1-855-SOCMROI (1-855-762-6764) info@netbase.com

How to Evaluate Social Analytics

Updated July 2019.

Social Media Analytics, powered by Next Generation Artificial Intelligence (AI), comprises several techniques and tools. When combined, they provide a thorough understanding of your place in the social landscape, as well as the audiences you most want to target. And lots more. Let’s explore the possibilities.

Social Media Analytics Strategizing Starts Here

Social insights allow for strategic actions in all areas of brand operations, making your business leaner, smarter, and more profitable – with greater potential for long-term growth

Our How-To Guide will break down several of these techniques in greater detail, but first you need a general awareness of the various moving parts, and how they fit together.

Seeing the forest before zooming in on a specific clump of trees is wise, because you’ll likely want to move around and explore other areas. Having a sense of potential indicators to look for will help guide that exploration in an organized manner.

Without an understanding of what’s on offer with social media analytics, it’s easy to lose your bearings, get lost, then overwhelmed – and the next thing you know, some other business unit has claimed your budget! Or, even worse, some competitor has claimed your customer base.

Let’s get that topography in your hands . . .

NetBase social analytics dashboard

Start with Social Listening

Perhaps the biggest change to marketing in the past decade is the shift to consumer centricity. Promoting the tenets of your brand’s quality used to be enough to sell products, but as technology has leveled the playing field of production, consumers have demanded more.

With the rise of social media, any marketing efforts not based in human connection are simply relegated to background noise, an unwelcome intrusion. Social media is a consumer playground, so brands must abide by consumer rules. All interactions must be human, authentic, personal, invited.

But this is actually good news – because all the consumer information you could ever ask for is freely available online. You simply need the right tools, and a little know-how, to mine for gold.

Social Listening is the first step in understanding who your audience is and what they want. Surrender any notion of demographic breakdowns as the end-all, be-all. Market to a segment of women aged 25-34 assuming they all have the same interests, and you’re not only turning off every woman who doesn’t fit into that mold, you’re likely not even addressing a huge chunk of your true audience.

Instead use demographics as a framework to start your research, and adjust your thinking as you discover outliers within the mainstream majority. You’ll want to explore interests, professions, ethnicity, bio terms – all ways these folks self-identify. But even that is just a starting point:

Evaluata beyond typical demographics using social analytics

These outliers will become visible as you “listen” for psychographic information – and that’s what’s most important now.

Getting Psychographic About Things

Psychographics comprise the opinions, attitudes, and behaviors of consumers. This can be in relation to your brand, of course – but the best insights come from what consumers talk about beyond your brand. This is what really helps you identify and connect to them.

For example, if you owned a pizza restaurant, you might use a tool like NetBase Pro to search on the term “pizza.” Among the various insights you’d learn the most popular post is this one:

Trainer Simeon Panda with a pizza

Just from the engagement on this one post you learn how many people care about being able to eat pizza while maintaining a fit physique. This user is also a potential influencer for your pizza brand – or gym. And this is before you’ve even looked at all the other data from your search.

What insights like this allow you to do is segment your audience so you can reach out with individualized messaging crafted just for people who love pizza and working out.

You might also have a segment of political activists who use pizza as a reward for their volunteers, or people who enjoy pizza during a particular TV show each week.


There’s no limit to the segments you can create – and you’ll notice there’s no talk of age or gender in the example above, just common interests.

That’s intentional. Because the goal is to hit your audience in “the feelz.”


As you create audience segments, and in every other tactic we’re about to discuss, you need to dial in to consumer emotion. The idea is to understand and appeal to their deepest passions – whether positive or negative.

We measure this in two ways:

  1. Net Sentiment, or whether consumers’ emotions are positive or negative
  2. Passion Intensity, or how strong those emotions are

Together, these values – ranked on a scale from -100 to +100 – determine Brand Passion, a critical metric in determining what actions you need to take next.

As we can see in this airline example, sentiment is pretty low right now, at 22%. And it appears a delay and emergency landing figure prominently in that result. But we can also see passengers impressed with the focus on safety – and that can provide an angle for airlines to capture and direct worried consumers toward:

sentiment and sentiment drivers to help understand audience motivations

You’ll also want to know what percentage of a conversation is trending positively and how much of it is negative, as this yogurt example illustrates:

Sentiment summary metrics supporting positive and negative conversations

Also, what is driving that sentiment:

Attriibutes found in online conversations about yogurt

Using this data, brands can drill down (to each item in a word cloud and analyze) and take advantage of unmet needs, or discover new product ideas that they may have missed otherwise.

To ensure Sentiment Analysis is accurate though, beyond offering the abilty to get extremely granular and transparent results, your tools must evaluate language and its varying idiosyncrasies – from global languages like English, French, Mandarin, etc., to slang, sarcasm, netspeak and more. This includes pop culture references, and “silent” languages like emojis and images. It’s called Natural Language Processing (NLP).

Every way sentiment is expressed online must be captured by your Sentiment Analysis Tools or your Social Analytics will be off.


