We've touched on various aspects of propagation in preceding sections, however, before we look at some of these in more detail we think it's important to consider social media in the wider context of an organisation's marketing and communications activity.
As we've already mentioned we believe that social media should not be considered as a separate ‘silo’ of activity. In fact we recommend a ‘social first’ approach to all campaign planning. That doesn't mean that we think every campaign should be run on Facebook or Twitter, far from it, however, it's important to think how any communications activity might work or be propagated across social channels from the outset of any campaign. ‘Social’, like ‘community’ is not something you can add at the end of a traditional planning and creative process.
Social media is a two-way channel and from a practical perspective you need to be prepared to deal with this. Obviously, you need to address issues around moderation, tone-of-voice and channel management before launching any major campaign. However, it's also important to understand that any network communications activity takes time to work. So, for example, Facebook propagates messages using an algorithm called Edgerank. Crudely speaking, Facebook rewards ongoing conversation - the more you say and the more your network says about you, the more your content will appear in other people's Newsfeeds. It is therefore vital that any campaign should become part of an ongoing program of engagement, and that any campaign should be proactively managed with moderators commenting and responding to any activity as this is not only picked up by users, but by algorithms like Edgerank.
Some parts of the communications industry have adopted a paid, owned and earned model for media planning. While we've found this model helpful to a degree, in our experience the term ‘earned media’ can be unhelpful as it tends to be used as a proxy for social media, which in turn tends to suggests that social media doesn't cost anything - that it's something you can get for free.
Social media is not free and it takes time and effort - you really have to earn it. The upside is that if you get it right the results can produce the sort of exponential returns that traditional media simply can't. You earn a lot more.
In our opinion, social media campaign budgets should be comparable to traditional digital campaign budgets. What differs is the emphasis of spend:
budget should be diverted from traditional advertising and media spend to content development, partnership and channel management/moderation. It's important to note that this is, to some extent, a question of emphasis, we always recommend that any campaign is supported by some targeted media spend and some outreach work.
The hub and spoke model of digital integration was first outlined by the team who developed the digital and social media campaign that was widely credited with helping propel Barack Obama to the US presidency in 2008.
In the hub and spoke model, the hub is the organisation's website - something over which it has complete control and the spokes are social networking and other social media sites - places over which the organisation has incomplete control. In this model the spokes are mainly used for propagation, reach and prospecting, the hub (in that case MyBarackObama) works lower down the marketing funnel at the level of desire, action, conversion and also post purchase loyalty. The distinction between functions is by no means absolute, it is entirely possible to generate revenues on social media platforms, and generate awareness on a corporate website. In our experience it tends to be intelligent integration of owned digital media with social media channels that produces the best results.
For example, the VW Golf driving game on the iPhone reportedly drove over $4 million in direct sales from the app as well as attracting millions of downloads and tens of thousands of ratings. Social media propagated the message on the spokes and the hub converted prospects.
We are sometimes asked by clients if there is any point in retaining their own website. In almost all cases we'd advise a client that they should retain their own site. As we've seen with Bebo and MySpace social media channels can fall from favour almost overnight. In addition, the user experience on social media channels is often poor and changes to functionality which are out of an organisation's control. Perhaps most importantly, an organisation's website is a place in which they can develop a one-to-one relationship with a customer, a place where they can understand their needs and behaviours in more detail.
Social media is becoming an increasingly important part of the digital media mix, but we always argue that social media spend should be incremental to existing digital spend, with budget derived from traditional brand advertising spend to pay for new social media campaigns. It's not a question of a website or a social media presence, most organisations need both.
At the time of writing Facebook has over 850 million users and rising, and while there are signs that Facebook usage is dropping in core markets it seems that for the immediate future at least Facebook is one of the most important destinations on the web (for example, Facebook currently accounts for nearly 1 in 4 display ads in the US).
This has lead to a perception that Facebook has in some way ‘won’ the battle of the social networks and that every social media campaign should be run on their platform. We don't believe this is the case. Facebook is an integral part of many of the campaigns we run, however, Facebook is not a great place to reach influencers, it is the place to reach mass audiences. And it doesn't just stop at Twitter either, Tumblr is currently the most influential social platform in the fashion industry, many of the biggest brands and media owners have adopted the platform, however, it has very limited commercial traction in most other categories. At the time of writing many of the web's key influencers seem to be using Instagram and Quora, in addition Google Plus is experiencing some early traction. And this is important because of the reasons we outlined earlier. Influencers are the key to propagating the campaign.
Secondly, the nature of engagement on different social media channels can be quite different. Facebook tends to be a channel for communicating with close friends and family, whereas Twitter is more of a mass personal broadcast tool.
In addition different social media channels have different demographic and sociographic profiles Twitter users are older, more affluent and better educated than those on Facebook.
The best approach to channel selection is often a function of the creative idea and the strategic target. When the primary channel has been selected, it is then important to understand its relation to other channels. So it's important to use other platforms like YouTube, Tumblr and Twitter to propagate content and maximise the reach of a campaign. The problem with this approach is that the cost of management can outweigh any benefit. However, tools like RSS and Twitterfeed can help syndicate content across multiple platforms and maximise reach with a minimum amount of effort.
Finally, if you want more control over the user experience of a campaign than the social media platforms allow, then the Facebook Social Graph (or Connect as it was once known) integrates Facebook login into an external site while allowing a site owner to access various features of the platform and user profiles. Other social networking services like Twitter offer similar functionality. Bringing social media functionality into a site can be as powerful as taking content and functionality out to the social networks.
