Fred, you might be interested in an article a colleague of mine wrote a couple of weeks ago about the same topic - Ian Kennedy calls it the "Unified Theory of Interweb Economics".
The theory goes something like this:
1. Advertising is a function of your traffic volume, the more traffic come to your site, the higher rates you can charge. 2. Social sites such as Facebook and Twitter have a lot of link sharing going on as friends post links to share them with each other. 3. When Facebook and Twitter send more traffic than Google search referrals, advertising dollars will follow the source of that traffic. 4. Google’s dominance in online advertising will be threatened.
As a marketer I like the theory but in practice I think the paradigm shift will come not only when the traffic moves but also when the trust factor shifts. Google has spent quite a bit of time, effort and money to brand themselves as the "good company". From the "do no evil" to the environmental funds, they have always believed that if people trust a company to do good then they'll also trust them with their money (in this case, with their searches).
That is something that I don't think Mark Zuckerberg and co. understand (or at least not until recently). Facebook operates under the concept that you interact with those who have similar likes and dislikes and that by building a bridge between people Facebook is inherently a good company and trustworthy. The problem is that their recent PR blunders with their terms of service and the whole Beacon fiasco have moved them into the distrust column. Add to that the fact that users are up in arms about redesigns and you have a company that has a somewhat harsh relationship with its userbase.
In my opinion, money will definitely flow into Facebook, but until they can create a feeling of trust within their "walls" they won't be able to generate ROI for their advertisers (trust flows through the web - if people don't trust the link they'll also mistrust the product or service being sold). And that's one thing that we can count as certain - marketers will ALWAYS follow the ROI, and that's why Google is still King. There is very little out there that returns the kind of ROI and volume than Google does.
Udo Szabo
· 8 months ago
@avertiz - absolutely agree that the notion of "trust" is critical to any social personal experience. You have to give credit to FB so that they are actively pushing the boundaries in this space, at the cost of tapping into the fire every once in a while.
What i don't know if the relation of "trust" discussion between the people here on the blog and the users in the field. Its always difficult to draw conclusion from small personal research, but as of know i haven't felt any major trust implications in the many people that use FB or other services in my personal environment. And lets be fair, every users who joined in the past 9 month probably doesn't even know what "Beacon" is.
What we found to be the most difficult part in terms of monetizing the rising amount of social traffic is to bring a users "intent" in "context" of the value it has for a site with an economic incentive. As you share all kind of things over the day with all kind of people that find bits and pieces of your stream more or less relevant optimizing the stream against economic interest is the big task.
I am sure somebody will figure it out on the grand scale. Until then expect more vertical models to work that filter the relevant "context" from the users stream by interest, like stocktwits and others do.
fredwilson
· 8 months ago
Hmm
I'll go read the post
Udo Szabo
· 8 months ago
Still pretty rough on the edges. Lets me know where you disagree.
Mark Essel
· 8 months ago
Thanks for the link to Ian's post Udo. Makes you wonder how fast the shifts will be in advertising dollars?
Where can we review trends for derivatives/rates of change from internet information flow statistics?
Prokofy
· 8 months ago
Re: Are sharing and “social discovery” more important sources of traffic than search in certain verticals? Is this a harbinger of things to come across a larger swath of Web sites?
There are two kinds of commerce, 1) the commerce of the bazaar, where you need relationships and intuitive reading of metadata and you pay the price you succeed in haggling, and 2) the commerce of the marketplace, where you don't need relationships and the price tag is what you pay. A lot of modern civilization is about freeing people from social relationships that made them dependent in order to be able to engage in commerce. All that is happening on the Internet now is that this is being reiterated; social media with its insistence on friending and following and reputational points isn't magical, it's retrograde and maybe even plunging us back to the dark ages.
I find having a blog in the niche "Second Life" or "virtual worlds", that other similar niche sites linking to my site, and Twitter and Plurk links produce a significant amount of traffic that pays off, i.e. clicks on the ads. Google search might produce more traffic all together, but not conversion traffic.
But what matters is not finding if this is true and answering your question "yes" but what to do about it, and more importantly, what effects it will have on the market in general. If you make all ad campaigns or news distribution follow in the tracks of passed links or sharing among likeminded networks of people, in the belief that this "pays out more," you are building the closed circuit of the bazaar, where you need a relationship and be a friend of a friend to close a sale. By contrast, open search without relationship metadata tacked on to it is the modern marketplace where you buy something without having to get a word-of-mouth or a handshake to close the deal.