A great example of this in action is 113 Industries’ work for Ocean Spray as they searched for their next $100 million beverage line. With the morning juice market in decline, they wanted to come from another angle. Using Social Analytics with Sentiment Analysis, they identified 3 consumer behaviors tied to cranberry juice:

  • Love for cranberry juice, but embarrassment of the stigma that drinking it is tied to having a urinary tract infection
  • Use beyond breakfast hours, like women drinking cranberry juice to rehydrate after Zumba class
  • Women enjoying cranberry juice in a wine glass in the evening, finding the rich color and texture appealing as a non-alcoholic cocktail

These insights allowed Ocean Spray to create two new product lines to grab additional shelf real estate in stores, as well as break past the traditional breakfast juice market. The results of their campaign exceeded all previous new product expectations.

This wouldn’t have been possible without exploring consumer sentiment around cranberry juice. Simple metrics like number of mentions would only have told them how much cranberry juice and/or Ocean Spray as a brand were being talked about. You need social sentiment for context – at every level of your analytics.

tweet of Ocean Spray mocktails

And what other levels are we talking about? Here are additional areas you must analyze for complete Social Media Analytics.

Channel Surfing 101

Just as you can’t assume who your audience is, you can’t assume where they love to interact. You can’t reach them if you don’t know where they are.

For example, a search on the terms “yogurt” and “#yogurt” in the US reveals the following domains:

If you’re a yogurt brand, it makes sense to focus on high-volume channels like Twitter and Instagram – which you might have come to anyway. But what about Reddit? Would you have thought to look there? Many brands don’t – and should:

Reddit sound bites about yogurt

There are also several smaller online community and blog sites, as well as travel sites. Brands shouldn’t ignore an opportunity to connect with a unique consumer segment on unexpected channels. These could turn out to be your most passionate and loyal customers in the end.

It’s never about jumping onto the newest or seemingly most popular channel – note the absence of Facebook from this list, for example. Would you expect that?

And just because it isn’t showing as a top location, doesn’t mean it’s not popular with a smaller, very specific niche of yogurt lovers that a brand may want to target.


To continue our yogurt example – what other niche audiences are out there anyway –  and where?

You can’t go wrong looking to topic influencers to see what they’re saying as a solid basis to start your search:

Analyzing authors found in a given topic=

The conversations they’re having reveal what is trending where – and with who. Are they capturing your target segments’ attention? You can segment influencers lots of ways to be sure:

Segmenting influencers by a variety of metrics demographics and sources

It’s about being where your target audience is and where the influencers are. And when that matches up and you create a successful collaboration for your brand. – it’s about knowing what to have these influencers (popular authors) post where.

Content Is Not Universal

The channel you choose must guide the kind of content you offer. This is of major importance in both catching consumers’ attention, and engaging with them.

When on Snapchat, use of fun filters like animal ears and such add the whimsy of the young-at-heart to your posts. But that wouldn’t make sense on LinkedIn. Not even close!

On Facebook, extended live videos offer a chance to truly show your audience what you’re all about, while bringing them information or entertainment they crave. It also alerts followers to your “live” status and encourages interaction.  Having lots of time set aside, understanding the right time of day to host these events – and being prepared for any questions that. participants may ask are all very important items to plan ahead for here.

And it’s all intel brands can sort out with the help of social analytics:

What are people saying:

Word cloud shedding light on what an audience is saying.

What times are they most active online:

Day and hour social analytics insight

Power of the Preview

Having this capability (to preview potential audience questions/concerns) is also key when hosting Reddit Ask Me Anything (AMAs) events! Particularly as AMAs are crucial to understanding and (inevitably) hosting for brand success on that platform.

On Instagram, user generated content (UCG) is amazing. Encouraging followers to create short bursts of video and images in support of an event, promotion or product can become a story for the brand.  And combining UCG later to create brand highlights offers brands exceptionally viral opportunities.

Enticing followers to view while they can.

But even on a single social channel there may be several ways to interact with your audience. Social Analytics helps you discover which are their favorites and least favorites – so you’re always speaking their language.

We can look to the UCL Final, a featured brand in the NetBase Media & Entertainment Industry Best Practices Guide, to illustrate.

During the game, most of the discussion is centered around text, and engagement is a bit lagging. However, after the game, image and video sharing spikes, as does engagement.

graph of UCL Final conversation trends by media type, and versus engagement

This is important information for understanding what your audience wants and when. There’s no one perfect type of content to drive engagement – you need to analyze social data to learn all the nuances.

Social Analytics for Every Business Goal

Once you understand the basics of Social Analytics, and the importance of using Sentiment Analysis, you can use specific tools to find further insights in focused areas of concern, including:

Additional “how to” topics include:

We’ll explore each of these areas in upcoming posts. In the meantime, start where you are, and reach out to us if you need help!

Want a customized demo of how to use our industry-recognized Social Media Analytics Tools? Reach out!

Experience the NetBase difference!

Understand your customers like never before.

Take a tour to see how NetBase can fuel top performing campaigns and marketing strategies.