Applying Social Network Analysis to Social Media


sna-map.jpgSocial network analysis has been a part of social media analysis (not the same thing) for a long time, but it hasn't been central to the social media discussion lately. Mostly, SNA shows up in the form of link analysis, which is used to identify online communities and influencers. A recent conversation on intelligence applications of social media data got me thinking about how much more could be done with the many expressions of connections online.

Looking for less obvious connections
Link analysis is relatively easy work, since the data you're looking for is helpfully encoded in HTML. Follow the link, map the connection, and continue. But think about all of the other connection data that is being generated, and how it could be used to map social networks or model influence in the real world:

  • Explicit social graph data
    Sometimes we make it easy, by making our connections on sites like Facebook and LinkedIn visible to the world.

  • Follower/following
    Twitter follow connections are probably weaker than other social network connections, but these connections are mostly public. Asymmetrical follow tells you something different about the relationship.

  • @replies
    Probably weaker than a social network connection, but stronger than a follow. @replies indicate some level of active connection (which may be one-way).

  • References in text
    A mention of an article or book may not include a link that a crawler could follow, but it's still a citation.

  • Mentions in text
    References to people, organizations, and topics within the text of a post. The text might even describe the nature of the connection (e.g., "my friend Bob," "Bob, my former boss").

  • Sharing
    Bookmarks, likes, and other sharing services provide another source of links from identifiable parties.

  • Book reviews
    What do you read? Which authors? Who comments on your reviews? Are your reviews voted up or down?

  • Community membership
    Besides direct connections with individuals, we're joining discussion forums and online communities, which connect us to other members.

  • Forum posts
    Active engagement in a community is a signal. Comments on a common thread suggest a connection, or at least common interests.

  • Blog comments
    Commenting on a blog indicates that you read it (unless you're a spambot).

  • Check-ins
    Check-ins reveal where people go. Who else checks in at the same place? At the same time? What about accidental check-ins?
The big picture
Each of these sources is connected to an entity—a user account that belongs to a person or an organization. If you can identify the same entity across multiple services, then you can build a more complete picture of that entity's connections. The differences between types of connections might lead to a deeper analysis of the network, too.

As social becomes a feature of seemingly everything online, the potential to use SNA to build richer analysis only grows. Social media are giving us many opportunities in indicate our connections, both explicitly and implicitly, constantly adding to the public data pool. Whether this is more of an opportunity for analysis or a threat to privacy depends on your point of view.

Image by Marc Smith.

This is one of those posts where the probability that you'll comment is inversely proportional to the probability that this idea is useful to your work.


Great post Nathan!

I'm actively working on projects such as this and and I believe it has huge potential in the areas you mention above...and more! Everything has a connection to something. If there's a connection, it can be mapped and analysed.


Hi Nathan, really happy to read this post :) ! At linkfluence we're focusing on social network analysis since 6 years and we put this activity at the center of our r&d and services development. SNA helps you to understand heterogeneity, diversity, complexity of social media, it is a way out the googlarchy, the flat ranking of all monitoring tools and shows that you're not influential everywhere but only in a context.

A quick example of a really concrete use of social network analysis in the field of fashion over Europe :

All the best !

Great ideas here, I tend to follow the same ideas when it comes to analysis, finding those deeper connections can lead to some great connections for your business as well.

Really useful ideas and for free :), thanks to the author for his kindliness!
I am pleasantly surprised to come across with explicit indices and metrics which (according to the author) should be focused on in order to analyze social networks and connections among SM users.

To my mind lots of SMM vendors are looking forward to enjoing the benefit of the connections matrix evaluation and social graphs measurement. But they don't have consistent approach and single understanding of the problem. Furthermore this question bore ethical issues as it was partially described here.

All the best,

Hey Its really good job buddy you are giving the clear cut ideas on Social Network Analysis which has become the backbone of our society .
Its also giving the opportunity to create communities and through this people are increasing the memberlist of their community

Comments are now closed for this entry.

About Nathan Gilliatt

  • ng.jpg
  • Voracious learner and explorer. Analyst tracking technologies and markets in intelligence, analytics and social media. Advisor to buyers, sellers and investors. Writing my next book.
  • Principal, Social Target
  • Profile
  • Highlights from the archive


Monthly Archives