You need both. It is undermining the modern open society to force people back into ancient tribal or medieval village grooves where you have to know someone who knows somone to get something or only certain approved castes or craftsmen get to sell things in a closed loop. Merely making this more scaled and automatic by putting it on Twitter so that the "friend" is really a stranger still doesn't undo the problem of the closed circuit and making the marketplace less flexible ultimately. Not every aspect of commerce can be driven by social media or you will create a very hierarchical society too dependent on powerful aggregators and influencers with too high a bar to enter the market for new business.
Igniter
· 8 months ago
As an aside, this post is another great reminder of how critical metrics and tracking are in the social web. If you are measuring it you are running blind. And tracking/measurement is going to be the basis of what's needed to make the realtime web deliver relevance.
theschnaz
· 8 months ago
This line really struck a chord with me, "I’m getting more and more of my information from the people I’m connected to through email, IM, RSS, Facebook, and Twitter. Also, the nature of the searching I’m doing now is much more targeted and specific."
I started to think about where/how I hear about the media I like (blogs, books, movies, music, games etc.) and I noticed a natural dichotomy between short and long time commitments.
I consider books and blogs to be a long commitment; you can put in a lot of time reading these. Now, I only make a long commitment if a friend or a trusted source recommended it. I thought about the last 5 blogs and books that I’ve read and all have been previously recommended. Seth Godin mentioned this idea on his blog recently, when you start a book, you're "supposed" to finish it, no one wants to read through one they don't like. How will you know if you'd like it? Your friend recommended it.
Short commitments include music, movies and games. I'll listen to a CD because I’ve heard a song on the radio or I'll see a movie because I saw the trailer. If a friend recommends an album or movie, all the better, but it's not required. A movie only takes 2 hours of my time, so if it's bad, no big deal.
I think this is why Daily Candy and Thrillist are so successful. They act as trusted sources (similar to a friend) that recommend things. If Thrillist suggests something, chances are I'll like it and their readers will too (I have personal experience with this.)
For search, now that I have a G1, almost all of my searches are hyper-specific/local (finding locations, store hours, bus schedules, etc.) I never search for new books or blogs.
To summarize, I'll only consider buying or reading media that requires a long commitment if it has been recommend by a trusted source. Media that requires a short commitment can still be "pushed" to me through commercials or "traditional" marketing techniques.
I’d love to hear this community’s thoughts. Think about the last 5 books/blogs you’ve read, how did you hear about them?
Greg
awilensky
· 8 months ago
It figures the least effective way to share links is email, that is used by all of my friends and family. They just can't be bothered to go to delicious, and wont use the share me widgets because the mode is too obscure. A instant click share for twitter would be best.
fredwilson
· 8 months ago
Email is the most effective he says
awilensky
· 8 months ago
I meant that it figures that what was found to be most often used is the most unwieldy. It take more clicks and the shared links are harder to manage - which doesn't seem to matter to most people, but mean everything to me.
fredwilson
· 8 months ago
Got it
Jay Meattle
· 8 months ago
Alan, have you seen www.bzzster.com? one-click way to email links.
leigh
· 8 months ago
Interesting to see how this starts to change with younger demo seemingly abandoning email.
However, when clients say (which they do often) that they think email is not for them, i'll now add these stats to my long list of "email is dead" myth busters
Alan Warms
· 8 months ago
Fred I commented via twitter to @fredwilson - be cool if I could have figured a way to make that happen back here. Wanted my followers to see the thread. Should I have started it here?
fredwilson
· 8 months ago
Yes, add twitter to your disqus profile and you'll see a checkbox where you can twitter the comment
aaronklein
· 8 months ago
Fred,
I can personally vouch for the accuracy of this. Interestingly, my "new visitors" blog traffic is driven as follows: 35% Facebook, 29% Google and somewhere between 10% and 20% Twitter.
(The Twitter ambiguity comes from the fact that only 6% of traffic is logged from twitter.com, but my bit.ly links published on Twitter show a lot of TweetDeck click-throughs.)
I only wish bit.ly and feedburner were directly trackable in Google Analytics, so I could get a per-post breakdown of exactly how much traffic is repeat from feeds and e-mail, and how much is organic growth from either passed or searched links.
