Defining social media

[Summary: I’ve been looking for a definition of social media to use in research, but without much luck. So, tentatively, here is an attempt to provide one: Social media is the creation, publishing and/or sharing of content from an author to a crowd, providing a locus for horizontal interaction across the crowd. This blog post unpacks, and seeks to justify, this choice of definition]


Definitions are useful things in research and critical thinking around a subject. When studying the youth work uses of social network sites, I found boyd and Ellison’s clear definition of what constitutes a social network site to be extremely useful in giving focus to the work. Right now I’m working with Kevin Harris on an essay around the potential uses of social media in frontline public services, and one of the first challenges I’ve hit is finding an operational definition of social media.

Search the web and you will find plenty of claims about ‘what social media is’ – but they tend either towards explaining what social media is by examples (e.g. ‘social media is sites like Facebook and YouTube’) or to conflate social media with a whole host of other concepts, often in a normative way (e.g. ‘social media is the open and free sharing of content and conversation between people operating as equals.’). A good definition needs to be general enough to pick out all those things which are, by consensus, examples of social media, and to identify new examples of social media, but to also be tight enough to allow us to ask questions about the empirical and normative properties of social media (rather than assuming them). So far, I’ve not found a definition to fit that bill – so, here is a very tentative attempt to provide one.

What do we want from a definition:

A choice has to be made in advancing a definition of social media.

  • Should the definition only pick out those ‘new’ forms of media that we’ve commonly labeled social media, excluding older media by definition? Or should it allow the use of the term (social media) to refer to older things and historical experiences?
  • Should the definition be tied to the digital? Or can there be non-digital social media?
  • How will the definition set out the relationship and order of sub-terms and related terms. For example, is a social media platform something which hosts social media; or is the platform prior, such that social media is anything hosted on a social media platform?
  • Should a definition try to take in all those things that people commonly call ‘social media’, or should it advance the claim that some of the things commonly called social media are, in fact, not.

My own choices are to: allow the terminology of social media to pick out historical phenomena as well as current ones; to talk first about people’s practice, and not to tie the definition to the digital; and to advance a definition which may exclude some things commonly called social media – but generally only as a result of a rejection of the primacy of platforms in defining what is, or isn’t social media.

Offering a definition

Based on those choices, the definition below is my tentative first attempt at capturing the concept of social media.

Social media is the creation, publishing and/or sharing of content from an author to a crowd, providing a locus for horizontal interaction across the crowd.

Or, for those on Twitter, the 140 character version:

Social media=creation, publishing &/or sharing content from author 2 crowd, providing locus 4 horizontal interaction across the crowd (@timdavies, 2009)

Of course, this is still a little dense – but hopefully with some unpacking I can show why I think this captures the essential concept of social media.

Unpacking the definition

Creating, publishing, sharing content

An interaction can only be a social media interaction when there is some media. Be that a video, a photo or a 140 character tweet. I’ve used the term ‘content’ to avoid importing any connotations from the idea of media that might lead people to exclude content such as simple interpersonal messages from their understanding of social media.

[I’m not entirely sure that these three terms are the best for this part of the definition, but retain them for the time being – open to alternative suggestions. They should be read as ‘Creating [and/or] publishing [and/or] sharing’.]

From an author to a crowd

One key difference between an e-mail and a blog post is that the author of the e-mail chooses, in a deliberate and technically defined way (by e-mail addresses), who to address the e-mail to, whereas, with the blog post, they publish to an unspecified or unknown audience.

It is not necessary that the potential audience of an item of content be ‘everyone’ for it to be social media. Social media can be published or shared within a relatively closed community, but it is always published/shared with the possibility of people within the community / domain where into which it is injected ‘overhearing’ or engaging with it – even if they were not the audience the author had in mind.

If you need an analogy – think of relating an anecdote at a crowded party. You have an idea of who is in the room. An idea of who you want to address the anecdote to. But you do not limit who may listen, and you accept injections from those in the room who overhear and engage.

This author->crowd aspect of social media operates at the level of both technology (e.g. the ease of publishing to a community) and at a level of social norms (knowing that people are not obligated to engage with the content produced, but allowing that they can).

A locus for interaction

Social media involves the possibility of interaction. But that interaction need not be described in terms only of some technical functions (e.g. comment boxes and rating boxes). Rather, in social media, content has the potential of becoming a social object around which interaction can be organized – and this can happen in at least two ways (which are not mutually exclusive):

  • The platform through which the content is published allows for or enables comments and interaction;
  • The content is licensed in ways that allow people to share / remix and engage with and around it in active, creative ways;

The interaction need not take place in a single location, or on only on platforms that call themselves ‘social media platforms’.

Horizontal interaction across the crowd

If the only interaction possible around content is between the author and individual members of the ‘audience’ (vertical interaction), then the potential social interaction around the content is highly constrained. In social media, there must be the possibility of audiences of content interacting with each other around or through the content, with or without reference to the originator of the content.

