The One Thing Wrong With Social Content Sites Like Digg

About This Blog

This site is for  entrepreneurs.  A full RSS feed to the articles is available.  Please subscribe so we know you're out there.  If you need more convincing, learn more about the site.

Community

Google+

And, you can find me on Google+

Connect on Twitter

Get Articles By Email

Your email:

Google

Blog Navigator

Navigate By : 
[Article Index]

Questions about startups?

If you have questions about startups, you can find me and a bunch of other startup fanatics on the free Q&A website:

Answers.OnStartups.com

Subscribe to Updates

 

30,000+ subscribers can't all be wrong.  Subscribe to the OnStartups.com RSS feed.

Follow me on LinkedIn

OnStartups

Current Articles | RSS Feed RSS Feed

The One Thing Wrong With Social Content Sites Like Digg

 

I'll jump to the punchline first:  What's the one thing that's wrong with social content sites like digg?  The content!

Social content sites have really grown recently and the top ones get a significant volume of web visitors on a daily basis.  Odds are, if you are a regular reader of OnStartups.com you found this site originally through one of the social content sites like digg, reddit or Netscape.
 
I'm a regular user of both digg and reddit and have submitted, voted and read articles on both sites on a relatively regular basis.  I met the founders of reddit when they first announced their product/company in Cambridge, MA and have been following the company ever since.
 
I really only have one major issue with the social content sites out there right now.  The content sucks.  Though this may seem like a relatively strong statement to make, I'm confident that if you visited any of these sites right now and looked at the top 5-10 stories/links, you'd find that the noise to signal ratio is really, really high.  There's a reason for this.  Most of these sites are not really designed for you.  If you're like me (and chances are you're more like me than you are the average digg user), and you took a critical look at the content on digg.com at any given time, you'd find the relevancy pretty low.
 
So, why does the most popular social content site on the Internet have content that sucks?  Clearly, the whole point of the site is that the content should not suck.  The reason is very simple:  The content is really not focused on you or your needs.  It's not meant to be.  It's meant to rank the most popular content based on the opinion of a bunch of random people you don't know or care about whereas what you're probably really interested in is content based on the opinion of a very small number of people you do care about.  These are the people with whom you share some interests and or/or have some respect for. 
 
Digg has categories (and sub-categories), but that doesn't really help.  The content that gets submitted in the first place (and then gets voted on) is primarily a function of the user-base and the user-base is broad and diffuse.
 
My co-founder at HubSpot wrote on an article on this very topic a little while ago.  The title of the article is "Let A Million Diggs and Reddits Bloom".  If you're interested in an analytical look at the social content industry (with comparsions to the publishing industry), it's well worth the read.  The basic point of the article is similar to the one I'm going to make here:  Social content should be about relevancy and the best way to improve relevancy is to narrow the focus. 
 
For the past few months, I've been working on a small project to test this theory.  The project has resulted in a small experimental site called DailyHub.com.  DailyHub is basically a social content site targeted at business geeks.  The idea here is to start with a specific target audience (business geeks), and then build content around that audience.  This is not unlike the whole blog phenomenon.  Blogs were largely successful because they were mostly targeted at a specific topic or group of people.  The whole idea was to create very focused content that catered to the needs of a specific audience.  I think the same idea will work for social content sites.
 
So, if you're a business geek, I'd appreciate it if you could wander by DailyHub.com and see if the content is relevant.  If you're a registered and approved member of OnStartups.com, your existing username/password will work there too and you can login immediately and submit content and vote on links.  As with all things, this is an experiment to see if a more narrowly focused audience can aggregate content that is fundamentally more useful.  The software that powers DailyHub has been developed internally, but the long-term intent is to make it a platform and let others build their own topic-focused social content sites.
 
If you get a chance to try it out, let me know what you think.  I'm more interested in feedback on the idea than the software (as the software is still very early).  If the software is bad, I can fix that.  If the idea is bad, that's a different matter.  Please leave a comment with your thoughts.
 

