Have you heard of Thematic analysis? If you are working in insights or data analysis, you will have heard of Sentiment Analysis. It answers the question “Is this mentioned in a positive or a negative light?”.
Sentiment analysis is important because companies want their brand being perceived positively, or at least more positively than the brands of competitors. Sentiment analysis, if accurate, can be a valuable tool for this specific use case.
But what about Thematic analysis? It complements sentiment analysis by answering the question “What are some recurring themes that people mention?” There are broad themes like “pricing” and specific themes like “competitors are cheaper”.
The more specific the theme, the more useful it is.
And here, I’d like to argue that thematic analysis, if accurate, always beats sentiment analysis, for these three reasons:
1. Thematic analysis is more nuanced
Let’s say, you can learn who is talking about “pricing”. Combine this knowledge with sentiment analysis and you learn whether people are happy or not happy with your pricing. At Thematic, we extract themes that are considerably more specific. We can tell you how many people mention themes such as “great pricing”, “pricing is ok”, “poor pricing”, but more importantly, themes such as “good value for money”, “competitors are cheaper”, “price increases” and “deals to new customers”. And the more detailed your knowledge the better decisions you can make.
2. Thematic analysis is more actionable
Knowing whether something is positive or negative does not answer the question “Why?”. If you don’t know why people like or dislike something, you can’t act on it. Consider these two examples:
“great customer service” vs “poor customer service”
“good user interface” vs “bad user interface”
Great sentiment analysis tools can categorize your feedback in this way. But knowing this raises more questions rather than provides answers. “What is it specifically that we are doing well?” “What is it that customers expect that we don’t deliver?”
Thematic analysis can answer these questions. It can find themes that impact customer service positively, such as “great market knowledge”, “answers all my questions”, or those that impact customer service negatively such as “long wait time”, “need to call multiple times to resolve an issue”.
If you know what these themes are, you can monitor your team’s performance over time to drive real improvements.
When it comes to themes related to user interface, unexpected themes such as “cannot find your phone number” can uncover valuable unknown unknowns.
3. Thematic analysis is more accurate
Finally, thematic analysis can be more accurate because it can capture themes that sentiment analysis can easily miss. The reason a simple sentiment analysis can miss things is that it lacks common sense knowledge. It’s incredibly hard, if not impossible, to teach computers common sense. It is common sense that tells us that “slow” is a positive adjective in the context such as “slowing down for passengers”, but a negative adjective in the context “slow loading time”.
Sentiment analysis will also fail when it comes to specific actions, whereas thematic analysis will capture them. For example, in responses to a student survey, Thematic found themes such as “spread exam days evenly”, “more practical courses”, or “student registration admin”.
In summary, without thematic analysis, sentiment analysis isn’t as actionable as it can be and can even lack accuracy.
So if you need to analyze people’s feedback accurately, make sure that you are discovering emerging specific themes.
After seeing how powerful thematic analysis can be to deliver insights from customer surveys, we decided to name our company after it, but it’s not just thematic analysis that we do now.
Get in touch to learn how we can get the most out of your customers feedback.