Let’s Have Another Talk About Content (Text) Analytics


Let’s Have Another Talk About Content (Text) Analytics

By Marc Edwards, B.Sc. , PMP, Content Analyticss, bpms, MBA

In prior posts, I talked about content analytics, specifically in healthcare, and how it could impact you. That being said, this is still a topic that can confuse quite a lot of people and I would imagine that trying to grasp such knowledge can be rather intimidating for some. So I decided to take another stab at content analytics and try to break it down as much as possible so that someone reading this can take this knowledge and implement it into their own work.

So…what do I mean when I say “content”? I won’t go into yet another definition of the exact word itself; rather, I will just say that for the purposes of this blog post, content = text (sort of). Think of all of the forms and web pages and word documents that you come across daily. Every single one of those things has text. The problems arise when you want to get ahold of all of this text to do some analysis with it but it’s difficult to do because there doesn’t seem to be any structure surrounding it. For most of us, it’s not instinctively clear how one could structure and organize a social media feed to be analyzed by a computer. How important would it be to you if you could analyze your organization’s social media feed to determine what people are saying about your company? Would you want to be able to predict consumer behaviour through such feeds? Would you be interested in discovering something new about your consumers that will lead you to enhance your services to better serve their needs? I think most of us would say “yes” in a heartbeat.

Text and Why It Matters

Text (words, sentences, paragraphs, documents, etc.) is most often in the form of unstructured data. This loosely means that it is not found in a structured database, with each word or sentiment in its own field. You may find text in the Comments section of a form but the sentiment of the text will still be unclear. However, just because text isn’t necessarily structured doesn’t mean that it has no business value. It’s just that text is more difficult to analyze because it has to be read, understood, and contextualized before it’s true value can be realized. All of this requires a lot of preparation.

There are two approaches to dealing with text that will have to be addressed before you start with your content analytics initiative:

  1. Do you want to start with the data and try to figure out what it can tell you, or
  2. Do you want to start with your own business knowledge to try to understand the data?

Whatever approach you choose will depend upon the business questions you are trying to answer; typically I find that starting with your business knowledge and business questions is the more effective route to take, especially when it comes to text and especially if you don’t know much about the data.

In a way, when you do text analysis, it’s similar to the work done when a graduate student conducts research to complete a paper. Meaning that any form of analysis done requires prior human knowledge, any result is not necessarily a good result, and you have to validate your results and make sure that they are replicable. Thoroughness is the key to doing this type of analysis correctly.

Digital or Bust

When considering content analytics, there are a few things that you will need to identify the sources of information that you want to analyze. These information sources must be in DIGITAL format. If they are not in digital format, get them there as soon as possible or consider other sources of information.


Earlier in this post, we touched on some of the benefits of content analytics. Here are some more:

  • You can discover new insights concerning your business and your consumers
  • You can detect new industry trends earlier
  • You can identify new revenue opportunities
  • You can increase customer satisfaction and grow the lifetime value of your customers
  • You can derive more value and understanding from those cursed “Comments” fields

How does all of this sound to you? Does it make you more curious about content analytics and what it can do for you? I will provide further posts that go deeper into this topic but let me know what you think in the meantime.


Leave a Reply

Your email address will not be published. Required fields are marked *