Activating social listening to protect online reputation (part 1)

EDITOR’S NOTE: this is a guest post by Leopoldo Martínez D., a researcher, consultant on social media corporate intelligence and lecturer at UCV and IESA (Venezuela), and it was originally published on his blog (in Spanish).

 

1. Introduction

In this previous post, I suggested that conversations taking place in virtual communities fostered by a digital marketing plan generated a feedback that was useful to assess and monitor the performance of a digital marketing strategy.

This feedback could generate a huge amount of valuable data (Big Data) that enable to create a knowledge base on what is being talked about, who is telling it, who is having a greater impact on brand image, products, people or organizations.

This knowledge base can also be fed by discussions arising from unexpected events, which are not part of the communication plan, but deal with the virtual community’s topics of interest.

To specifically assess the conversation’s impact, it is necessary to pay attention (more than just listening) to what is being said through metrics (qualitative and quantitative) that reflect the perception that the members of the virtual community have on brands, products, people or organizations, being this perception an online reputation measure.

In this sense, the purpose of this post is to show in a simple way how to carry out an active listening of conversations in social networks to evaluate the online reputation.

2. Assessing online reputation in an ocean of information

In a digital marketing strategy, online reputation management is carried out through active listening. This process consists of four phases:

  • Monitoring: listening to what is said through the search and extraction of data from conversations using key words or phrases. Conversations can be past or real-time.
  • Filtering: conversations are filtered using criteria that extract what’s meaningful (comments, praises, complaints, etc.).
  • Analysis: consists of the observation and intelligent analysis of conversation data through the study of the sentiment of the interactions (positive, neutral, or negative mentions).
  • Action: once the conversation’s topic and sentiment are understood, it might be necessary to carry out some tasks for cleaning up the online reputation or improving it.

 3. Sentiment as an indicator of online reputation

Understanding the sentiment of a message or conversation is an ability of human intelligence. But when we are talking about hundreds of thousands of messages, we need to employ computer tools that automate this task and, at the same time, emulate human intelligence to understand the underlying sentiment.

There are several techniques to carry to out this type of emulation, among which the best known are sentiment analysis and opinion mining. This analysis allows inferring the sentiment of a message through its polarity (positive, neutral, or negative). This polarity expresses a person’s perception of a subject.

Another feature that the sentiment of a message often presents is that it usually reflects possible causes and effects of a situation or topic discussed in a conversation. This aspect is vital to assess online reputation, as it may reveal key details about potential problems (causes) and consequences (effects) around the image of a brand, product, person or organization.

For this reason, another technique required to evaluate online reputation is text classification, which, through a specific computer application, assigns a topic or category (economy, finance, society, police, security, entertainment, food, etc.) to a conversation. With the category identified and depending on the situation considered, the human professional will determine if the category collects messages of cause or effect.

The combination of polarities and categories (causes and effects) can be represented graphically to show whether the perception on a subject is favorable or not for the online reputation. Figure 1 shows a hypothetical example of 3000 messages analyzed on a day like any other.

Figure 1. Polarity of messages (causes and effects)

In Figure 1, you can see that there are 2000 messages with negative polarity (N): 1000 of them belong to cause categories and the other 1000 to effect categories. 600 messages have neutral polarity (NEU), consisting of 400 cause categories and 200 effect ones, whereas 200 of the 400 messages with positive polarity (P) belong to the cause categories and the other 200 to the effect ones.

The calculations 2000/3000 = 66.67%, 600/3000 = 20%, and 400/3000 = 13.33% can be taken as measures of the perception of the hypothetical situation taken into consideration, and hence as measures of online reputation. Also, the axes Cause categories and Effect categories provide intensities of the possible causes and effects.

In the following post, I will present this online reputation evaluation approach in a real situation related to tourism.

 

Leopoldo Martínez D.


Leave a Reply

Your email address will not be published. Required fields are marked *

*
*