Aaron
Mike Su
· 8 months ago
Just a quick point specifically on #4 - I don't think you get much linkback credit via social sharing. I believe most of FB, since it's behind a password wall, will never get crawled by the spiders (not sure about this), but I know for a fact Twitter no-follows all external links, so you definitely won't get any credit for that. Throw in the mix various URL shortening services that handle redirects in varying degrees of SEO friendliness, and I believe the impact is minimal. But the traffic is gold. And, if every 100 or 1000 social links shared leads to one blog link on a decent pagerank blog, then that's still worthwhile.
avertiz
· 8 months ago
I can truly attest to no. 6 with my own findings - I'll try to add a bit more insight...
One of my key ethnic/multicultural marketing strategies is to find the right niche sites and build strong direct advertising relationships (whenever possible we go beyond traditional IAB placements). What I've found and what compliments Ben's comments is that there is an inherent trust level attributed to each site/placement/etc. People are more likely to pass on links when they trust the source (not just the person providing the link but the site itself). In our trials, we've seen that even when shown the same link/ad people pass it on more if it comes from a news source (particularly if focused on a niche/vertical) followed by a message board, a blog, an ethnic portal and finally by social media sites and vehicles (Twitter).
What we are seeing is that even if the offer is spot on and the content is compelling we still have a better ROI on passed links when "pushing" them through a trusted source (within a vertical). In the case of many ethnic/multicultural markets that source tends to be ethnic news media. Sites like Facebook, MySpace and such tend to have a lower "trust" value and although they offer high visibility they don't generate enough ROI (convert) for a direct marketing campaign.
Just like Ben mentioned, a lot of the conversations our target markets are having have moved to social media and RSS, however the "trust factor" has not yet made the move. Outside of one-to-one recommendations (IM is a good example), traditional outlets still outweigh social media when it comes to ROI for passed-on links.
The big question is when will it change? When will someone put as much stock into a link passed on via a Facebook wall update as one passed on via a news site, blog or message board posting?
Ben Straley
· 8 months ago
Hi all. It's good to hear that others are seeing similar trends in traffic sources etc. To be clear, I do not believe traffic from passed links will surpass search for the foreseeable future. I just think shared links are already driving a meaningful % of visitors for many sites and are one of the most cost-effective and fastest growing sources. What's not to love about that if you're a publisher or advertiser?
Ethan Bauley
· 8 months ago
What makes you think that passed won't surpass search sooner than later? A hypothesis I'm playing around with is that Google relatively overvalues links from more complicated publishing technologies (blogs/standard HTML) versus simple publishing tech (Twitter, FB, etc). So, if the velocity and quantity of links from the latter increase at a rate faster than the former, it should change the mix.
I think there is some value in looking at things from the POV of the link publisher (and the demo/psycho/technographics of that population)
At any rate, thanks for sharing that analysis, Ben really killer! I'm going to pass along a link to it to all my pals...via a link I hard code into a static HTML page
;-)
Udo Szabo
· 8 months ago
Fred & all, we continued the discussion on Ian´s blog with another interesting discovery. What if not only the way traffic get directed is shifting to more social exchanges, but also the quality of this traffic, aka the attention/time spend on the site is higher. Ian has digged out some numbers from his own blog, would be interesting to see if other sites see a similar behavior.
Here is the link to the full article "Traffic Sources and Attention": http://bit.ly/wceXq
fredwilson
· 8 months ago
Thanks. I'll check it out
Mike Su
· 8 months ago
Here's a data point for you: Ashton Kutcher has 1.3mm followers on twitter. He shared a link with great headline to a video of ours yesterday:
How many twitter.com referrals would you guess we got from his tweet to his 1.3mm followers, and all of their re-tweets?
If you guessed 30k, you'd be right.
fredwilson
· 8 months ago
A ~2.5% ctr
That's not bad
When I update twitter with a link to a new blog post, I often get a chartbeat notice that there are 80 or 100 people on my blog at that moment
I have close to 19,000 followers so that could be a ctr of as much as 5%
I'd be curious what other people see
Mike Su
· 8 months ago
I also suspect ctr's are generally much higher for people with fewer followers. Each follower is probably a personal relationship that is far more apt to click through. For example, I have <100 followers, and when I look at bit.ly stats for links I share, each link gets ~20 clicks (look at those 10 lines of code in action!!). So that's >20% ctr, but still only 20 clicks. But it does imply that when it comes to passed links, it's probably better to have 1000 people with 100 followers share a link than to have 1 person with 100,000 followers share the link. That probably then implies a bunch of other stuff that I'm too lazy to figure out right now.
brendan mcnulty
· 8 months ago
i work in games and this would be a metric i would love to get my hands on, and more importantly find some way of stimulating the pass along. its always said if you can measure it, you can optimise it.In terms of undersatnding your dashboard and getting a full perspective on where your traffic is coming from, this could swing spend away from advertising and towards more compelling content creation.
