by Gerrie Coertzen and Daniel de Bruin
The internet is now mostly social and social media platforms are fast moving towards a gated ‘ecosystem’ approach, where updates and news stories are no longer equally shared between user accounts. These new ecosystems are highly controlled platforms where updates are collated and selected by a small group of employees, acting as online censors, which results in the filtering of information feeds.
As we discussed in part 1 of this series of articles, the internet is changing fast, and the way information is shared is becoming much more controlled.
Facebook recently launched their new Facebook Paper app, which shows users only news items and user updates which have been pre-selected for them. It is not yet clear whether the amounts paid by certain newspapers will affect which news items a user will be shown. Similarly, Google has integrated their services to such an extent that user data now seem to flow seamlessly between, Gmail, search, Google+, YouTube and other Google products.
Quality, not equality
Online searches, comments, reviews and social media interaction creates an enormous amount of data. Users have many online connections, ‘followers’ or ‘friends’, and create many updates, tweets, and shares every day. Social media platforms are starting to selectively filter what is shared, how items are shared, and who can share. All this is done in the name of quality – however, the equality of new updates is sacrificed in the process.
To understand how much data is created, we can compare the numbers for a tiny proportion of this data. Around 150 billion email messages are sent per day, and over 2 million searches a minute are executed on Google. On average, more than 70 hours of video footage is uploaded on YouTube every minute, more than 600,000 bits of content are shared on Facebook per day, and over 340 million Tweets are sent and received every day.
To any user, most of this information is of no interest, and it is becoming very difficult to filter through all the updates and shares to get to the meaningful, or interesting content. Social media platforms are now trying to solve this problem, by filtering updates on a user’s behalf. The danger is that the filtering process is no longer within the user’s control, and can be prone to censorship. All updates are no longer equal.
Are you following me?
Many people are now starting to carry more location-aware devices, and this can be utilised in marketing research such as buyer behaviour, and in conjunction with other sources of data, can help to spot trends.
Using data analytics in social media, organisations can establish where there is a need for a specific product or service, if people in specific areas are more prone to buy or use a product, or if specific factors contribute to increased customer satisfaction, recommendations and sales.
Targeted advertising
Analytics of social media activity (social signals) are used to deliver more targeted advertisements to consumers online. The information is used to deliver advertisements that are more relevant to the individual consumer and location-based services can help to push those advertisements to consumers at the right time. For example, if someone is an avid swimmer and has shown an interest in Arena’s Viper goggles, a targeted advertisement or coupon is sent to her mobile when approaching the Arena retail outlet. Or, the swimmer’s friends can receive targeted advertisements about Arena’s products when her birthday is approaching.
Combining social media analytics with other sources of data
Customer sentiment on services and products purchased based on reviews and social shares such as ‘Likes’, ‘Tweets’, ‘Favourites’ and ‘+1’s’ can provide a much better understanding of purchasing behaviour. This data can also be used to collaborate on sales forecasts, where items should be stocked and when.
Data from social media activity during TV programmes, allow marketers to understand their audiences and the impact of their campaigns better. The BBC can immediately see which programmes get more audience participation and can plan their broadcast schedule better. Other channels and media houses can use this data to optimise advertising slots to charge more for delivering content to targeted audiences. Netflix uses its algorithms to anticipate which programmes will be most successful as it did with ‘House of Cards’. Data from many different sources, including social media, was used to determine the popularity of actors, the relevancy of story lines and how viewers would prefer to watch programmes.
Personalised search
Google also uses social media analytics to provide a more personalised search experience. Based on the user’s likes and comments, and those of their friends, more relevant search results will show up on their search engine results page (SERP). Google uses this to reduce the amount of SPAM advertisements and comments. For example, when using Google Plus and YouTube, users will see the comments of their friends and other people in their ‘circles’, or from contributors they have liked in the past, ahead of more generic feedback.
In the next article in our series, we will discuss the new ways users and apps are approaching privacy and how users can secure their data.
Gerrie Coertzen and Daniel de Bruin are directors at Modelling Design Partners, a business intelligence company implementing the latest techniques in data analytics and machine learning.
