How to Choose the Right Text Analysis Software

Analyzing text sounds like an easy task to do. You just look at the text, find some patterns, and categorize the text based on those patterns. If you have received a small amount of free text feedback from your customers, you should be able to do this without too much difficulty or dedicated resources.

However, let’s say you’re working in a company that is growing fast, and the amount of feedback is increasing – you are feeling the pressure now- you’re struggling to keep track of the whole picture and frankly, you’re on the verge of losing your sanity! In these situations, you need to remind yourself that we are living in the 21st century and we have machines that can analyze text much more efficiently and accurately than us humans.

So please, this is undoubtedly the best piece of advice you will have been given in a while – stop using manual labour and processes and get yourself some text analysis software instead! 

In this article, I will tell you a bit more about text analysis – what it can do and how you can benefit from it. I will also tell you what you need to pay attention to when deciding on a text analysis software to invest in.

What is a Text analysis software?

Text analysis software, also called text analytics or text mining software is as the name suggests – used to analyze all different kinds of texts. Lumoa is an example of a text analysis software.

Text analysis includes a set of techniques that structures information arriving in text format, for example, free text customer feedback. The purpose is to convert unstructured text into meaningful structured data to support business analysis and decision making. The structure created by the software is done with methods like sentiment analysis, language analysis or simply look at key phrases.

For example, sentiment analysis involves analyzing subjective material and extracting attitudinal information. Simple sentiment analysis divides the sentiment into three buckets: a sentence can be positive, neutral or negative. Machine learning technologies can detect the degree of sentiment as well: if someone hates your product, the negative sentiment is stronger than if he just dislikes it. Similarly, if someone loves your customer service personnel, the positive sentiment is stronger than if she just feels it’s good.

If you do text analytics in the context of customer feedback, what you typically get is topic volume, volume trends and a sentiment for each topic. The analysis can for example reveal that in March 35% of the feedback was about your customer service personnel, and 70% of that feedback was positive. The results can look like this:

Example of sentiment analysis

What can you do with a text analysis software?

You can use a text analysis software for different purposes. For example, some use it for research purposes, where you have a lot of transcribed data and the software helps you to structure the data.

Text analytics software is also used for business purposes. Companies that receive customer feedback in an open text format are perfect candidates for text analysis software and would reap many benefits.

How do you choose the right text analysis software

Finding the right software is a real challenge! Every business is different and reasons for buying a text analysis software are often different as a result. Maybe you need a software that is able to analyze text in multiple languages or maybe you want a software that can connect the analysis to some important metric. Maybe you need both?

Here is a summary of important things to consider: 

1. Analytics and reports

I would say that this is one of the most important features to consider. If you think about it, you have this awesome software that can magically turn unstructured data into structured data. Great! You know how it works, but are you able to show your co-workers and superiors that it’s working and what it means?

Most of us working in business have some KPIs or metrics that we are working towards. So, it’s good if the analysis software you invest in allows you to you link the data received from customers to these KPIs or metrics (perhaps NPS or CSAT) that your company is trying to achieve.  So that you can see what is driving the metric up and down and take action if necessary.

Tips: Tie your data to a KPI so that you can measure and report your development. You can easily do that with Lumoa.

Another important thing to consider is how the data is displayed once it is analyzed.  Maybe you want to be able to show your colleagues how the NPS score has performed during the last three months and how it spiked the other day when you implemented the changes in the online store. Is the visual impact of the analytics important if you are trying to share insights easily with senior management and obtain buy-in for a project?

2. Actionability

Text analysis software doesn’t only categorize and structure your data, some text analysis software can even tell you what you need to do next.

You need to consider, based on the insights generated by the text analysis software, how easy is it to prioritize and take action on the issues that you just detected? Can the software somehow prioritize and point out what issues you need to fix and improve?

One of the most important outcomes you want to have from a text analysis software is to know if you need to make any changes to the current ways you are operating. You can also use the insights to justify your arguments and to make fact-based decisions that will help you take action.

By having a software that is able to show you what is more important vs. what is less important (with regards to the metric that you are trying to achieve), you won’t need to spend time on figuring out what to improve first, the software will point out what is the most critical issue to fix right away.

3. User Friendliness

Depending on who is going to use the software, you should consider the user-friendliness of the software.

