Mobile Menu

Text mining tools: 3 key considerations before taking the plunge

cog icon with question mark on green background

Picture the scene. You’re mindlessly scrolling through your socials when a shiny sponsored advert hits you right in your weak spot. It’s hypnotizing; it’s perfect; it might just be the answer to all your problems. But making a purchase without having a plan of how you want to use it is a surefire way to waste your money. This is especially the case with bigger purchases.

For organizations considering adding text mining tools to their business intelligence toolkit (or any other type of business software for that matter), it’s important to have a clear set of objectives and consider how processes and ways of working need to be adapted to realize the promised value.

While a sports car may be an effective way of getting from point A to B, it’s probably not the best vehicle for a five-person family that needs more seats and storage. When it comes to business software, the same logic applies. It’s important not to be blinded by the headlines and invest the time in really understanding how you want to use potential new tools.

Before investing in text analysis software, we always recommend asking yourself three important questions:

  • What sources of text data am I going to analyze?
  • Why am I investing in text analysis software?
  • Who is going to use the platform?

What sources of text data am I going to analyze (and why)?

All text mining tools demand one core ingredient: text. Even the best text analysis software cannot live up to its potential without a rich input of data to analyze. Anything otherwise would be like attending a tennis class without a racquet.

There are a wide variety of text data sources that can be analyzed, and chances are, you already have access to many. Before choosing the best text analysis software for your organization, identify the data sources you already have and consider how you might enrich your pool of data to get more value from your new tool.

Popular text data sources include:

  • Customer and employee feedback – How often do you survey your customers and/or employees, and do you include open-end questions to yield text data?
  • Social media conversations – Do you have access to these via a social media listening tool?
  • Customer service and focus group transcripts – Do you transcribe your interviews and focus groups in order to get verbatims that can be analyzed?
  • Open-end survey responses – Are you phrasing your questions in a way that encourages both quality and quantity in text responses? Can you segment responses according to associated data points (e.g. region) so that you can compare audiences?

Identifying the sources of data to be analyzed is a good first step, but it’s also important to understand what you want to learn from each. Text analytics can be used for a range of business purposes, such as delivering rich insight into customer experience, market trends, employee engagement, new products, changing behaviors, audience segmentation and more. So it’s important to be specific: what are you hoping to learn from each data source? how will this impact your decision-making? and in what ways will this support your business objectives?

Answering these questions about your data will ensure that the work you’re doing is not only interesting, but also powerfully useful.

Why am I investing in text mining tools?

Choosing to invest in text mining tools can be a big decision – and a difficult sell – especially when stakeholders aren’t attuned to the limitations or problems with the status quo.

Mapping out the primary objectives for buying can help ensure your investment is successful. In the case of data mining tools, these objectives typically fall into one of two categories:

  • Time and cost efficiencies vs. manual alternatives – If you already have a team devoted to manually analyzing text data, a good text mining tool can save time and money while freeing up your team to craft the story around the data for stakeholders and focus on driving actions from insights.
  • High quality insights – Data mining tools not only do a faster job than manual alternatives – they can also do a better job. Being able to apply quantitative rigor to your qualitative data, the best text analytics tools can spot insights that the human eye, word clouds, or sentiment analysis tools cannot. This enables users to better navigate the nuances in customer feelings, thoughts, and behaviors.

Articulating your objectives from the outset is crucial to getting stakeholder buy-in and budget approval and will help you stay focused once you’re using your new platform.

Who is going to use the platform?

While intelligent text mining tools can yield better insights than human analysis, it’s important to remember that they are not a total replacement for human involvement. The most successful adoptions of text analysis software require skilled users who are able to operate the tool effectively and intelligently. Preparing your team for the new adoption is important, and presents an opportunity to break old habits and define new processes. This typically involves:

  • Creating an implementation strategy for your new text mining tool that puts users at the core.
  • Selecting a small group to pilot the tool and forging them into internal experts that can field questions from other users.
  • Considering whether other teams in your business would benefit from using the tool, and getting them involved if so.
  • Planning for how you’re going to share the business objectives and train new users within your company (many software providers, like Relative Insight, will offer dedicated support to help you).

With the right people driving your implementation, you’ll be setting yourself up to realize a good return on your investment.

Setting yourself up for success with text mining

Text mining tools are good for such a wide range of purposes that it’s easy to become vague about how you intend to use yours – or to become too excited by the potential without drilling in to how it might work for you.

There’s no reason why you can’t use your text analysis tool for all of the purposes we’ve explored above, but the important thing is to be specific about your intentions. By answering the three golden questions – what, why, and who – you can prepare yourself for a successful implementation.

If you’d like help in defining the objectives and uses that are right for you, book a discovery call with one of our experts.