Posted by Dharmesh Shah on Thu, Apr 19, 2007

COMMENTS

Hey Dharmesh, I agree with you completely that the content on digg and reddit has gone downhill as their popularity exploded. My only question is beside branding the social content application for business geeks, what from the technology side will prevent a new social content site become a free for all like digg. Don't get me wrong I think making it clear from the beginning will setup a context for future submissions , it seems to working well for news.ycombinator. Just curious if you had any ideas for implementing other tech features besides the vote up and down model to focus content.

posted on Thursday, April 19, 2007 at 12:44 PM by marvin


Marvin: Great question. I've answered this (and other questions) in the DailyHub Manifesto page, accessible here: http://www.dailyhub.com/Manifesto/tabid/7872/Default.aspx The idea is that because the site is positioned as being for a specific target audience, it will likely attract those kinds of people. So, if you're looking for celebrity gossip content and you mistakenly wind up on DailyHub (despite my best efforts), chances are, you'll figure out quickly the site is not for you and not register, submit links or vote. So, the idea is that simply by making it clear who the site is for will keep those that are not interested from participating. Of course, the idea is that there should be a site out there for everyone based on their area of interest. Once again, I think of this a bit like blogs. Though it is possible for a blog to have readers outside the target audience, most of regular readers (and particularly RSS subscribers) are those that have an interest in the topic.

posted on Thursday, April 19, 2007 at 12:53 PM by


I agree that personalization is the direction that things are heading in the future - as does Google apparently, judging from recent comments they've made. If you're interested in this approach, you might want to take a read through Greg Linden's blog (http://glinden.blogspot.com/). He's very focused on personalization, and blogs about it often. In fact, he actually launched a personalization-oriented news site like you're describing www.findory.com) a while back. However, he recently put it on hold, for various reasons. So you might want to take a look around the site and read up on it (and its competitors) before you invest too much more time in your idea. HTH, DAR

posted on Thursday, April 19, 2007 at 1:12 PM by DAR


I've never been an enthusiast of social content ranking - the results are just so random. As a reader of numerous blogs (including this one), I count on bloggers (and others) to track down the best, most relevant info and to then filter, distill and feed it back out to me.

Plus, I read enough stuff already that I really don't have bandwidth to sift through yet more content - really, I can't even digest what I get today from sources that I generally find really interesting.

So, the fact that you are trying to focus the users down to a more relevant base is a step in the right direction, but unfortunately for me it doesn't go far enough that I'd become a regular visitor. Perhaps now and again when I have some spare time and I want to expand my horizons to see if there is anything great out there that I am missing.

Not to mention that like marvin I am skeptical that you can keep either the membership or the content focus tight enough to assure high relevance and quality.

Sorry to be negative, but you asked...

That said, I think some shape of your idea is potentially interesting. One thought somewhat along these lines that I might find valuable would be a thing that organically tracks the most popular posts on a variety of blogs, media sources, etc. - where this info is sorted by source, topic, etc. By 'organically tracks', I mean where citations are based on actual counts of times a post is read/forwarded/etc. vs. a scheme that requires readers to take the additional step of tagging or marking something as popular. Perhaps this is already being done somewhere. I guess this is a little Technorati-like, except that is purely business focused, ideally with an ability to personalize the topics, sources, etc. The result would be a little like my customized NetVibes page and the upteen blogs feeds I have there, except you are synthesizing and organizing things automatically for me.

posted on Thursday, April 19, 2007 at 1:18 PM by


Relevant comment thread: http://programming.reddit.com/info/1d46g/comments/c1d6v4 .