The requirement that the potential must be for horizontal interaction ‘across the crowd’ distinguishes cases where content is broadcast into multiple small ‘crowds’ where it becomes a social object (e.g. a TV programme watched & discussed in living rooms across the country) within sub-units of the whole crowd, from circumstances which can tie together interaction from right across the crowd who constitute the potential audience of the content (e.g. the use of a hash-tag on Twitter to discuss a broadcast media programme or a weather event).

This does not mean that social media necessarily equalizes people – or that all interaction around social media content is horizontal and peer-to-peer in character. But it does suggest that without the potential for horizontal interaction around content, that content is not social media.

Extensions of the definition

The definition unpacked above is essentially the definition of a process (creating, publishing, sharing content) under certain conditions. But from these we can derive a number of further definitions:

  • Social Media Content – content is social media content iff it is created, published and/or shared from author to crowd in a way that can provide a locus for horizontal interaction across the crow
  • A social media platform/tool – is a platform or tool which, to a significant and noticeable extent, intentionally, or unintentionally, facilitates the creation, publishing or sharing of social media

  • Etc.

Testing the definition

I’ve tried in composing this tentative definition to apply a number of tests to check it’s utility. Example tests check if it rightly rules in, and rules out, certain examples of things that may or may not be generally considered to be social media. Question tests check whether the definition can be used to guide substantive enquiries into social media without including the answers to interesting questions in the definition itself.

For example tests, I believe that:

  • This definition adequately rules in as generally social media: YouTube, Facebook & Twitter. I welcome other suggestions of examples to test.
  • This definition rules out TV, E-mail, Telephone Conversations and Podium Speeches as not being examples of social media. It also rules out use of tools such as YouTube solely as media publishing platforms when all interactive features are turned off. An online video with no interactive features is only made into a social object when shared by someone who adds opportunities for interactivity to it – in which case the ‘social media’ consists of the original non-interactive video, plus the sharing of it in ways permitting horizontal interactivity across a crowd.
  • This definition would rule in content-mediated discussions at an unConference of BarCamp; and the intentional facilitation of a participative workshop using media content – be that multi-media or paper-based media.(I expect this set of examples to be more controversial – and ones that go beyond most people’s commonsense ideas of social media

For question tests, I believe this definition should facilitate the answering of questions such as:

  • Is social media a democratizing force?
  • How can social media be used in front line public services?
  • How do specific examples of social media structurally differ?
  • What properties of social media contribute to collaboration?
  • How can social media contribute to greater community building, rather than to greater individualism?

For questions about the properties of social media – the definition does not only pick out examples of social media, but also gives a framework for assessing different properties of social media – but I feel this is a justifiable bit of additional work done by the definition in this question context.

Use and development

This blog post (at is the first attempt at forming and sharing this definition. I will be giving it a practical test in at least one upcoming project – but wish to subject it to a wider critical test.

Perhaps, for purposes other than my own, it is a non-starter – and a definition that cannot achieve some wider than individual acceptance is little use at all. But I hope it can prove useful (in current, or a revised form) for others.

(All comments and feedback; pointers to other works etc. are welcome. Thanks to all who have contributed to conversations on this topic with suggestions of links to follow, or pointers to existing definitions also, and apologies that version 1 this blog post, written whilst I’ve been without Internet access, does not offer specific references and credits.)

10 thoughts on “Defining social media”

  1. Hi Tim,

    I’m really enjoying your blog – been reading some this evening.

    I’m currently doing a lot of frameworks with Social Media and your systematic approach is similar to mine and what I think is needed in the industry (albeit, you are far more academic than me!)

    Will be sure to keep in touch.


  2. Just noticed that this rules in mailing lists (e.g. Google Groups / Majordomo lists) as being ‘Social Media’ in most cases.

    This wasn’t (to my fault) one of the tests I subjected the definition to – and I’m wondering if this is acceptable. Should e-mail lists qualify in most cases as social media? Would welcome intuitions of others on this…

  3. Some useful discussions on Twitter I’m logging here to get beyond the 140 Character limit:

    In reply to: fairsay @timdavies A challenge! I consider all collab=social media since my defn is that social media = sharable ‘content/state’ (public or private)

    I’d agree that, any use of media between two people for collaboration is a social use of media. But I would argue that there is subtle difference between ‘a social use of media’ and ‘social media’. The difference is structural – and in terms of the addressing of the message. It seems that for mediated communication between two people, or between a set group where all are expected to interact with it, would be best afforded a term such as ‘interpersonal media’, whereas the use of media ‘from author to crowd’, that is, from an individual to a loosely specified audience or set of collaborators, is what should earn something the designation ‘social media’.

