pay websites more per click than AdSense. It's a winner-take-all market. But AdSense is very vulnerable: intent (the search string), which is the one reason Google advertising works, is absent from AdSense. As an example, I once wrote a blog post with a title like "startup financing: take the elevator, not the stairs". Google placed ads on that post for Thyssen elevators. It had just no clue that the topic was startups, not actual elevators. Not impressed.
- earl 14y ago
  Google uses search activity from the past several hours to help match ads in adsense for content[1]. Hiding those from other ad networks is, imo, one of the best reasons they should be sued for antitrust. We've recently started to expand the use of the query words in referral URLs to a few hours so we can so we can continue to deliver more relevant ads. The technical way that we're doing this is by associating the relevant query words in the referral URL with the existing advertising cookie on the user's browser. After a short period of time (a few hours) the query words are no longer used for the purposes of matching ads. [1] http://adsense.blogspot.com/2010/02/better-contextual-matching.html http://adsense.blogspot.com/2010/02/better-contextual-matchi...
  - maxerickson 14y ago
    That article doesn't discuss the details of the implementation, but it could very well be the Adsense js that is tagging the cookie with the referral data. Any ad network can implement that.
    - earl 14y ago
      Perhaps I wasn't clear. Google is hiding referer data, particularly the query string, for a growing number of searchers. Today it's anyone logged into google, or about 60% of traffic to my blog. Tomorrow, they may well do it for anyone. Right now for the logged in users, and potentially in the future for anyone, the only ad network that can still see the search terms in the referer string is adsense, because Google search can pass them to adsense internally. No other ad network can see them. This is obviously incredibly valuable for adsense. And it should trigger an antitrust violation because they're using their monopoly / near monopoly in search to juice their offsite context and display ad networks.

mikecane 14y ago

So basically this is saying that Facebook needs a search engine to be effective at advertising. Is DuckDuckGo up for sale?

suking 14y ago
LoL - Fb could build that pos in 2 hours.
3lit3H4ck3r 14y ago
Why would they buy DDG when they (FB) already have Bing? Oh...yeh. Is DDG for sale? :)

amitvaria 14y ago

But consider Facebook's "friction-less sharing". Facebook now knows you just listened to a song, read and article, took a trip, had a baby, got married.... Now think of the targeting you can get there.

antr 14y ago
All that sounds very nice, but that is what FB has all been about. This kind of non-quantitiave arguments won't validate a $100bn valuation.
sp332 14y ago
The targeting goes all the way down. Gabriel Weinberg (who runs DuckDuckGo) targeted an ad to one person! http://www.gabrielweinberg.com/blog/2010/05/a-fb-ad-targeted-at-one-person-my-wife.html http://www.gabrielweinberg.com/blog/2010/05/a-fb-ad-targeted...
tomelders 14y ago
You sell cars. I listen to Donovan, The Beatles, The Stone Roses, Mozart, Au, REM, Dj Shadow, Handsome Boy Modelling School, Chopin, Ludovico Einaudi and Daft Punk (to name but a few). I've liked pages for five star greek hotels, the burger place on the corner, the guy who sells awesome falafel's on Hoxton market. You know I saw this video of a talking dog that I thought was totally awesome. Just this month, I read an article on the Greek debt crisis in the economist, an article on David Beckham in the Daily mail website, I shared a wikipedia page about Rommel, I liked an article about continuous integration with node.js and Jenkins. Now sell me a car. I actually like Volkswagens and Volvos, but you won't find that anywhere in my Facebook data.
- cageface 14y ago
  I actually like Volkswagens and Volvos, but you won't find that anywhere in my Facebook data. I'd bet there's actually a pretty high correlation between many of the things you just listed and a preference for VWs and Volvos. It doesn't really take that many data points to start filling in the blanks. But I think this kind of targeting is going to be more useful in selling you a product you don't yet know about but are likely to want than it is a car that you are almost certainly already aware of.