posted on Thursday, April 19, 2007 at 1:41 PM by Bill Mill


When I read about DailyHub on Small Business Hub I thought it sounded like a great idea and it appealed to the business "geek" in me. So I added it to my RSS feed to try to figure out how useful and interesting the content would be. The first day I had a look at some of the articles who's headlines caught my attention, but from there on....it has just not gotten any attention from me. Pondering on your post, this question comes to mind, "Is popular content relevant content?" One really needs to take a moment and consider that deeply. Popular thinking is not typically the thinking that leads to revolutionary businesses - normally it is the people who are thinking on a totally different wavelength that make the breakthroughs. What are those people reading? That seems to be a more relevant question than, "What are most people reading?" One of the things that I don't see a lot of talk about in general is forums - maybe they are too "web 1.0". One of the advantages of a forum is that it allows a certain level of relationship to develop due to the group conversation nature of it (yes, blogs are like this to a degree, but that is another debate). Maybe take a look at what is happening at The Forum SA (http://www.theforumsa.co.za). It is a forum for business owners in South Africa. The conversations that have happened there have had a huge impact on my business, and the relationships are priceless. I'm probably taking a relatively far-out-there attitude, and I haven't gone in to all the details of my thinking...but, hopefully it stimulates some debate ;)

posted on Thursday, April 19, 2007 at 2:55 PM by DUncan Drennan


This isn't a problem of size, it just that the technology behind Digg and Reddit haven't come along far enough. I think these type of social sites need to look at things like Amazon's "Also recommends..." and Last.fm, where your likes & dislikes are profiled and then more relevant content is shown to you. That would allow the site to scale much better with more users, instead of just becoming more generic.

I don't think a more targeted site like DailyHub really solves this problem. All it does is reduce the number of people inputting data, and it's really no more targeted than just picking a category on Digg. How would a popular DailyHub be better then a digg business category?

posted on Thursday, April 19, 2007 at 3:08 PM by Doug


You've diagnosed the problem nicely: :The content sucks because it is not relevant to us, why? selected by a large group of people who do not have the same interests with a smaller group (business geeks)" Your proposed solution is also promising: Have a site that is focused on an area, where content is selected by people who are more like us. Yet you are focusing on improving the software. The software is not why Reddit or Digg is failing. It's the community. The success of DailyHub is almost completely depend on the community that may grow around it. The community that will find the stories, post them, rate them, etc. So the challenge is how to create this community. This is the difficult part. And judging from the level of activity in the onstartups forums, I dare say, community building is not going so great. The focus needs to shift from software building to community building. It's harder :) but it's the key.

posted on Thursday, April 19, 2007 at 3:40 PM by Berkay Mollamustafaoglu


I totally agree which is the same reason why I started my site. I think that this also accounts for why we see ideas rebuilt from the ground up so often. An example would be the web browser. Rebuilding the browser, firefox, allowed a whole new generation to, in effect, dictate its direction. Again we are seeing it with social networks as new ones pop up everyday. There will always be problems just waiting to be solved.

posted on Thursday, April 19, 2007 at 4:33 PM by Raffy Banks


I think the problem is the uncontrolled population of data. It is great to see a site bloom but without ORGANIZATION the massive information is sure to be overwhelming. When we createdwww.econolution.com it was focused to a specific group, for a specific purpose, and organized by location. Now the relevance is more obvious even with loads of information. Sometimes things don't have to be recreated, when they just need to use organization that is already working, the phone book seems to serve people well by grouping by location -- so if you wanted to find community information - why not look in that community? The second thought is trust, having random users contributing without authenticating your information makes users question the source and soon pollutes the real value of content, we still believe the source has to be from a trusted member to be 100% accurate, if you want to know what a group is doing why not let them share the information directly from their site to their community!

posted on Thursday, April 19, 2007 at 5:45 PM by Clinton


It's funny that people are just starting to realize this. I have been working on a focused social content site for a couple of months now. I have come up with what I believe is a better "mousetrap" for voting on the content, not just a popularity vote. Can't say much more about it now, but I should have a beta out in June. When it's ready I'll let you know.

posted on Thursday, April 19, 2007 at 6:42 PM by Jeff


Dharmesh,
Your solution to relevancy problem (make content scope narrow) would work, but not very efficiently.

There is considerably better solution, which is applicable to unlimited set of topics. In fact the more topics are rated - the better.

Here's the idea:
If person recommends the same article as you do - chances are that you would like other recommendations of this person.

So, here's the solution:
when generate rank list for the user -- take into account how he votes for other articles.
Based on these votes personal rating of other experts can be generated.