    (I apologize for any pedantry. I’m trying to explore and defend the definition above for an upcoming project – and very grateful for all comments that are prompting me to examine and reflect on it more…)

  4. I’m enjoying your challenge (which I only picked up on by chance in the continuous flow of twitter as I am not watching it 99.9% of the time).

    I think I was intrigued by the idea of even trying to create a definition which I view as always resulting in an inadequate final result but having an enlightening process!

    I see ‘author to crowd’ as an over-restrictive, specialised characteristic of ‘social media’ and just one of many cases in which it can be used.

    For instance, radio, TV and print media is ‘author to crowd’ but I would not call it social unless it is live in a public place (in which anyone there can interact in various ways), or through phone-ins or other methods that ‘complete’ the circle of communication. Mass emails also fit this.

    Some large scale events (e.g, Make Poverty History, Dr Who final episodes, big popular protests) can be social as people talk and share about them online and offline. In these cases it isn’t the media that is social but the experience of a common event.

    Looking ‘social’ up on Wikipedia doesn’t help (as I found out) since it has different meaning to different people in different contexts. What you seem to be trying to do is create a ‘systhesis’ of what is out there and apply it to the phrase ‘social media’ (which is a valid but difficult approach!).

    So the ‘author to crowd’ approach doesn’t convince me 🙂

    You want an approach that does draw the line (as artificial as most lines are) between some types of collaboration vs. other forms of sharing. In thinking about it, would it help to view social media as ‘media where the opportunity / potential for it to be amplified or transformed by group behaviour’? If so, you may wish to look into social psychology for inspiration too!

  5. Hey Duane

    Many thanks for joining in. I think my approach heres is mainly structural – picking out features of the structure of things generally regarded as social media, and using the line that draws, and the features as variables (i.e. who is the author, what is the nature of the crowd) to ask questions about the social impacts of social media, without assuming any such social impacts in the definition itself. As you say – the interest is mainly in the process of drawing the definition – rather than the result itself, which is ever adequate, will only be so across a limited set of purposes.

    However, I think the full definition I’m working with:

    Social media is the creation, publishing and/or sharing of content from an author to a crowd, providing a locus for horizontal interaction across the crowd.

    in talking about the ‘locus for horizontal interaction across the crowd’ partly responds to your objection – as that would exclude radio, TV and print as required. It would also, probably exclude the talk about large scale events – but only because those are social experiences, rather than social media – of which there can be media that becomes social media within specific localized crowds.

    I like, however, the reframing as “media where the[re is the] opportunity / potential for it to be amplified or transformed by group behaviour” which seems an altogether more user friendly definition…

  6. Hi Tim,

    Glad you like the counter proposal 🙂 Of course, I’m not looking into it in-depth from an academic perspective (as you are) but more from an intellectual curiosity perspective since the process is more interesting and useful than the result!

    Thanks for the provocation and good luck with your work. (and maybe see you at ECF 2010 in March since it is in Oxford and you are at OII in Oxford I believe)

  7. Hi Tim, thanks for pulling this altogether. Makes a really interesting post. I’ve been thinking about similiar issues with some work I’m doing looking at the relationship between volunteerism and social media.

    I think it’s incredibly challenging to define social media not least because it’s a concept that’s really challenging the conventional meaning of some of the base concepts on which it’s building, such as: what creation is (in a world of digital reproduction) or what publishing is when the line between public and private spheres are blurring to they extent they are with the advent of social media.

    Anyway, I just wondered whether it’s worth trying to unpick more what ‘providing a locus for horizontal interaction across the crowd’ is. I think your definition is fresh and interesting because of this phrase.

    Part of what’s allowed social media to emerge and earn the tag ‘social’ media is because it’s cheap, quick and simple to participate in and facilitate participation. Social media overcame the hitherto constraints of money, time and knowledge that you previously needed to access many types of media.

    But I’d add that the real significance of social media is as much in it’s potential to be social, its potential to be a locus for interaction should the crowd be able to participate.

    Traditional media just didn’t have the technological infrastructure to make horizontal interaction possible even in theory. Nowadays, any post on any discussion board provides a potential locus. This potentiality changes the nature of the media significantly even before anyone actually takes up that opportunity to post (and transform a creation into a locus for interaction). Social media is as much about providing a potential locus, as it is about providing an actual locus.

    Btw – I think email groups have to be included in the definition. In many ways they are a very pure form of social media with their roots in such a basic and universal protocol as with email. I’d personally be tempted to draw the line at digital media (for purely pragmatic reasons), as the arguments around the social merits of non-digital media often take the discussion down (interesting) but ultimately off the topic side roads.

    Cheers, Patrick

  8. Hi Tim!

    Thought I’d share these links with you (presuming you already haven’t come across them!


  9. Oops! The link wasn’t displayed earlier

    Chk out ‘Laura’s presentation’ on Slideshare



Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.