Crimson Hexagon & Equity Research

This guide introduces the concept of “brand fitness” and lays out how equity researchers may apply it to derive value from social data by:

  • Quantifying conversations about brands & their direct competitors
  • Understanding fluctuations in sentiment over time 
  • Identifying potential brand health issues

Financial Services

+

Equity Research Case Study

Problem: Online conversation about Uber had skewed overwhelmingly negative for years. At the beginning of 2015, 44% of all tweets about Uber were negative. Researchers needed to quantify topics of conversation about the brand among consumers to identify any fundamental issues with the product. Segmenting the conversation with Crimson Hexagon provided a solution that scaled (45MM tweets).

 

Insight: The research found only a small percentage of complaints about bad drivers, ride costs, or poor experiences. Specifically, the analysis of social conversation revealed a healthy, working product that people liked using. Instead, overwhelmingly, Uber’s negative sentiment was driven by press surrounding then CEO Travis Kalanick and several high-profile legal issues (1).

Key Takeaway: Social listening data should be included as part of a comprehensive research program.

How do I get started?

Start with brand fitness. Focus your research on the “owned conversation” for a single brand and its direct competitors. Measure the volume, tone, and content of the conversation that consumers are having directly brands. These metrics provide a good snapshot of what people think and feel, and require minimal effort to analyze using a repeatable framework.

 

Conceptualizing & Measuring Brand Fitness

Deriving key performance indicators

Offering a lifetime warranty or “satisfaction guaranteed” indicates that a brand has confidence in the reliability of its products or services. Similarly, marketing activities send signals about a brand’s fitness by way of perceived expense and effort. Research demonstrates how signals of brand fitness have a powerful impact on consumers’ attitudes and purchase intentions (1)(2)(3).

Social listening tools provide a clear view of a brand’s perceived expense and perceived effort (the sum of these signals could be said to add up to perceived caring) as signaled by the quality of their interactions with people online. We may, therefore, measure customer-based brand fitness by quantifying the volume, tone, and content of conversations between consumers and brands. Further, utilizing engagement conversation as an indicator of brand fitness allow for a repeatable research methodology with comparable like-for-like metrics and benchmarks.

The Data

Defining the sample frame

The biggest challenge for researchers using social data is separating the signal from the noise. Brand fitness metrics are derived primarily from enagement conversation, a subset of data that is simple to isolate and rich in meaningfulness.

Online conversation about a brand or organization is composed of two streams. First, “owned conversation” centers on social media posts directed at a brand–or more specifically, a social account or property owned by the brand. This compares loosely with traditional media analytics in terms of paid advertising. This can provide insight into a brand’s “fitness” by measuring its efforts to gain traction through the lens of consumer engagement.

Second, organic or “earned conversation” aligns with the concept of earned media, traditionally thought of as brand mentions in newspapers and magazine articles. This measures what people are saying among themselves. This data has much to offer, however, it can be time-consuming to execute. An “earned conversation” dataset is best employed for industry-level research or trend spotting.

The Analysis

Surfacing actionable insights

An essential first step in the analysis is to ensure there is a clear definition of the question that is being asked. This methodology allows you to identify how a brand performs on key social metrics and how these compare with its industry. For example, we might ask: “How does LensCrafters brand fitness compare with other eyewear brands in terms of volume, tone, and topics of social engagement?”

The brand fitness framework uses this basic structure:

1. Volume

2. Sentiment 

3. Topics

Example Analysis: Eyewear

A share of voice (SOV) analysis compares the total volume of brand engagement conversation vs. top competitors (see below).

Basic Sentiment analysis of all engagement conversation by brand provides a baseline for comparison (see below).

Net Sentiment scores by month will provide a view into overall stability and can identify potential red flag events (see below). 

 

Sample Frame Design

Data Sources

What content sources are available that provide data on consumer engagement with brands?

Twitter Only Analysis

Some researchers rely primarily on Twitter data only for brand fitness analysis. Twitter has been validated as a broadly representative source of online conversation data that can serve as a surrogate for the larger social sphere when necessary.