So when personal rating list would be calculated - all experts's ratings would be included into that rating with different weights, ranging from -1.0 to +1.0.

Do you like this idea?

posted on Thursday, April 19, 2007 at 8:54 PM by Dennis Gorelik


This is a great idea. One feature i would like is OpenID registration.

posted on Friday, April 20, 2007 at 12:54 AM by Oren Golan


Oren,
What doyou mean under "OpenID registration"?

posted on Friday, April 20, 2007 at 1:08 AM by Dennis Gorelik


some suggestions: DailyHub must track only the Top Ten for the day in the home page. Additional Top Tens can be added.. and viewed in separate pages: Weekly Top Ten, Monthly Top Ten, Quarterly, Yearly and ALL-TIME Top Ten. DailyHub Recipe: Narrow focus : Business Geeks Limited Topics : 10 at a time I would subscribe to all of the DailyHub Top-Ten feeds (Daily, Weekly, Quartely, Semi, Annual, AllTime) And all of the Top-Ten keeps changing.. How cool is that? And yeah call it the "super-dynamic time weighted model".. ;)

posted on Friday, April 20, 2007 at 4:14 AM by Muthu Ramadoss


"loser generate content" - that's a total hoot! thanks for sharing.

not only catchy, but has an awkward ring of truth to it.

posted on Friday, April 20, 2007 at 4:41 PM by


+1 to what Berkay said. A strong community is vital to the success of a community-based recommendation service. Some companies are having success in this area -- Yelp being the first to come to mind. They've build a very tightly knit group of core 'yelpers' that are key to the ongoing health and success of the site.

posted on Thursday, April 26, 2007 at 2:20 PM by fiat lux


Here is my startup idea:

Instead of creating a site where people can vote, create a platform where people can start there own topic specific voting for their own interests and then promote it to their peers.

It can be hosted on the site or on the users site and will have options for differnt voting models and customization.

What do you think?

posted on Thursday, April 26, 2007 at 11:37 PM by Elisha Klein


Totally agree - the content at Digg sucks, and it sucks big!!! I went from a passionate user, to a hater, and now an ex-user.

Look forward to further experimentation in this field.

posted on Sunday, June 03, 2007 at 8:01 PM by Ryan Armasu


Dharmesh,
Sorry to take so long to get over here to address your request to evaluate DailyHub.com. I find it interesting and hope you keep it as non-commercial as I see it today...it would be a great tool for reviewing other real business people's thoughts.

posted on Tuesday, June 19, 2007 at 11:11 AM by Scot Duke


What about using social content software for more focused attempts at improving the world?  
 
For instance, there is a site that is interested in improving the way Windows Vista looks called Aero Taskforce.  
 
But you could also use this website software to make sites for your local city, such as New York City or Birmingham, and local users would log on and submit articles or suggestions and topics about their local town... and people living nearby -- the IPs could be tracked for geographical relevancy -- could upvote or downvote people's suggestions and ideas. Does this or that road need fixing? Is their blight? Is a certain street riddled with unwanted prostitution? Are some areas too boring?  
 
 

posted on Tuesday, September 09, 2008 at 2:11 AM by matthew stale


What about using social content software for more focused attempts at improving the world?  
 
For instance, there is a site that is interested in improving the way Windows Vista looks called Aero Taskforce.  
 
But you could also use this website software to make sites for your local city, such as New York City or Birmingham, and local users would log on and submit articles or suggestions and topics about their local town... and people living nearby -- the IPs could be tracked for geographical relevancy -- could upvote or downvote people's suggestions and ideas. Does this or that road need fixing? Is their blight? Is a certain street riddled with unwanted prostitution? Are some areas too boring?  
 
 

posted on Tuesday, September 09, 2008 at 2:12 AM by matthew stale


i saw another interesting site. You are able to challenge other links. to see who has a finnier link. then users vote on it.. and winner gets points and points let you get extra cool features on the website.. well amongs other fun things 
 
http://www.highcow.com/

posted on Wednesday, April 22, 2009 at 11:15 AM by plaker


Comments have been closed for this article.