This guide will teach you why and how you can use text analysis to understand your customer feedback and interactions, reviews and other user-generated content in a matter of minutes, without a single line of code, no expensive software and no data analytics skills.
By the end of this guide, you’ll be armed with the information needed to extract meaningful insights from texts to learn what your customers are thinking and feeling, enabling you to get to the heart of the matter and inform your decision making with data.
First off, what is user-generated content?
User-generated content (UGC) is any content, like text, reviews, feedback, customer chat logs or blogs, created by your customers or your users.
From social media feeds to product reviews and customer support interactions to blogs, UGC is ubiquitous, but it is often woefully underused.
Whether it’s to refine a product or service, develop a new feature, or figure out where consumers find value in your offering, UGC can be utilised for many purposes. I’m sure you’ve seen the testimonials and highlighted reviews on a website or storefront, explaining why a product is great or how a service made their life easier. Perhaps you’ve even handpicked a few of the best for your site, to show off your selling points and validate your offering. Maybe, as our case study shows, your feedback might help you to improve your service.
The above chart shows each care summary for a client over a 4 month period scored with VAYU’s built-in sentiment analysis. The company found a clear trend in positive sentiment over the past 2 months. Given the nature of care work, it’s inevitable that sentiment analysis will find negative scoring terms, but trends like the above can tell the company as much about the wellbeing of its carers as those they are caring for.
These are all important things to consider. After all, consumers place trust in companies that are authenticated by real people.
But there’s more to it than that. You need to look beyond the star rating and the surface level observations.
This begs the question: what can you do to find REAL value in these texts? How can you go beyond the surface and dig a little deeper?
Text analysis holds the key to understanding
The best way for you to dig deeper into your data is through text analysis.
Without some analysis of the review text, the reviews are ultimately worthless. Yes, they might look great in your testimonials, but are they improving your product or service?
Text analysis allows you to not only categorise reviews, feedback, or customer interactions into the most frequently discussed topics, it also enables you to rank them by sentiment, to understand how positive or negative these topics are in the eyes of your customers.
In doing so you can increase retention and advocacy by really listening.
Simply put, text analysis is imperative for any business collecting user-generated content. So how do you get started?
Step One - Identify the questions you want to answer
The first step is to ask questions like these:
- What do our customers think of our brand?
- When should we launch a new product?
- How can we improve or refine our services?
- Where do our customers find the most value?
- Who should we target in our next marketing campaign?
- Is our product regarded as too expensive or too cheap?
These are all valid questions and ones that need answering, but so often businesses find it difficult to get conclusive answers.
For the purposes of this guide, let’s use a common example; Amazon reviews. Maybe you want to know what your customers are talking about the most or the one feature they find the most important in your product.
Identifying and analyzing drivers like these is key to success, so you must find a solution that allows you to analyze all of your data.
Let’s face it, your customers are generating tons of this stuff, so why not use it to understand them better?
Step Two - Processing your data and choosing a tool
You’ve got reviews flooding in and customer service agents chatting to people daily, generating row upon row of text data in the process. People are talking...are you listening?
Manually handling all of that data is definitely not the answer. Analyzing them yourself is a time sink, and prone to human error and bias.
There are many tools out there that can process text data, but many of them are convoluted, packed full of unnecessary features that are difficult to make heads and tails of. Added to this, they require either training or hiring expensive talent.
This amounts to a significant outlay of capital, resources, and time.
The company uses VAYU to explore its datasets with zero overhead. VAYU’s no-code interface made the data manageable and understandable, all without them having to train staff members.
The above shows almost a year's worth of data on a single wellness measure for one of its clients. Previously, they would’ve been unable to perform these types of analyses.
With VAYU, they have been able to identify a marked deviation in the last quarter of the chart, which could be a predictor for changes in other wellness measures.
Understanding early warning signs can be vital in providing the appropriate level of care, but getting comfortable with a model for generating triggers or safeguarding alerts can be difficult.
We’ve addressed this need with VAYU, a tool so intuitive that any knowledge worker can pick it up and perform data science in minutes.
Once you’ve chosen a suitable tool, gathered your data and processed it, the next step is to take action! But communicating these findings to stakeholders can sometimes be difficult. That’s where data visualization comes in.
Step Three - Visualize and present your findings
The ease with which VAYU enables anyone to visualize their data is the most important thing of all.
Collecting and processing the data is one thing, understanding it is something entirely different.
What’s great about VAYU is that in just a few clicks you can see your insight presented before you in easy to digest graphs, charts or visualizations.
Taking that information and putting it into a visual context like this makes big data much easier for us to understand.
It can also allow us to see patterns and trends at a glance, enabling us to communicate these findings quickly and effectively to other stakeholders.
Layering on your domain knowledge, whether it’s in marketing, retail, eCommerce or whatever it might be, allows you to make data-driven decisions in no time at all.