Menu
Back

Text analysis examples for better business intelligence

The most successful organizations have a range of business intelligence (BI) tools at their fingertips to help them make informed and innovative decisions. From cutting-edge performance management tools to the latest in reporting software; intelligence tools are a necessity, not a nice-to-have. But there’s one crucial element of the BI toolkit that is too often overlooked or even missed entirely. Text analysis.

Also known as text mining, text analytics refers to the systematic, efficient, and objective analysis of large volumes of text data (in other words, words). It’s a powerful way of bringing quantitative rigor to your qualitative data to help reveal the ‘why’ behind the ‘what’.

Harnessed to brilliant effect by market research, customer experience, HR, and marketing teams in particular, text analysis tools help you to make the smartest decisions.

Perhaps you’re already using text analysis yourself, perhaps you recognize the need for it in your BI toolkit and want to learn more, or perhaps you’re wanting to explore some common text analysis examples…

Whatever the state of your current business intelligence tech stack, continue reading as we explore the most popular applications of text mining for data-driven businesses.

What is text analytics used for?

While there are near-endless use cases, we see three primary types of text analysis with our customers: customer insights, market research and employee insights.

Let’s explore these three real life examples of text analysis in more detail…

1. Customer insights

Text analysis for customer insights is all about understanding what customers want, think, feel, like, and dislike – and how these elements might differ across different customer segments. Online customer reviews and satisfaction surveys are a classic example of text data that could be analyzed for this purpose.

2. Market research & target audience understanding

This type of text analysis can help you understand what your target audiences are saying about the topics that are important to your business. Open-end responses in a market research survey are a common example of this type of text data.

3. Employee insights

Text analysis for employee insights is about understanding what and how your employees are feeling, as well as learning how this might differ across teams, functions and roles. The results of an employee survey are a classic example that would be used for this purpose.

While by no means an exhaustive list, these three types of text analysis are some of the biggest and most effective examples that businesses are using today.

Insight-rich text data examples

There are countless examples of text datasets that hold a goldmine of unearthed insights.

Words are pretty much everywhere you look, meaning that many businesses are sitting on a multitude of text data types. Some of the most common places to find text datasets are in surveys, social media, online reviews, customer service transcripts, and marketing content. And hidden within these text datasets is a labyrinth of sentiment, emotion, sub-conscious bias, brand perception, and behavioral preference. Within text data examples lies razor-sharp insight into all of this.

If your goal is to draw out customer insights, you might look to open-end survey responses, where reams of valuable words are just waiting to be mined to tell you more about your customers. Another of the most powerful types of text data is online reviews, which contain thousands of words, detailing your customers’ thoughts, experiences, and feelings. You can also unpick your NPS (Net Promoter Score) or CSAT (Customer Satisfaction Score) to gauge customer loyalty and contentment on a more granular level. And if you have customer service agents or chatbots, you can even analyze the seemingly innocuous text data to reveal the priceless (and often emotionally charged) insights into what your customer wants, needs, and feels.

If you’re more interested in conducting a general health check on your industry through market research, a key text data example is the heaving universe of social media. Between the words, emojis and memes that your target audience uses across Twitter, Facebook, Instagram, and a multitude more platforms, you can find public opinion in its most organic form. There are also insight-rich open-end survey responses and mediums like online forum discussions, where your target audience is directly having their say. Text analysis gives you access not only to what your audiences are saying, but how they’re saying it – all at a vast scale.

And finally, if it’s employee insights that you want to reveal, you can look to the text data contained in employee engagement surveys, or in the documents and transcripts of 360 feedback or performance reviews.

In short, there’s no short supply of text data to hand when you look a little deeper. And it is when you look deeper, that far-reaching insights can be unearthed, actioned, and used to make a tangible difference to businesses. Here’s how.

Text analytics use cases for improving customer experience

You’re never going to have a product or service that every customer loves. But by comparing positive feedback against the negative, you can reveal the underlying drivers of positive and negative customer experiences. These insights can then inform your customer service strategy, product development, and brand positioning.

Similarly, by comparing from customers across different countries, regions, or markets, you can get under the skin of how and why the customer experience varies across international operations, and take steps to make improvements.