A blogger outreach plan should be part of any social media campaign. However, in our experience it should be a relatively small part of the propagation strategy and not the main focus. There are reputational risks involved in badly managed outreach campaigns. And as we've already explained we believe that partnering is a more effective and authentic way of propagating messages across a network. However, a small and targeted piece of blogger outreach is important to start any campaign. There's more detailed advice in this short introduction to blogger outreach.
That social media has delivered great return on investment for a range of organisations is undisputed. For companies like Zappos it is a clear point of competitive advantage. And there are similar success stories when it comes to the world of marketing communications. The social media component of the Old Spice Guy campaign raised sales by over 107%.
However, as we noted in the introduction there is no established, generalised ROI model for social media either as part of a marketing campaign or as part of a wider business initiative. There are three main reasons for this:
However, it is clear that a social media campaign can deliver enormous amounts of value to an organisation. And while it might take some time to develop more predictable campaign and ROI models, that does not mean social media campaign activity should not be measured. In fact, measurement is key. It is only by experiment that we can develop more predictable social media ROI models.
We recommend that any organisation considers the following four steps when it comes to developing an ROI model for social media campaigns.
There are a wide range of tools available to measure social media sentiment from enterprise solutions like Radian 6, Sysomos, and Visible Technologies to smaller scale and free solutions like Social Mention and Hootsuite. License models vary along with capabilities. And there is no one ‘best solution’. It is important to understand what any tool is monitoring, Twitter for example sells different levels of access to data to different providers, so it may well be that you are only getting a representative sample rather than a full picture of any campaign. And while this is less of an issue when measuring sentiment it is of course extremely important when trying to measure reach. However, sentiment analysis tools are not enough on their own to accurately measure social media activity.
For an example, we ran an agency promotional campaign earlier this year, which featured a picture puzzle of 20 things that happened on the Internet in 2010. The picture appeared on the front page of Wired, Gizmodo, The Telegraph and dozens of other influential blogs and community sites. At the height of its popularity it was generating over 20,000 unique visitors a day to our site. However, most of the action happened elsewhere, as you can see from the Gizmodo link above, they had over 60,000 views on their US site alone, and over 150 comments. Similar results were replicated across about 150 blogs. It had thousands of tweets and retweets, but again these can be hard to track down as people were tweeting and retweeting other people's URLs. Furthermore, we got lots of traffic from Tumblr, and various reblogs of the content were themselves reblogged hundreds of times. There is no way of telling how much traffic and awareness this generated on that platform. Sometimes you simply have to make an educated guess at the ‘propagation effect’ on other platforms that cannot be tracked.
So you can see, even for a simple internal agency project there are lots. We counted over 450,000 directly measurable impressions, however, we estimate that the real reach of the project was over twice that number - we cannot count the impressions on blogs or communities that don't refer back to the original picture.
Measuring social media activity is complex and the total number of engagements is often an aggregate of a number of different metrics - comments, shares, likes, retweets, reblogs, etc. It is therefore important to do some manual checking and investigation. Check website referrers and follow the links. Use traditional search engines to track down content. Follow the trail of users back from content sharing communities like Reddit and Tumblr.
It will only become apparent over time which benchmarks are the most important for your organisation. Therefore, in the early phases of any social media activity it's important to capture as much information as is possible so that significant data points can be identified. These can then be used as part of an internal benchmarking process that can in turn be used to develop a return on investment model.
It is tempting to expect social media channels to deliver results experienced by some of the outliers and early adopters. External benchmarking should be conducted against direct competitors, which will give a better sense of progress made.
Social media activity can deliver a wide range of value to an organisation. It is important to understand the different types of engagement and measure them accordingly. As we've explained in the section on participation it's important to understand which metrics measure depth of engagement, and which measure reach, and use them to develop a model which values each of these. Don't judge the success or failure of a campaign based on the number of creative participants alone, because unless you're Nike or another high-engagement brand, this number will almost always be in the low hundreds or thousands. Shares and recommendations, along with more traditional digital metrics like page views, unique visitors, and dwell time help paint a much broader picture of the impact of a campaign.
It is therefore important to integrate social media campaign tracking into the broader brand and campaign tracking work, so that brand sentiment uplift as a consequence of social media activity is correctly attributed. If you want to develop direct attribution models it is necessary to create direct, trackable calls to action. This is entirely possible to do, but not always desirable.
Furthermore it's important to understand all the value that social media campaigns deliver to a business. As we've already noted all digital media is becoming more and more social. So for example, Google and Bing now rely on what are often referred to as social signals to rank pages in search. There is no definitive information on how significant social signals are, however, it is thought that in certain contexts around 40% of Google Page Rank is determined by social signals. And this is a trend that is only going to increase. (Indeed there are those, like Eli Parriser, who worry that social filters will compromise our cultural objectivity). This means that as much as 40% of your SEO budget may well hinge on your social strategy, a value that should be attributed.
And as we noted in the previous section on content, social media content has a life beyond the campaign. Good content can go on driving traffic and engagement for years. In our opinion this often gets missed when trying to develop ROI models for social media which only look at value over the life of a campaign (because that's how traditional paid media advertising worked).
Finally, in recent months, it's become apparent that the Facebook Like button is becoming something of a focus for campaigns. Campaigns with explicit mechanics to drive Facebook likes have two major points of appeal:
However, while it's tempting to focus campaigns on the Facebook like we don't advocate this as a general strategy. For example, campaigns that ask for a like in return for some content for entry into a competition have limited reach - in our experience there needs to be a very clear exchange of value for fans to like a campaign or brand to access content.
These recent figures on the influence of the Facebook like button come from Facebook themselves so it's worth treating them with some scepticism, however, it's clear that Facebook likes do drive sales, however, opinions differ widely about the actual value of a Facebook like. From our experience the value of a like will vary between brands and over time.
Flickr some rights reserved by gurdonark