tomelders 14y ago

For me, this is the strongest case yet for niche social networks being the future of social media. FB is too broad and demographics tell you little about the people you are speaking too unless you can inject some sort of context. If (via a Distributed Social Networking Protocol) people were subscribed to several social networks focused on their hobbies and interests, then I believe Social Media Advertising would be more viable, and more useful to the audience. I also think that when you look past all the fanfare and hyperbole surrounding social networking, what you're left with is really sophisticated bulletin boards and forums. I can see that from a marketing point of view, having one point of access to that many people might seem like a dream come true, but the numbers are suggesting that's not the case at all. Facebook is not the answer to an online advertisers dream. I don't think it ever will be.

ajdecon 14y ago
For me, this is the strongest case yet for niche social networks being the future of social media. FB is too broad and demographics tell you little about the people you are speaking too unless you can inject some sort of context. If (via a Distributed Social Networking Protocol) people were subscribed to several social networks focused on their hobbies and interests, then I believe Social Media Advertising would be more viable, and more useful to the audience. You don't need DSNP: niche social networks already exist. Ravelry and GoodReads, as examples, seem to be making a decent go of it. Would love to know how their advertising does.
- tomelders 14y ago
  I think DSNP is essential. Imagine if you could only email someone with a GMail account if you had a Gmail account. Imagine if you could only check your Hotmail emails on the Hotmail website or through one of the official Hotmail Apps. That's how Social networking works right now. API's won't cut the mustard. We need a DSNP.

antidoh 14y ago

I am so glad that email didn't depend on advertising to eventually support itself "sometime in the future." Email is useful to me as email, and useful to my providers as a billed service. Yes, they're different things.

cletus 14y ago

I agree with this and have posted numerous comments here to that effect, particularly on the value of intent (with respect to the effectiveness of online advertising). Th attitude of "eyeballs not revenue" and finding that repeatable, scalable formula are almost a religion on HN and in startups. It's generally a strategy I approve of. Even dropping 20% since the IPO (currently trading at just over $30 as I write this; and bear in mind that even Groupon--at least initially--enjoyed a nice IPO bounce), the company is trading at a huge P/E ratio. I get why this is: it's speculative. FB is still viewed as a growth company and the speculators feel that there is huge unrealised monetization potential. I remain a skeptic regarding the value of "social" in advertising. The OP is right: all this data just means--maybe--a slightly higher CTR, at which point Facebook is just another display ad network and that doesn't justify their valuation. As an aside, IMHO Twitter is in this same "put up or shut up" boat. I don't believe Facebook is doomed (IMHO Twitter is). I just believe the value of the data silo they have is both overstated and transitory (at some point--one way or the other--Facebook won't be the gatekeeper to your profile and social graph). Facebook has done a lot of things right as a business (the Like button being foremost among those IMHO). Personally I believe their biggest mistake was spurning Apple: Apple wanted to use Facebook for their Ping boondoggle. Disclaimer: I work for Google in display advertising.

RegEx 14y ago
I'm confused over the significance of the P/E ratio. It seemed like a big deal when I was reading these stories about the FB IPO, but then I got to look at other ratios in tech. Amazon's is much higher than FB's 93.72 at 175.23 [0] [0]: http://www.google.com/finance?q=NASDAQ:AMZN&authuser=0 http://www.google.com/finance?q=NASDAQ:AMZN&authuser=0
- tatsuke95 14y ago
  Oh, don't worry, the bears consider it out of whack and a good short as well. I'm not exactly sure what the deal is with Amazon, but their p/e was much more realistic in the not-so-distant past: http://ycharts.com/companies/AMZN/pe_ratio http://ycharts.com/companies/AMZN/pe_ratio
- 14y ago

fakhrazeyev 14y ago

It does seem obvious. There is just a wall of hype hanging. And factor in the fact that it was reported that approximately 40 % of all social network accounts are created by spammers.

pbharrin 14y ago

Google's ads are better because you have intention when you are searching. That's not true when you are checking into Facebook to see what other people are doing. Are there other times when people use Facebook and they have some intent? Planning an event, or asking a question, those come to mind.

