Failure to invest in sentiment analytics could lead to brand damage

Early adopter companies in retail and investment banking are exploring social media sentiment analytics

Hasan Syed is every company's worst nightmare. If he's not happy with a product or service he's received, he complains. And if he doesn't like the way his complaint has been dealt with, he's prepared to take it one step further, as British Airways (BA) found to its cost earlier this year.

Increasingly frustrated at the perceived lack of assistance he was receiving from BA's customer service department after the airline had lost his father's luggage, like thousands of other consumers, Syed took to social media to vent his spleen.

But he wasn't satisfied with a derogatory tweet that would merely be seen by his modest band of Twitter followers. So Syed took the drastic step of putting his complaint into a paid-for promoted tweet that could potentially be seen by thousands of people. 

His message – ‘Don't fly @BritishAirways. Their customer service is horrendous’ – quickly went viral receiving thousands of retweets and Twitter comments, in the process generating global media coverage. It took four hours for BA to pick up and respond to Syed’s tweet, by which point the story had already spiralled out of control.

More on social media sentiment analytics

If only the airline had invested in the latest software that’s proving a godsend for marketers and customer service departments alike. With millions of conversations taking place across channels such as Facebook, Twitter, YouTube, LinkedIn, blogs and other online communities, it’s nigh on impossible for companies to physically keep tabs on everything that’s being said about their services or products. 

However, businesses such as Thomson Reuters, IBM and Xerox have invested significant sums of money developing social media sentiment analysis software, which is capable of accurately determining the sentiment of comments, allowing companies to identify potential problems before they escalate.

The automated nature of these platforms is crucial, according to Tong Sun, who heads up the data analytics laboratory at Xerox’s research centre in New York.

“When humans have to step in to evaluate the context of a tweet or route a post, it slows things down and reduces the overall value or social media data,” says Sun. “We’re piloting a platform that lets computers do all of the heavy lifting.”     

This latter point is the key advantage of these types of platform, argues Amanda Sinclair, director and co-founder at Watch My, which provides a “self-service, business intelligence portal” on the web that gives users the ability to monitor social media channels, websites and news channels via the company’s SaaS platform.

Crucially the cloud-based service, which is used by a number of leading FMCG (fast moving consumer goods) companies, automatically emails users a couple of times a day to alert them about activity surrounding their company or a competitor – an important function given how quickly things can spiral out of control on social media.

“Every day Twitter users are talking about brands, asking people for recommendations and praising or complaining about companies,” says Sinclair. “At the very least companies should be keeping an eye on Twitter for mentions of their brand name so that they know if people are talking about their company.”

Social media sentiment analytics nascent

Every day Twitter users are talking about brands

Amanda Sinclair, Watch My

But it’s not just a case of monitoring these conversations. You also need to act on what’s being said, advises Kirke Lovesey from IBM’s business analytics products and technology expert team. He cites the example of a high street retailer whose clothing products were attracting derogatory comments on social media. Thanks to the use of social media sentiment analysis tools the company was able to identify an issue that might otherwise have gone unnoticed and rectify it.

“The thing about social media sentiment analysis software is that it allows you to find out what you don’t know,” explains Lovesey. “On these sites people feel anonymous so you get a much more honest opinion.”

At the moment, Lovesey says this type of software is still in its infancy and, because there are lots of different products on the market, there’s a lot of confusion about what system people should opt for – the fact that most of the products are offered on a SaaS basis, with purchasing decisions made by marketing departments or senior management teams rather than IT departments, might also be adding to the confusion.

“When you go into a large arena like this there is a lot of noise that you have to filter out sensibly,” says Lovesey. “It’s a complex science. For example, it can be tricky to detect sarcasm rather than sincerity. The only way you can generate decent sentiment analysis is by creating a pretty hefty dictionary and then putting syntax expressions on top of that. That’s the sort of functionality IBM offers.”

Investment banks turn to sentiment analysis

Early adopters of social media sentiment analysis include retailers such as supermarket chains, high-tech companies and the financial services industry. Another area where these tools are increasingly popular is the world of high finance – not that you’d know about it.

More on social networks

James Ross, founder of social media stock sentiment trading analysis software Hedge Chatter, says his company is currently working with several “globally known” banking firms and they’ve all requested silence.

“There are many investment banking firms trying to explore how to incorporate social media sentiment analysis into their trading strategies,” says Ross. “However, the hard numbers on this are difficult to acquire because correlating social media with investments is not widely accepted yet, so it’s pretty much ‘hush, hush’ on the inner circles. Another reason is if these investment banks can prove some type of modelling it would be their secret sauce and therefore they would not want to alert their competitors.”

In the current economic climate any business advantage over a rival – no matter how small – can make all of the difference between success and failure. As a result, the dangers of failing to keep tabs on what’s being said about your products or services on social media should not be underestimated, cautions’s Sinclair.

“Monitoring what people are saying about your products and industry can help you design your products and propositions for the future and in that sense Twitter acts as a great market research tool as well as a lead-generation tool,” says Sinclair. 

“Similarly, if you monitor what people are saying about your brand it can also help you with customer service and PR. There are many examples of companies who have found themselves under social media attack. Failure to invest in these kinds of tools could easily result in significant damage to a company’s reputation and brand.”

Read more on Business intelligence and analytics