Is it only the customer analytics department (or the person analyzing customer feedback) that is going to have access to the tool?  What kind of technical knowledge does this person have?

Some analysis software can be more technically demanding than others, some may even require you to be able to do some basic coding in order to use. If this isn’t practical you may need to look for software that is built for users that don’t necessarily have any special technical skills. If the software is going to be used in different departments (perhaps including the customer service department, sales department and the management team) make sure that the software you are getting considers their capabilities.

4. Analyze Text in Different Languages

Most text analysis software can analyze text in English. However, let’s say that you are operating in a market where English is not the first language, then you maybe need a software that is able to analyze text in some additional languages. Or maybe you are working for a global company and you receive customer feedback in a multitude of different languages. Naturally, you need to make sure that the software is able to manage all of those languages specific to your business.

Translate text from different languages to English – Quite handy right?

Some software is able to offer text analytics in the more popular languages such as English, French and Spanish. Which is great if you need to be able to process text in just those languages. However, if you operate across the whole of Europe for example, you will need software that can process many different languages – some software can analyze up to 60 languages.

5. Filter Comparison

In business, it is important to be able to measure your results and compare them to previous results, or results from another business area or product type, so that you can look for trends.

Comparison features definitely add impactful value to text analysis software. You want to be able to identify differences between different departments, products, countries or survey sources so you can try to figure out what is causing them.

Example: compare trends and differences between countries

Something to think about is how customizable you want the comparison filters to be. Some analysis software has a default set of categories that you can use, with other software it’s more customizable and you can amend the filters so they are more relevant to your business.

6. Set-up time

Even though this may not seem like the most vital factor to consider when choosing a text analysis software, it’s still important. There is a wide range of text analysis software available on the market, some which are easier to set up and start using than others.

Let’s say that you’re searching for a text analysis software and you found a good candidate with all the features you need. However, the salesperson has indicated that it will take 1 month before you will be able to see the first results from the analytics, and then they need to fine-tune and optimize it for your specific needs which will add another 3 weeks. In the end, it will take closer to two months before you actually can start using the software.

In comparison, there is another software vendor that promises you to have the software up and running within 2 weeks.

As I mentioned, this isn’t maybe the first thing you would think about when you’re about to invest in an NPS software, but it’s definitely something to consider. Remember the length of the set-up time could also be an indicator of how fast the vendor’s customer service is – if you ever need support!

7. Tagging set-up

This is a topic that is pretty much related to the set-up time, but I will dedicate this subtopic its own small section because it’s so important!

What do I mean with tagging? Even though you might think a software with machine learning features can do everything automatically, this is not always the truth. As I mentioned earlier, a text analysis software establishes categories based on the text it analyzes. To make the result optimal for your industry and specific company, most services require that certain words, word pairs, and longer word combinations are linked to categories (this is called tagging) so that the machine starts learning. However, words do not mean exactly the same thing across all industries: “charge” refers to different things if you work for a smartphone manufacturer, for a bank or for a retail company. Similarly, “Apple” should maybe be tagged with “competition” if you are Huawei, but with something else, if you are Apple Inc or a fruit retailer.

Depending on the software you are considering, some software providers offer tagging as a service to their customers. Some do not.

If you choose a software vendor that can help you with this, then you need to check with them if they are able to customize categories or if they just use some industry-standard tags. If you choose a software vendor that offers you industry-standard tagging, the result can be fine as long as the terminology you use inside your company is pretty standard. If they are able to customize to your needs – meaning that you indicate what types of words fall into what kind of category – that is the absolute best!

If you are left alone with the task of tagging your words, you need to be sure that you have enough resources to do this. Some text analysis software can be set up and in use within hours or days, however, there are others that might require months of training and fine-tuning. Remember the opportunity cost, this is valuable time you could be spending actually improving your customer experience instead of teaching the software. If you decide to do this on your own, check with the software provider if you need to do this in all the languages separately or whether establishing the rules in one language is enough.

Brief summary

As you can see, there is a lot to consider when you are choosing a text analysis software. The most important takeaway from this article is that you really need to make sure that you are clear about your needs and the specific features that you require. It’s so important that you do your research!  Use review websites such as G2 Crowd and Capterra to find out what users have to say about specific software.

And good luck, if you choose wisely the impact on your customer experience could be a gamechanger!