Of all the windows through which a business can peer into an audience, social media seems most enticing. The breadth of subjects, range of observations, and, above all, the ability to connect and draw inferences make social analytics hugely exciting for anyone who is interested in understanding and influencing past, present and potential customers, employees, or even investors.
As individuals leave traces of their activities - personal, social and professional - on the internet, they allow an unprecedented view into their lives, thoughts, influences and preferences. Social analytics attempts to draw useful understanding and inferences, which could be relevant to marketers, sales persons, HR managers, product designers, investors and so on. Thus, as social tools like Facebook, Twitter, LinkedIn, WhatsApp, and many more, host a plethora of social activities of many people, a humongous amount of data is generated about people's preferences, behaviour and sentiments. Like any data, it is amenable to analysis to gain useful insights.
The challenge comes from the sheer volume, velocity, and variety. It is very difficult to ensure that the analysis is relevant and reliable. Besides the daunting technical intricacies of setting up the appropriate analytics, the aspects of choosing information sources, filtering the right data, and its interpretation and aggregation are susceptible to errors and biases. For example, some social activities are relatively easier to access (like activity on Twitter, or public updates on Facebook), many are not. Some types of data (like text, or location) are easy to search and interpret, many (like pictures) are not. So a good analysis model must judiciously compensate for the nature of the sources included, and hence it could be at times very difficult to assess if the analysis is useful or just meaningless mumbo-jumbo.
The first and the most critical step in a social analytics initiative is to identify the business goals that the analysis might contribute to, and the key indicators that relate to those business goals. Typical objectives include increasing revenues, reducing customer service costs, improvement in products and services and influencing public opinion. Some key indicators that may be relevant in a business context could be, thus, audience reach, customer engagement, product related sentiment, public awareness and acceptance, etc. An important step would be to see if these indicators can be measured, at least in parts, in terms of inferences from social media. For example, customer engagement might be measured by the numbers of followers for a Twitter account and numbers of retweets and mentions of a company's name. At the same time, for a company offering specialised products for an audience which is not active on Twitter, this could be irrelevant.
As marketing is increasingly shifting online and digital and social marketing is becoming a bigger part of the marketing efforts, the most common use of social media analytics is to mine customer sentiment in order to support marketing and customer service activities. Social analytics can help listen and analyse customer interactions about products or services and address customer concerns promptly. As the tools for social analytics allow for assessment of the impact of most of the business's social media activities, these initiatives can be optimised for maximal impact.
More so, we can track conversations to identify leads and business opportunities. A Twitter interaction about, say, problems with a competitor product, could lead to a sales opportunity. In B2B scenarios, social analytics can lead to highly valuable insights and profiling information about an account. There is generally a wealth of information that can be gleaned about a company by looking at the activities of various individuals in that organisation's ecosystem - employees, suppliers, buyers and so on. Activities on Twitter can signal new initiatives, hiring, projects, or investments. The job postings, people profiles on professional networks can provide information about products being used, extent of usage, etc. Blogs, community postings etc can reveal about the influence levels. All of these can lead to identification of specific product opportunities and purchase drivers. We have built sophisticated processes, which build on domain and product knowledge to listen, organise and analyse this data and enable and accelerate sales by providing sales teams with relevant information about their key target accounts, thereby exposing new business opportunities.
Social listening and analysis has to be an ongoing process, requiring specialised skills and investments. We have, at Advaiya, built and maintained social dashboards for products and companies. These dashboards allow an aggregated view of the relevant happenings on a social platform leading to recommendations for social media intervention (for example, to respond to an influential negative chatter), or for product development, or for marketing (like, what messaging could be more relevant, target group characteristics etc). Companies with highly complex and specialised products can also get deeper market intelligence based on the professional activities of the target audience leading to actionable information about the relevant business opportunities and threats, thereby implementing relevant marketing tactics to engage and convert these into tangible business.
Let me add that social analytics is highly relevant for individuals also. For anyone serious about building her online presence, social media analytics tools are essential. It is important to know, for example, how many people did your post reach or what sort of links do your followers like best, or, even, does anything you do online even matter? There are a range of analytics tools available where one can assess and get a score, about one's social presence.
For all its potential, the challenges with social analytics may not be completely addressed. The important issues of privacy and reliability remain. Use of personal data for targeted marketing can be construed as violation of one's privacy. More so as seemingly know-all messages litter (in many cases, distastefully) our screens. A related risk of hyper personalisation is alienating customers or creating negative opinions, if messages are wrong (after all, any analytics is prone to some errors) or intrusive or not sensitive to the person's context.
Trust in social analytics must also be tempered with recognition of the divergence of online and offline lives. The 'real' personality could be very different from what is indicated by the online persona, aspirations and behaviour. For many businesses, substantial portions of the relevant population may not be generating adequate social data. Care must be taken whether data pulled from social media sites is credible enough to influence business decisions.
Used correctly, social analytics can yield enormous benefits to an organisation's goals. It is critical that the models correspond to the business context and the intervening mechanics are accounted for.
Founder & CEO, Advaiya Solutions