 

“Because of the way it is used and perceived by users, Twitter seems to us to be most representative of the broader social sphere… When the wealth of data from [Facebook and interest-based social networks] is shared into the broader social stream, it usually comes through Twitter. Thus Twitter seems to us the best aggregation of discovery and sharing,” (Millward Brown Digital).*

 

Additionally, Crimson Hexagon provides access to the full firehose of Twitter data since 2014 (and partial data as far back as 2008), allowing researchers access to the entire universe of data on the platform and the ability to make rich year-over-year comparisons.

Incorporating Facebook

Facebook Page data sometimes provides an additional data point. However, Crimson Hexagon cannot provide Page data prior to 2017. As a result, like-for-like annual comparisons are more limited when incorporated into the analysis.

Instagram Unavailable

Instagram place strict limits on what data they provide, specifically in terms of historical data. Please refer to our summary Instagram policies for more information.

Search String

Search for @brand

This search includes content that mentions the brand’s account (does include @Chevrolet, does not include “Chevy”), direct replies to content broadcast by the brand, and re-tweets of content broadcast by the account.

Remove author:@brand

Users may isolate engagement conversation for a brand using either a Buzz Monitor or Social Account Monitor (SAM). When using a Buzz monitor be sure to remove any posts made by the brand itself. Replies to tweets made by the brand will still be included in the results even when content broadcast by the original author is removed.

Keep Re-tweets 

A re-tweet may at first appear to be irrelevant data as it seems to be either duplicate content, content from the brand rather than consumers, or both. However, the reality is that each re-tweet is a piece of content that a user has chosen to share directly with their followers.

Each re-tweet should be understood as an endorsement of that particular thought or idea or an expression of its significance (i.e., newsworthy). As a result, each time a post is re-tweeted it is measured as a new post in the conversation and positively impacts the volume metric.

Setup

SAM: @chevrolet

Buzz: @chevrolet AND -author@chevrolet

 

Executing The Analysis

Volume – Are the brands being talked about?

  • The share of voice (proportion of total conversation)
    • When does proportion change?
      • One brand increases – Possible increase in market share/ marketing spend
      • One brand decreases – Possible drop in brand salience/ product or services becoming stale
  • Volume comparison over time (volume for each brand)
    • When does volume spike?
      • Do the brands move together? – May indicate seasonality
      • Does it increase momentarily? – May indicate an event or viral content
      • Does it increase steadily? – May indicate brand growth or investment in marketing

 

Sentiment – What is the tone of the conversation about each brand?

  • Positive/Negative by brand (all time)
    • How does sentiment compare?
      • High positive/negative sentiment for all brands? – Possible industry trend (e.g., high negative sentiment for all cable companies)
      • Low positive/negative sentiment for all brands? – Possible low-interest industry (e.g., mostly neutral sentiment for life insurance)
  • Positive/Negative change by brand (year over year)
    • How does sentiment change? – Possible consumer attitude shift (e.g., the brand has improved customer service)
      • Sentiment change by brand (volume each brand or Net Sentiment)
  • When does positive/negative sentiment spike or skew? (all time or past year)
    • Does either pop momentarily? – Possible product issue or bad PR
    • Does either slowly change? – Possible change in product quality or aging facilities

 

Content – What are the topics of conversation about each brand?

Please Note: This piece of the analysis requires the most work and is the most subjective. The primary tools to assist you are found under the “explore” tab, however, our machine learning solution is ideal when a more quantitative analysis of content is required.

  • What are the overall topics and sub-topics of conversation? (overall or by brand)
  • Are any topics defined by an emotion or sentiment?
    • What are the topics among positive and/or negative conversation?
      • Are people expressing concerns about product/service fundamentals?
      • Are people excited about a particular feature or experience?
  • What are the top sites and URLs about the brand being shared online?
    • Which articles/content are being amplified by social media?
  • Can machine learning provide additional support?
    • Do you want to quantify the components of positive and negative sentiment?
    • Do you want to easily compare conversation topics across brands?

 

 

Authored by Jason Potteiger, Sr. Manager, Customer Success

 

* Source: Czernek, Anne (Senior Research Analyst): “POV: Social Measurement Depends on Data Quantity and Quality”, Millward Brown Digital

Request a Demo

Ready to transform your business?

Get a walkthrough of Crimson Hexagon and learn how consumer insights can help you make better business decisions.