These comparisons and explorations don’t just have to be a one-time thing either. You can keep a pulse on customer experience by monitoring changes in customer feedback over time. This can be especially useful for assessing and quantifying the impact of a specific business change or new initiative such as a product launch.

Text mining applications in marketing

A huge advantage of using textual analysis in market research is that it allows you to analyze public discourse around a topic of interest to your business, both efficiently and objectively. By understanding how your audiences talk about topics you care about, you can develop marketing communications that have more impact, and that genuinely reflect the needs and expectations of your target audiences. You can also demystify the drivers behind changes in buying behavior as well as anticipate future needs.

Let’s imagine you’re a high fashion brand with a baby boomer target audience who is especially active on Facebook. After analyzing their commentary and shifting responses over time during red carpet events such as the Met Gala and The Oscars, you unearth that this highly engaged and trend-forward audience is critical of certain fabrics being used – fabrics that you use with certain suppliers. Armed with this insight, you could reassess your product development and supply chain, as well as unlock the broader theme of sustainability in your communication with this audience.

Social media text analysis like this is a highly effective method of drilling down into the most important aspects of online conversations in ways that often social listening tools and traditional outputs (like word clouds) simply can’t.

The same applies for text mining in quantitative target audience surveys. Primarily, these text analysis examples in research surround the intimidating landscape of open-end questions – which sometimes leave researchers scratching their heads over the piles of text they’re lumped with working through. However, when analyzed thoroughly and at scale, an open-end question like, ‘How did feel when travelling on your most recent trip on the London Underground?’, can tell us much more than a multiple-choice question like, ‘How would you rate your most recent experience of the London Underground? (1-5)’.

Uncovering employee insights using qualitative text analysis software

While the vast ocean of text data available on social media, in open-end surveys and in customer feedback questionnaires naturally lend themselves to consumer research; there is also an often-overlooked opportunity much closer to home: employee feedback and performance reviews.

Much like with customer feedback surveys, businesses often invest much more time in designing employee questionnaires and collecting employee feedback, than they do in actually analyzing the results.

However, this application of text analysis can prove invaluable to businesses. In the case of employee surveys, text analysis can help to create a more inclusive, engaged, and positive company culture. Circulating an employee feedback questionnaire is great for helping employees to feel listened to; but they also provide an opportunity to actually listen to employees.

The opportunity exists when employee feedback is reviewed using qualitative text analysis software. Such software helps to identify trends, sentiment and key topics at scale – and also helps to compare attitudes within the business. For example, comparing the feedback of employees who feel positively about the company, against those who feel negatively, helps to spotlight what the business is doing well – and not so well. Additional filters can then be laid on top of this, allowing comparisons between different teams, different seniority levels, and perhaps even different locations within the office building.

Imagine a creative agency, for example, made up of designers and account handlers, split across two floors. Even before a thorough analysis, it is clear from feedback forms that the account handlers generally have a more positive attitude towards their work than the designers. By using qualitative text analysis software, however, this insight is pushed further to reveal that those who are most positive generally talk about career progression and social opportunities, whereas those who are most negative often refer to environmental conditions like bad light and crowded desks.

In part, this appears to be a cultural difference between account handlers and designers; the former are more sociable, the latter more sensitive to their aesthetic surroundings. However, when an additional ‘geographic’ filter is applied, a different truth is revealed. The complaints about environment are coming exclusively from those who sit downstairs – it just so happens that the majority of designers sit here, while most account handlers sit upstairs. The solution? Downstairs needs a facelift.

Without a thorough qualitative text analysis, the conclusion may well have been frozen at the assumption that account handlers have a more positive outlook than account handlers.

Ready to complete your BI toolkit?

Text mining can unlock new perspectives for businesses in any industry. It is a useful tool for turning instinct into insight – for really listening to audiences, understanding and comparing what they’re saying and how they’re feeling, and removing the guesswork.

Thousands of businesses are already using text mining to make better and more confident business decisions. If you’d like to join them and add text analysis to your BI toolkit, get in touch with us to book a discovery call.