Analytics Systems

The New Voice of Customers Online

In the midst of the digital revolution, the telecom industry is undergoing major changes. One of the changes affecting telcos is the increase in data sources from which to get customer feedback. Where this used to be fully controlled by companies through their call centers, websites and shops, today much feedback is expressed in social media, blogs, news sites, app stores and forums.

In the digital world, telcos need to systematically listen to the voice of their customers online.

In the digital world, agility in adapting to customer feedback is very important. At Telefónica SA (NYSE: TEF), when we launch a new digital product, we actively seek quick feedback using the Internet, and in particular social media such as Twitter. The Internet quickly tells us how the product announcement is featured in press and blogs, and also shows what announcements resonate more in the digital world by tracking re-tweets and other sharing mechanisms.

Businesses can consume a graphical and interactive dashboard to get the most out of this data and turn it into relevant insights. Based on the need of our businesses, we update dashboards on a daily, weekly or monthly basis. At Telefónica, we deliver value from this "freely" available data by listening. This "listening" is built around three main concepts: crawling the Internet, concept identification and sentiment analysis/visualization.

Commercial crawlers are available (e.g. www.sysomos.com) to crawl public data sources including Twitter, blogs, news, media and forums. A crawler typically takes as input a combination of keywords (and, or, not). Selecting the right search terms is important to avoid inclusion of noise. The output is a set of posts, Tweets and articles containing the specific search terms. Where available, geolocation is provided. In case of apps, reviews from app stores (e.g. Google Play and iTunes) are also considered.

Once the data is collected, it needs to be analyzed for relevant concepts and sentiment. Several solutions are available on the market (e.g. www.bitext.com). Leading market products come up with the following results:

  • The concepts mentioned in the items
  • A set of possibly multiple opinions that constitute the items -- a Tweet or news item may contain several opinions about different concepts
  • The neutrality or degree of tonality of the opinions (how positive or negative)
  • The concepts the opinions are about
  • The phrases used to express the tonality of opinions (sentiment)

Concept cloud representing how sentiment about objects is expressed. Size represents number of opinions; color represents tonality as in next figure.
Concept cloud representing how sentiment about objects is expressed. Size represents number of opinions; color represents tonality as in next figure.

Once the data is analyzed, it needs to be visualized for consumption by business users. There are many visualization tools (e.g. www.tableau.com) for building interactive dashboards. Interaction allows users to filter for viewing only negative or positive comments, or for different languages, to drill down into more detail, and to always review the original content.

Distribution of tonality of opinions detected in the mentions.
Distribution of tonality of opinions detected in the mentions.

We have learned that -- apart from the typical reputation tracking that social media analytics is used for -- it is a valuable tool for getting quick and economic customer feedback and insights for products. The tool is able to detect specific issues people complain about, such as for example customer care quality, price and pricing issues, registration process, and specific product features like crashes, unclear interfaces, battery drain, etc.

One thing business users see as very positive is the fact that the tool is available from day one of launch, which enables quick responses to typical overlooked product issues, and complements the internally available product KPIs such as downloads, registrations, active users, etc. In general, we see that in the early days after commercial launch, comments are mostly positive, reflecting the fact that most are announcements and promises of the great features of the product. Over time, more and more feedback comes in based on actual usage of the product.

An interesting lesson has been that for some business owners, it is not easy to deal with a lot of negative feedback. And the fact that it is so easy to get feedback and that it is based on publicly available knowledge makes it harder to "hide" the insights. This is, however, above all a cultural issue. In the lean, digital world, negative feedback should be embraced and taken as an opportunity to quickly improve products based on real customer insights.

— Richard Benjamins, Group Director BI & Big Data – Internal Exploitation, Telefónica SA (NYSE: TEF)

Joe Stanganelli 11/13/2014 | 7:48:59 PM
Re: Linguistic sentiment analysis @jabailo:  Interesting.  Of course, I'm sure this has all been taken into account.

For instance, there was one modestly successful hedge fund (Derwent) that was based on a an algorithm that monitored the sentiment of Tweets (sentiment on ALL sorts of topics...not just about business/stocks) and linked the performance of the stock market to the positivity or negativity of Tweets, finding that its performance could be predicted up to 6 days in advance.

Presumably, snowball effects of positivity or negativity do take place -- and affect markets accordingly.
jabailo 11/11/2014 | 11:29:12 AM
Re: Linguistic sentiment analysis It would also be interesting to see if there are memetic chain reactions.

That is, does one person complaining about something, influence others do mock and do the same, and pass it along.    Same with praise, though something about human nature tells me complaining has a stronger draw.

Then, of course, if you want to start influencing, you could see the effect of injecting your own statements into the social noosphere.

Joe Stanganelli 11/11/2014 | 6:40:23 AM
Linguistic sentiment analysis Linguistic sentiment analysis is very powerful indeed.  It has been used to successfully predict the stock market, elections, and sales.  Seems like telecoms looking to it is an "about time" decision.
Sign In