glesica 14y ago

Seems reasonable to me. Facebook is gathering what amounts to a static picture of each user then selling ads based on the pictures. There really isn't a time component in the data (though perhaps they can add one now that the service has existed for many years). On the other hand, Google has more or less assembled a bunch of archetypes based on behavior (people who search for X later purchase Y, etc.). Then they match queries to these archetypes. The time component is front-and-center here because the archetype match was prompted by a user action. A purchase is really a time-related event that transforms the user in some sense. By purchasing something you're changing "who" you are from the perspective of an advertiser (e.g. a Chevy driver could become an Audi driver). So Facebook has a great picture of "who" each user is at a the moment, but no real way of knowing "who" each user would like to become or when. Hence, perhaps, some of their difficulty in selling ads. Or not.

kevinlu310 14y ago

Facebook's advertising model is a simple classical displaying advertising model. It need be disrupted seriously. Nowadays, the most popular marketing model is displaying advertising model, which is actually an attention-driven marketing model. However, no matter how good they are context-awared and personalized, they are always guesswork, which can hardly achieve high relevancy. It is extremely difficult to guess users’ actual intention for most cases, even you may have a lot of information of them. For example, it is super difficult to guess what an individual is exactly looking to buy when he walks into a Trader Joe’s, no matter how much information you obtained on him, like his income, profession, age, education and etc. Users’ intentions are always dynamically changing from time to time. There are no general equations that can perfectly predict people’s intentions. As a result, no matter how perfectly a system can attract attentions, attention-driven marketing can never achieve high efficiency and accuracy. As a matter of fact, nowadays advertisement industry is the biggest bubble of the world. Every year, trillions of dollars completely wasted by irrelevant advertising unnecessarily. What is even worse is, the user experience is badly hurt by ads and spams. Such marketing system can be called as attention-driven marketing system. Everything is done to attract attentions. However, only a few attentions are really attracted, but majority of the attentions may not be finally turned into intentions. As a result, majority of the marketing costs are totally wasted. The future marketing systems will be built on intentions. Intention-driven commerce systems need let every user to express his/her intentions freely and help him/her achieve his/her intentions in highly optimized way by socialization and crowd-sourcing. The future world needs to be a world without displaying ads, but only pure relevant information that matches users' intentions dynamically and real-timely. This is the very objective of my current startup project as well.

alain94040 14y ago
How do you promote events in your non-bubble view of the world? An event can be in person (imagine TechCrunch Disrupt for instance), or a new TV series launching... I can imagine very well how I would target people who like the series 24 if I'm launching a new thriller. Or readers of Hacker News for a tech conference. Etc...
- kevinlu310 14y ago

mag487 14y ago

Not trying to defend the viability of FB ads here, but this analysis seems a little shallow. Falling within a certain demographic may not "make" you click an ad, but the same could be said of performing a search on Google. If circumstance Y makes one want to buy a product, then as long as falling within demographic X is correlated with being in Y, targeting X could be a good idea. It is (obviously) the reason why toy commercials air during children's shows and not during basketball games.

ma2rten 14y ago

You are mistaken to think that the search-engine show-people-what-they-were-searching-for-anyway kind of advertisement is the only kind of advertisement that can work on the internet. I used to work for Hyves, which was until a year ago or something the biggest social network in the Netherlands. Hyves lost to Facebook in terms of number of users, but something that they did do right was modernization. Hyves had an in-house sales team and studio, which sold custom viral campaigns to cooperate advertisers. These campaigns could usually go together with a TV campaign and/or other media. For some products search engine advertisement is the right kind of advertisement (parrot secrets, maybe cars, I don't know), for others not such much (food stuffs for example).