All Comments
The theory goes something like this:
1. Advertising is a function of your traffic volume, the more traffic come to your site, the higher rates you can charge.
2. Social sites such as Facebook and Twitter have a lot of link sharing going on as friends post links to share them with each other.
3. When Facebook and Twitter send more traffic than Google search referrals, advertising dollars will follow the source of that traffic.
4. Google’s dominance in online advertising will be threatened.
The full story can be found on Ian´s blog here: http://tinyurl.com/de9l22
As a marketer I like the theory but in practice I think the paradigm shift will come not only when the traffic moves but also when the trust factor shifts. Google has spent quite a bit of time, effort and money to brand themselves as the "good company". From the "do no evil" to the environmental funds, they have always believed that if people trust a company to do good then they'll also trust them with their money (in this case, with their searches).
That is something that I don't think Mark Zuckerberg and co. understand (or at least not until recently). Facebook operates under the concept that you interact with those who have similar likes and dislikes and that by building a bridge between people Facebook is inherently a good company and trustworthy. The problem is that their recent PR blunders with their terms of service and the whole Beacon fiasco have moved them into the distrust column. Add to that the fact that users are up in arms about redesigns and you have a company that has a somewhat harsh relationship with its userbase.
In my opinion, money will definitely flow into Facebook, but until they can create a feeling of trust within their "walls" they won't be able to generate ROI for their advertisers (trust flows through the web - if people don't trust the link they'll also mistrust the product or service being sold). And that's one thing that we can count as certain - marketers will ALWAYS follow the ROI, and that's why Google is still King. There is very little out there that returns the kind of ROI and volume than Google does.
What i don't know if the relation of "trust" discussion between the people here on the blog and the users in the field. Its always difficult to draw conclusion from small personal research, but as of know i haven't felt any major trust implications in the many people that use FB or other services in my personal environment. And lets be fair, every users who joined in the past 9 month probably doesn't even know what "Beacon" is.
What we found to be the most difficult part in terms of monetizing the rising amount of social traffic is to bring a users "intent" in "context" of the value it has for a site with an economic incentive. As you share all kind of things over the day with all kind of people that find bits and pieces of your stream more or less relevant optimizing the stream against economic interest is the big task.
I am sure somebody will figure it out on the grand scale. Until then expect more vertical models to work that filter the relevant "context" from the users stream by interest, like stocktwits and others do.
I'll go read the post
Where can we review trends for derivatives/rates of change from internet information flow statistics?
There are two kinds of commerce, 1) the commerce of the bazaar, where you need relationships and intuitive reading of metadata and you pay the price you succeed in haggling, and 2) the commerce of the marketplace, where you don't need relationships and the price tag is what you pay. A lot of modern civilization is about freeing people from social relationships that made them dependent in order to be able to engage in commerce. All that is happening on the Internet now is that this is being reiterated; social media with its insistence on friending and following and reputational points isn't magical, it's retrograde and maybe even plunging us back to the dark ages.
I find having a blog in the niche "Second Life" or "virtual worlds", that other similar niche sites linking to my site, and Twitter and Plurk links produce a significant amount of traffic that pays off, i.e. clicks on the ads. Google search might produce more traffic all together, but not conversion traffic.
But what matters is not finding if this is true and answering your question "yes" but what to do about it, and more importantly, what effects it will have on the market in general. If you make all ad campaigns or news distribution follow in the tracks of passed links or sharing among likeminded networks of people, in the belief that this "pays out more," you are building the closed circuit of the bazaar, where you need a relationship and be a friend of a friend to close a sale. By contrast, open search without relationship metadata tacked on to it is the modern marketplace where you buy something without having to get a word-of-mouth or a handshake to close the deal.
You need both. It is undermining the modern open society to force people back into ancient tribal or medieval village grooves where you have to know someone who knows somone to get something or only certain approved castes or craftsmen get to sell things in a closed loop. Merely making this more scaled and automatic by putting it on Twitter so that the "friend" is really a stranger still doesn't undo the problem of the closed circuit and making the marketplace less flexible ultimately. Not every aspect of commerce can be driven by social media or you will create a very hierarchical society too dependent on powerful aggregators and influencers with too high a bar to enter the market for new business.
