Sentiment of the Union: Was it All About Obama?

Kimberly Surico |
 01/16/16 |
3 min read


Haters gonna hate, amirite? Consumers are quick to express their disdain towards a concept, product, or person on social, and so it’s important that brands be efficient in understanding what exactly drives and motivates negative sentiment online. Let’s use social media monitoring to pinpoint how sentiment-based insights can make or break the way you interpret consumers in real-time with the State of the Union address.

As President Obama discussed the last seven years of executive accomplishments and forecasted the next year to come, we at NetBase were on the edge of our seats to see what users were discussing on social media. A quick analysis yielded the following as an initial result:

sotu sentiment graph 1

It’s evident that prime Twitter engagement occurs in the moment; mentions take off at 5 pm PT, peak at 6 pm PT, and precipitously decline until 8 pm PT.

But what do these social media users say? How do they really feel about the debate, word for word? NetBase’s Sentiment Drivers allow us to deep dive and quickly slice into the content to better understand what ignites fans to a subject.

Social media sentiment analysis can be examined in three ways:

1) Attributes

sotu attributes 1

2) Emotions

sotu emotions 1

3) Behaviors

sotu behaviors 1

But there’s even more to ensuring what you see is accurate.

The “one bad apple” effect

A brief glance at the attribute word cloud clearly announces that the President didn’t have so hot a night: the social media audience categorized his speech as lethargic, boring, and non-substantive. But this doesn’t quite match the emotions and behaviors above, does it?

So where is this sentiment coming from? Who is driving it? Luckily, NetBase makes it simple to dig into the overall conversation and uncover the context around sentiment.

sotu trump tweet

And voilà! It turns out that our word cloud sentiment was heavily influenced by The Donald. So what, right? Trump antipathy towards the President is nothing new. But as the below tweets demonstrate, we can begin to understand that the target of user invective wasn’t necessarily the president, but instead Trump himself.  It seems that Twitter users are just sharing his post and tethering on their own commentary to temper, if not redirect, the negative sentiment.

sotu Trump RT 1

sotu Trump RT 2

So, what else is going on then? We want to make sure we’re getting a complete picture of the data.

Removing the Trump-colored glasses

Analytics is an iterative process; we learn by performing and re-performing analyses, but that doesn’t mean the process need be unnecessarily tedious. NetBase makes it easy to jump into the conversation to uncover what’s of utmost importance: your audience’s opinion, passion, and interests on the subject at hand.

So what would the conversation look like without Trump’s naysaying? One strategy is to apply filters – one to exclude retweets (but include replies and original posts) and another to exclude @realdonaldtrump mentions. And a pretty dramatic shift occurs.

sotu sentiment graph 2

A striking contrast to our original analysis: overall sentiment jumped from 24% to 45%. And look at the word clouds:

1) Attributes

sotu attributes 2

2) Emotions

sotu emotions 2

3) Behaviors

sotu behaviors 2

These behaviors, attributes, and emotions paint a completely different picture. Twitter users are impressed by a “strong” and “great” speech, and encouraging others to watch; in fact, 48% of tweets that shared behaviors used the word “watch” positively. Their exuberance is palpable: they’re “proud” of their POTUS and nation; they “love” the president and feel optimistic about the future. This is a dramatically different perspective than what we started with!

The best part? I was able to uncover these insights buried in the content within minutes. The visuals provided quick guidance on how to refine research and understand the SOTU’s digital reception.

And brands can just as easily harness the power of NetBase’s robust natural language processor (NLP) to better understand their customers’ likes and dislikes –and align to what consumers really care about.

Imagine being able to spot negative sentiment – like we saw above – and solve a customer issue in real-time, before it goes viral. That’s the advantage of deep social media listening, and sentiment-based insights. Use them well, and the state of your brand will always be strong too.

Does NetBase get your vote? Still deciding? Reach out for more about our social media monitoring tools!

Image from Nicholas Raymond

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