How well do you know your brand’s audience? You might know some basic demographic information, such as their age-range, geographic location, or loosely-defined ‘interests’. But do you know how they feel, particularly regarding your brand? Sentiment analysis can find out the answers.
What is Sentiment Analysis?
Don’t let the technical-sounding name deter you, Sentiment Analysis or ‘Opinion Mining’ is the process of determining the emotional tone behind a series of words (or even emojis).
This allows you to understand the attitudes, opinions or emotions of users when they post and comment online.
Sentiment Analysis is commonly used by brands to gauge how online audiences feel about them. Naturally, social media is the best place to find out, as self-expression plays a huge role. From comments on your brand’s FB posts to indirectly mentioning you in tweets, customer sentiment is everywhere and publicly accessible for free.
The best part is you don’t need a degree in Marketing Analytics or expensive software to undertake basic sentiment analysis, as we will demonstrate later.
How Will It Help My Brand?
Being able to quickly observe how your fans feel about your latest festival announcement, social media post or product innovation is a major benefit. It’s almost like gathering customer feedback without needing to formally ask for it. This allows you to better tailor your product or social media output and in turn, boost your sales.
Sentiment Analysis can also be used to snoop on your competitors and gain an insight into what they’re doing well (or not well). You can even analyse Twitter hashtags that are related to your industry. For example, you could use #LetTheMusicPlay to understand consumer attitudes towards nightclubs re-opening. The possibilities are endless.
How to Create a Sentiment Analysis Report
For your first dive into Sentiment Analysis, we recommend performing it on one of your brand’s social media posts. Whilst using an analysis tool is faster, doing the first few manually will you better understand the process. It’s also free – you can always invest in software later down the line.
In our example, we analysed the sentiment of We Are FSTVL’s recent FB announcement of a new show – We Are Mexico:
The start of any Sentiment Analysis report should provide some context for the reader. Describe the content of the post and how many engagements it had (likes, comments and shares).
A ‘like’ on your post is pretty self explanatory – it’s a positive reaction to the content. Comments are far more important as their tone can vary wildly. Sarcasm is something you need to be on the lookout for, as you may perceive a comment to be positive when it’s actually negative. There’s no way to know for certain, so pay close attention to the wording and look out for clues.
Emojis and use of punctuation can often reveal the true sentiment of a comment. Take these three sentences:
All contain the same words, but are conveying different sentiments. Summarise how many of the comments are positive or negative. In our example, 100% of the sampled comments were positive.
If you can’t establish the sentiment of a comment, simply label it as ‘unknown’. It’s better to admit you’re unsure rather than taking a stab in the dark.
Identifying Key Themes
Grouping comments by their purpose or theme and displaying this in a pie chart allows you to quickly identify key points:
In our example, most of the comments simply consisted of fans tagging their friends, with a few asking questions. What we can deduce from this is that We Are FSTVL’s audience are hugely excited about the upcoming show, and a few want to know the specifics. It helps to list these questions in your report:
This helps us establish the fans’ priorities. Are they enquiring about the price? The line-up? The date? Use these questions to both tailor your social media output and flesh out your FAQs.
Sentiment Analysis is far more straightforward than many businesses realise. How complicated it becomes depends on the level of detail you want, and the size of your audience. But even the most rudimentary Sentiment Analysis report can reveal how your fans truly feel about your brand’s online presence.