Reading Social Sentiment Tea Leaves, Proceed with Caution

Niraj Sharma |
 08/24/18 |
4 min read

Anytime communication happens via the written word, images, or anything beyond in-person conversation, there’s the possibility of misinterpretation. This is especially true when language nuances like sarcasm are involved. Thank goodness there’s social sentiment to clarify things!

What Is Social Sentiment, Exactly?

Social sentiment is the measurement of emotions shared by social users online. It’s the barometer that tells you where to focus based on what has your social audience in a tizzy at any given moment.

We measure social sentiment by combining two values:

  • Net Sentiment: whether emotions are positive or negative, on a scale from -100 to +100
  • Passion Intensity: the strength of those emotions, on a scale from -100 to +100

The combined, weighted value – called Brand Passion – tells you a lot about your audience, and how they feel about you, your competitors, and everything else.

Sounds pretty simple, but it’s actually fairly complex – meaning, you need tools able to decipher the nuances of human language to get accurate readings. Not all can.

Consider this post:

GIF of BTS singer

If you speak fluent “Dramatic Teen,” it’s clear this post demonstrates a lot of love for Jungkook, a member of the mega-popular Korean boy band BTS (Bangtan Boys).

But where in your analytics would this post be counted? The band isn’t named or tagged, Jungkook isn’t named or tagged, and unless you understand the phrase “trying to kill me” is a common expression of major love, it’s unlikely you’ll find it in a keyword search.

How would you ever know this GIF is getting hearts and retweets if you were running social analytics for BTS?

You wouldn’t. Now think about how many similar posts you may be missing. How reliable are your insights?

How about this one:

Emoji-laden tweet about Slack app

In this case, workplace app Slack is mentioned by name and tagged – but what’s being communicated exactly?

You’d better speak Emoji if you want to know!

What about Facebook Reactions? Certainly, those are indisputable… aren’t they?

Screenshot of Facebook Reactions

Not really. A thumbs-up can mean you agree with the poster – or simply that you’ve read the post, which is far less intense an emotion. But that’s what the heart is for, right? Well… the heart can signify love for the post, or for the person posting. When someone shares an obituary, a heart doesn’t mean you love that the person died.

Reactions are just as confusing as other social languages.

What should be clear from the examples above is it’s not enough to look at obvious references to love or hate and assume you’re getting the full picture. So, what do you do?

Tools of the Sentiment Trade

Just as it’s not enough to count mentions and retweets in your analytics, it’s not enough to look at sentiment as merely positive or negative. You need a lot more information if you’re to connect with your audience at the deepest level – and keep your brand out of hot water at the same time.

Here are the features that matter for a reliable assessment of social sentiment:

Natural Language Processing (NLP) – You can’t measure sentiment of language if you can’t understand it. Your tools must be able to recognize a sarcastic tone within the context of a social post – to know when words like “love” or “hate” and their synonyms are meant to convey the opposite of their meanings.

Analyzing Emojis – When characters are limited, or social users are short on time, emojis often stand in for text. If they don’t stand in for text completely, they may modify the meaning of text to convey greater Passion Intensity. If they aren’t part of your analytics, your insights will be lacking.

Image Analytics – Sometimes a picture is worth a thousand words – and sometimes it’s worth 280 characters. A picture may say it all to the person viewing it, but it says absolutely nothing to your analytics if your tools can’t recognize people, places, facial expressions, brand logos and more. This is a fairly new part of sentiment analysis, but its importance will only grow.

Social Monitoring – Used to alert you to both trends and trouble, social monitoring uses your keywords and settings to let you know when there’s an uptick in sentiment in either direction. This way you can jump on emerging trends as early as possible – or just as quickly nip any damaging posts in the bud.

Understanding sentiment – and things like sarcasm – all help you know what your audience wants. Maybe a little sass is exactly what should be on the menu. It clearly should be according to Hamburger Helper:

Snarky tweet from @Helper

And that engagement proves the case.

But sarcasm is extra tricky – especially in print. This is why you need to know your audience well enough to know whether they are in on and appreciate the joke – or whether more traditional styles of conversation are warranted. Failure to recognize sarcasm can even land you in court.

Luckily, you don’t need to rely on tea leaves to keep your brand on track. Social sentiment tools do this work for you. As long as you invest in those with the key features noted above, you’ll never have to wonder what your audience “really” thinks again. You’ll always know.

Want a closer look at our social sentiment tools? Contact us for a personalized demo!



Premier social media analytics platform

Expand your social platform with LexisNexis news media

Power of social analytics for your entire team

Media analytics and market intelligence platform

Enrich your media analytics with social data

Social media benchmarking
and competitive intelligence

Data streams & custom KPIs for advanced data science

AI, Image Analytics, Reporting Tools & more

Out-of-the-box integration with other data sources