CookWithMe 14y ago

I agree that Google can target alot better IF you have a product that solves a problem people search for. IF. > What researchers discovered was that people bought milkshakes as a breakfast replacement because it was entertaining during a long, boring commute, and would keep them full until lunch. How do you advertise for them on Google? If they search for "breakfast replacement for long, boring commute"? For anyone who can match their product to search queries, Google is the way to go, agreed. But not all products can, because people are not actively looking for solving a problem/need that your product solves. If I look at my current facebook ads, these are not things I actively search for, e.g. drinks, shavers, parties, facebook games, ... If I had to spend marketing $ on these, I'd probably choose facebook ads over AdWords as well.

redwood 14y ago

There are two primary types of ads 1) ones that give you what you want 2) ones that drive demand Google dominates (1) and I assume Facebook is trying to dominate (2) which includes a lot more display advertising (and TV, and billboards). Facebook probably thinks a lot of demand-driving advertising can be accomplished through viral online marketing through FB platforms and they are likely trying to exploit these channels with services for paying customers. Imagine paying Facebook to simulate something being trendy among people's trend-setting friends. It'll be a fine line but this must be what they hope to do. Agree the $100B is too high, but certainly they're easily a profitable company, long-term.

billybob 14y ago
> 1) ones that give you what you want > 2) ones that drive demand A big part of Google's success is that #1 is much easier to prove than #2. "People bought our product more because they had become familiar with it over the last month via newspaper/radio/Facebook ads" is a hard-to-prove claim. "We got X clicks from Google ads leading directly to Y purchases" is much easier.

kelvin0 14y ago

Ever since FB has gone public, I've noticed so much 'negative' stories/comments, and I really wonder why these are only coming out now... FB has been around for a long time and nothing has changed in its marketing strategies since it seems. Personally, I don't have a FB account, or own any stock but this targeted 'bashing' seems useless and has the smell of targeted anti-FB propaganda. Why don't we focus on the real issue being people rights to privacy and the shady practices of FB? More focus on the real issues please, do not get distracted by the 'anti-hype' of the week.

tatsuke95 14y ago
>"Why don't we focus on the real issue being people rights to privacy and the shady practices of FB?" These things are all connected. The public is being sold the idea that this is a $100BB company, one of the biggest in the world. Yet when some of us look at it (yourself, even) we see a great tool that doesn't justify the price as a business: as cletus says at the top of the thread, Facebook may just be another ad network that depends on its massive, engaged user base. Yet FB has problems with how it actually treats those users, through privacy issues and shady practices..
- kelvin0 14y ago
  I don't think that anyone who did their homework would have bought FB stock as a 'sure thing for long time investment'. FB didn't fool anyone who wasn't too greedy for their own good. Obviously I'm not saying FB is not a hugely successful product with an immense user base, but obviously the fact that this IPO is not 'going as predicted' by it's underwriters was clear to me waaayyy before the stock hit the markets ...

alain94040 14y ago

I call survivor bias. Historically, the only player in online advertising is Google, which is 100%-focused on intent-based advertising. So the article, logically concludes that intent is the only way to advertise (despite decades of TV and radio ads). Some products do great with intent-based advertising. Other products (that you didn't even know you needed) need other forms of advertising.

colinshark 14y ago

This argument continues to come up, but I don't buy it- At least not without some data. My personal anecdote: Facebook ads are the only online Ads I click with any regularity. They are targeted to my interests, such as programming. They also hit me at a time when I am mostly bored and clicking around on anything that interests me. When I find myself on Google, I am on a mission for very specific information. Even if I am shopping, I do a research phase and a purchasing phase. Ads in the research phase don't woo me. I google straight to trusted sources. Once I'm in the purchase phase (shopping for best price), I am not wooed either because the ads are too broad. E.G, I want the D3100, not just any camera in general. I completely understand Google's strategy of "intent" based ads. I just think Facebook's user data is even stronger.

rajpaul 14y ago

If I create high quality, high traffic Facebook content, do I get any money from advertising?

torstesu 14y ago

Cached version for those getting the 509: http://tinyurl.com/c7ozldn http://tinyurl.com/c7ozldn

xero_cuil 14y ago
Thanks for this!

mbailey 14y ago

Was just reading the JTBD literature a few weeks ago. An obvious concept, but so oft ignored.