(101 Introduction To Social Analytics+ 201 Boolean Mastery+301 Data Analysis And Insights Generation)

871% Completed
 

301.2 Quantitative Analysis

Please watch the video, read the lesson below and take a quiz at the end of this lesson (bottom of this page).

Quantitative analysis focuses on the numerical data and metrics available for social data. This allows the analyst to make comparison and benchmark performance objectively based on the initial research design. Most quantitative charts are automatically populated in the Insights Section in Radarr – which makes it easier for analysts to identify trends and relationships at a glance.

 

Common quantitative metrics from social listening include:

  • Impact Score™
  • Conversation over Time
  • User Analysis
  • Share of Voice
  • Channel Analysis
  • Sentiment Analysis
  • Competitive Analysis
  • Influencer Analysis

301.2.1 Impact Score™ 

 

Likes, shares, comments – these are the social media metric that everyone is familiar with. While it may be straightforward to compare likes from one post with another, what happens when you want to take into account other metrics, such as shares, comments, or even retweets? Does all these metrics mean the same, or will one outweigh another?

This is where Impact Score comes in handy. Impact Score is based on a proprietary algorithm that helps users sort and identify posts and content that is gaining high volumes of engagement (likes,comments, shares and views) online. Simply put, it is the one score that you need to take into account when considering engagement of certain posts, or the cumulative performance of your campaign.

Example of Use Cases

  • Comparing content performance on social media
  • Discovering posts that are trending to be assimilated into content strategy
  • Determining virality of both positive and negative organic content

 

Conversation Over Time is a very useful analysis to consider the volume breakdown over specific periods – which indicates the level of activity around certain events, topics, or trends of interest online. Following the chart will allow users to benchmark volume prior to any trigger or incident, which gives an objective view on the movement of conversation across time.

In addition to overall conversation, Radarr is also able to display the sentiment breakdown for each period – which means you will automatically be able to gauge the sentiment around your topic of interest and how it has shifted over time.

Tip: Click on the legends below the chart to customize the sentiment you wish to display on the chart!

Example of Use Cases

  • Awareness and buzz around campaign launch
  • Prominence of customer complaint and crisis management
  • Traction around emerging topics of interest

301.2.3 User Analysis

 

Conversation Over Time is a great way to show how interest around certain topics have shifted across time, but it does not take into account the repeated posts from a single source – which means that one person may be artificially inflating the sentiment around an issue. This is a fallacy that social analysts should be mindful of, especially when brands are looking to direct their business strategies according to social insights.

Unique User, when charted against time, shows the number of different accounts that are participating in the conversation online. This will allow you to discover the conversations that may be dominated by a small group, i.e. if there are any keyboard warriors who are actively antagonizing certain topics of discussion, or even when a single advocate is overly enthusiastic about your new product.

Example of Use Cases:

  • Check the authenticity and extensiveness of new trends
  • Audit backlash towards certain brand, products, services, or campaigns

Identify group of advocate or critics that may eventually shape online brand perception

301.2.4 Share of Voice

 

Apart from analysing data across time, another way to dissect the conversation of interest is to compare the volume against similar topics of interest. This will help contextualize the buzz received from certain topics, and can even help create perspective and benchmark against competitors.

The breakdown of Share of Voice ultimately lies in the topic set up on the social listening tool – which will require pre-planning in the topic configuration stage if it is important to make comparison between specific topics or subtopics. For instance, if you wish to determine which snack is most popular on social media, it is advisable for the topics to be configured according to each snack, rather than grouping all the snacks into a general topic.

 

Examples of Use Cases

  • Benchmarking organic brand mentions against competitors
  • Qualifying top themes for conversations around campaign launch
  • Identify which step in the consumer journey that requires attention from the brand

301.2.5 Channel Analysis

 

While it is true that certain social media platforms are more popular in some countries, it is important to be mindful that there tend to be differences in the demographic and setting where discussion happens on each platform. A topic that is trending on Twitter may not see the same on Facebook – which is why it is important to assess the source of where the conversations are coming in from before expanding on the different business strategies.

 

In addition to checking where the conversations are coming from, it may also be useful to deep-dive into the different components to get a more granular view on the happenings within each social media platform (e.g. platform-specific sentiment, platform-specific media type).

301.2.6 Sentiment Analysis

 

As mentioned in the earlier section, there are multiple layers that data from social media can be analysed from – and one of the most prominent is by looking at the sentiment of the post. What this does is to tag a sentiment to each post based on the text content – implying the attitude of the poster and the group of netizens discussing the topics in general.

In addition to a simple sentiment breakdown, it is also useful to look at the top terms across the different sentiment – which will give an idea of whether the sentiment was due to a specific unifying theme. It will also allow you to identify posts or accounts that are dominating the sentiment of discussion – would the key terms turn out to be certain handle, name, or hashtag

 

 

No.

Positive Sentiment Negative Sentiment
Keyword Mentions Keyword Mentions
1 cookies 1536 cookie 233
2 dessert 951 cookies 210
3 enak 351 dessert 126
4 yuk 348 @mecookiemonster 95
5 bikin 289 cookie happened 94

Table 301.2a. Top keywords for positive and negative sentiment around snacking in Indonesia. Highlighted in blue are conversations driven by the same accounts/posts – which can be inferred through the similar number of mentions.

301.2.7 Competitive Analysis

 

While Share of Voice considers the volume of conversations across different comparable topics, there are many other metrics that can be used to benchmark the performance of competing topics for a holistic view. Below are some of the typical metrics considered for measurement:

  • Estimated Reach: Estimated number of people who see your content
  • Mentions: Volume of Conversations within the topic
  • Engagement: Summation of the different engagement metrics available
  • Likes
  • Shares
  • Comments
  • Views
  • Impact Score™

Under the Insights section in Radarr Leaderboard and Heath is a great way to have all the competitive metrics at a glance. This will help you identify the top performing topics (which can be brands, products, or campaigns) and see how the rest of the tops measure up.

301.2.8 Influencer Analysis

There are mainly 2 types of influencer analysis, namely:

  • Influencer Identification
  • Influencer Performance Measure

As the name suggests, Influencer Identification is a prospective method used by brands to scope out potential influencers to engage for any of their products, services, or campaigns. While influencers may be engaged simply based on their popularity in the past, it is no longer the case. Consumers now have different preferences and relationships with influencers, and that changes across different industries.

This is why social analytics works great for influencer identification, where we can look at organic conversations around the industry of interest and determine who are the key opinion leaders within the field. Some metrics to consider include Impact Score and general sentiment of posts.

On the other hand, Influencer Performance Measure is retrospective – where existing influencers are benchmarked against the brand’s pool of competitors, or even industry social performance. This means that the individual account performance can be pitted against one another, similar to what we saw in Competitive Analysis.