I started to think about where/how I hear about the media I like (blogs, books, movies, music, games etc.) and I noticed a natural dichotomy between short and long time commitments.
I consider books and blogs to be a long commitment; you can put in a lot of time reading these. Now, I only make a long commitment if a friend or a trusted source recommended it. I thought about the last 5 blogs and books that I’ve read and all have been previously recommended. Seth Godin mentioned this idea on his blog recently, when you start a book, you're "supposed" to finish it, no one wants to read through one they don't like. How will you know if you'd like it? Your friend recommended it.
Short commitments include music, movies and games. I'll listen to a CD because I’ve heard a song on the radio or I'll see a movie because I saw the trailer. If a friend recommends an album or movie, all the better, but it's not required. A movie only takes 2 hours of my time, so if it's bad, no big deal.
I think this is why Daily Candy and Thrillist are so successful. They act as trusted sources (similar to a friend) that recommend things. If Thrillist suggests something, chances are I'll like it and their readers will too (I have personal experience with this.)
For search, now that I have a G1, almost all of my searches are hyper-specific/local (finding locations, store hours, bus schedules, etc.) I never search for new books or blogs.
To summarize, I'll only consider buying or reading media that requires a long commitment if it has been recommend by a trusted source. Media that requires a short commitment can still be "pushed" to me through commercials or "traditional" marketing techniques.
I’d love to hear this community’s thoughts. Think about the last 5 books/blogs you’ve read, how did you hear about them?
Greg
However, when clients say (which they do often) that they think email is not for them, i'll now add these stats to my long list of "email is dead" myth busters
I can personally vouch for the accuracy of this. Interestingly, my "new visitors" blog traffic is driven as follows: 35% Facebook, 29% Google and somewhere between 10% and 20% Twitter.
(The Twitter ambiguity comes from the fact that only 6% of traffic is logged from twitter.com, but my bit.ly links published on Twitter show a lot of TweetDeck click-throughs.)
I only wish bit.ly and feedburner were directly trackable in Google Analytics, so I could get a per-post breakdown of exactly how much traffic is repeat from feeds and e-mail, and how much is organic growth from either passed or searched links.
Aaron
One of my key ethnic/multicultural marketing strategies is to find the right niche sites and build strong direct advertising relationships (whenever possible we go beyond traditional IAB placements). What I've found and what compliments Ben's comments is that there is an inherent trust level attributed to each site/placement/etc. People are more likely to pass on links when they trust the source (not just the person providing the link but the site itself). In our trials, we've seen that even when shown the same link/ad people pass it on more if it comes from a news source (particularly if focused on a niche/vertical) followed by a message board, a blog, an ethnic portal and finally by social media sites and vehicles (Twitter).
What we are seeing is that even if the offer is spot on and the content is compelling we still have a better ROI on passed links when "pushing" them through a trusted source (within a vertical). In the case of many ethnic/multicultural markets that source tends to be ethnic news media. Sites like Facebook, MySpace and such tend to have a lower "trust" value and although they offer high visibility they don't generate enough ROI (convert) for a direct marketing campaign.
Just like Ben mentioned, a lot of the conversations our target markets are having have moved to social media and RSS, however the "trust factor" has not yet made the move. Outside of one-to-one recommendations (IM is a good example), traditional outlets still outweigh social media when it comes to ROI for passed-on links.
The big question is when will it change? When will someone put as much stock into a link passed on via a Facebook wall update as one passed on via a news site, blog or message board posting?
I think there is some value in looking at things from the POV of the link publisher (and the demo/psycho/technographics of that population)
At any rate, thanks for sharing that analysis, Ben really killer! I'm going to pass along a link to it to all my pals...via a link I hard code into a static HTML page
;-)
Here is the link to the full article "Traffic Sources and Attention": http://bit.ly/wceXq
http://twitter.com/aplusk/status/1578292226
How many twitter.com referrals would you guess we got from his tweet to his 1.3mm followers, and all of their re-tweets?
If you guessed 30k, you'd be right.
That's not bad
When I update twitter with a link to a new blog post, I often get a chartbeat notice that there are 80 or 100 people on my blog at that moment
I have close to 19,000 followers so that could be a ctr of as much as 5%
I'd be curious what other people see