As a growing insight field, social data and by extension social intelligence have only recently adopted more rigor and discipline. Consider and compare its introduction to our lexicon with the below graphic, which compares the prevalence of ‘market research’ and ‘social media listening’ across print sources.

Evolution of Key Phrases Found in Print Sources

Why does this matter? Well, standards and best practices support consistency and improvement. Without them, finding social insights feels lucky, as if you stumble upon them.

While we can’t unpack all facets of social intelligence, let’s focus on one big, critical element – its underlying data sources, i.e., the social platforms themselves. Specifically, we’ll look at the types of human behavior and conversations you can expect to find across various social media platforms.

Quality or Quantity?

While we know that meaningful insights require thoughtful questions and hypotheses, we still may be in the dark in terms of where to look and what to expect.

Our intuition suggests that more is better. The more social data sources, the more mentions, the more likely we’ll find representative insights.

This is true to an extent, but with more data comes more potential noise and effort.

An alternative view suggests combining quantity with quality. Yes, volume always matters. But the more relevant consideration is volume across relevant sources.

To identify relevant sources, we must first define our question or hypothesis, and then possess an understanding for the types of conversations and behavior we can expect from the available sources.

The Real Self Versus the Projected Self

When we think about our own social media usage, the truth is that each social platform reflects some degree of either the real-self or the projected self.

The projected-self encompasses who we aspire or pretend to be. The real-self more accurately portrays who we are offline, in the real-world. Critically, each self impacts our decisions and motivations.

Sources like Reddit and forums attract more of the true-self, with honest questions and discussion. Platforms like Instagram project highly curated and visual aspirations of how we want others to perceive us. Each source has its value, with forums surfacing unmet needs and product feedback, and Instagram surfacing moments of consumption and creative insights.

Knowledge is power here. Knowing that your questions relate to product feedback or the purchasing experience places you closer to real-self platforms. Creative or marketing-based insights place you closer to the projected-self platforms.



In addition to the real and projected selves, social sources run a functional-to-emotional and interactive-to-declarative matrix.

Functional sources serve to exchange information and direct people to a decision or action. Emotional sources exchange in-moment experiences or feelings. Interactive sources enable back and forth and engagement. Declarative sources are typically one-way modes of communication.

functional-to-emotional scale

Again, depending on the nature of your question, you’ll want to index on the related source.

Download Our Guide – Reddit Explained

Conversation Style by Platform

But the above frameworks don’t tell the whole story. To truly understand the fit or relevancy of each source, you must dive deeper and understand the nuances of each platform. Next, we’ll look more closely at three types of social data sources.

Consumer Review

Consumer review sites are a goldmine for product and competitive insights. And with 95 percent of shoppers reading reviews before purchase, they are also key to manage and improve over time.

Sources:,,, Sina Weibo

Nature of conversation: Authors describe their real-life experiences pre, post, and during a product purchase. They are evaluations. Readers gain functional insights that inform their next purchase.

Content Structure: Compared to traditional social platforms, reviews are longer formed and follow cleaner case and syntax rules.

Value: Surface high-quality product and competitive intelligence. Structured data, via star ratings, allow for tracking of product performance.


Micro-blogging sites like Twitter surface in-moment, opinion-based discussion.

Sources: Twitter, Sina Weibo, Tencent

Nature of conversation: Twitter and other micro-blogs serve as soapboxes for airing grievances, rapid transfer of trending information, and producers of pop-culture and viral content.

Content structure: Short form due to character limitations and unique use of hashtags. Heavy use of slang and disregard for syntax. Increasing use of videos.

Value: Good to find ‘in the moment’ experiences, new trends, customer experience issues, feedback on events, etc. As it is one of the places global brands have branded channels, it is also relevant for campaign impact analysis.


Forums are typically interest-based discussion groups that promote questions and answers among a community. Reddit is the prime example of a forum. And we all know it’s influence is growing.

Sources: Reddit, Yelp, Quora

Nature of conversation: Forums are interest based communities, that, above all, promote question and answer style discussions. Authors and readers mostly exchange product and experience-based information. While mostly functional in nature, many forums also instill shared feelings, opinions or emotions across its community group.

Content structure: Longer in form than micro-blogs, forums tend to adhere to syntax rules and are predominantly text-based.

Value: Because of their question-based style, forums are great for unearthing challenges or friction points across the customer experience. The long form and honest exchange of information unveils deep insights related to unmet needs, concerns and recommendations.

Of course, the above list is not exhaustive. But as you can see, each platform produces unique conversations. Knowing the nature of these conversations and their corresponding value helps you define the most relevant sources and obtain insights faster.

The opportunity with social research is vast and not fully tapped. By installing best practices and shared knowledge, however, we can introduce rigor and more consistently find the insights we desire.

Request A Demo