Every organization wants to uncover actionable insights when it comes to their content. You may already be monitoring your content analytics to understand the performance of your content marketing efforts, but there’s more you can learn.
Unless you’ve been living under a rock for the last decade, you’ve probably heard of IBM’s Watson™ computer system that was originally launched in 2011 to compete on the gameshow Jeopardy. Since then, a variety of commercial applications have been produced with Watson’s deep learning and natural language processing to solve real-world challenges.
In this post, we’re going to take a quick look at IBM® Watson™ Content Analytics, an enterprise analytics solution that can unlock the value of both unstructured and structured content to help you save time finding information, identify new opportunities, and improve your content marketing efforts.
What Is Watson Content Analytics?
The platform, officially called IBM Watson Explorer Content Analytics, enables the collection and analysis of structured and unstructured content for various channels, including documents, email, databases, websites, and more. Content analytics in this scenario is the combination of linguistic analysis and enhanced computational power.
The robust platform can be connected to a wide variety of content repositories and then proceeds to crawl and import content, parsing it and analyzing it along the way. The result is a searchable index and a holistic view of your entire enterprise’s content, with the data available for both analysis and search. Users can find what they need much more quickly, in context, and then act.
Why Does Your Business Need to Use Content Analytics?
You’re creating and generating more and more content every day, while also struggling with the best way to manage, find, and understand how it is impacting your business. One of the biggest hurdles in the enterprise content world is how to capture and analyze unstructured content.
Think of unstructured content as any article, blog post, email, picture, PowerPoint, video or social post that your organization produces. It’s pretty much any type of content data that doesn’t fit neatly within a column or row of a database.
A large percentage of the information your business holds is in an unstructured format. Now, since a growing percentage of the content we produce is digital (ex: it ends up in some form of content management system) and we can connect javascript-based tracking to capture content marketing metrics, some analysis can be done to understand “what happened”.
But this scenario has plenty of analysis gaps and certainly doesn’t account for all the other types of unstructured content that live within your company’s four walls. Bottom line, it’s hard to search and analyze this data because of its lack of structure, keeping you in the dark when it comes to new insights. What solves the gap here is automation.
Before the existence of AI and machine learning, this type of evaluation of unstructured data would require human labor, reading and understanding each element. This approach isn’t sustainable because of the cost and time required.
Now, a cognitive-computing platform like Watson Content Analytics can be introduced to streamline these activities. For example, let’s say that you’re a CMO of a manufacturing company and you have a traditional marketing and sales dashboard that is showing you that you are missing sales targets. The natural next question is “why are you missing them?”
When you have a enterprise data repository that is pulling in all unstructured content across your organization and analyzing it, you may quickly recognize that customers are expressing negative sentiment for product failures within emails, support forums and on various social networks.
How Can You Use Watson Content Analytics to Improve Your Content Marketing?
Watson Content Analytics can be a valuable tool for your business. You can leverage its capabilities to dig deeper and find content gaps, new opportunities to be more competitive, provide the best customer experiences, and gain market share.
Sentiment Analysis
Within the application, you can analyze sentiment, something essential for content marketers as they strive to deliver useful content. But it’s somewhat elusive to capture this kind of information.
Now, you can use the tool to review all content regarding specific products by using a custom annotator. Think of a custom annotator as a tag that you place on a piece of content when a certain product has been referenced. Everything associated with that product would be compiled and then categorized as positive, negative, ambivalent, or neutral. You can also look at the specific details of positive and negative words and phrases and associated trends. You can uncover “why” customers praise or critique your solutions.
A good use case for sentiment would be your social media profiles, especially if you have a highly engaged community or depend greatly on reviews of products. Compiling all this data from multiple websites would be a heavy lift, but now it’s possible with Watson. Imagine what you could learn from analyzing all the content about how your customers feel about a product.
Drive Higher Customer Engagement and Loyalty
Every brand is tirelessly seeking engagement and loyalty from its audience. It’s a fiercely competitive world, and Watson Content Analytics can be of assistance to meet these goals. Watson is a cognitive learning solution that allows for interaction between the platform and humans, which can then accelerate expertise and optimal outcomes.
One way businesses are executing on these cognitive capabilities is by embedding interpretive features such as image recognition or question-answering natural language in applications. With these enhancements, data, analytics, and cognitive insights are aggregated at scale.
Why does this matter? Here are a few examples.
1. Information Gathering
Employees spend too much time looking for information. According to a McKinsey report, workers spend around 19 percent of their workweek searching and gathering information!
Image Source: McKinsey
You can decrease this with cognitive solutions, allowing your content, sales, and customer service teams to have a unified view of customers and products in near real time. With this knowledge, you can detect patterns around buying activities, understand customers better, and anticipate what they’ll do next. This puts you in a position to engage and win them over.
2. Identifying Attributes that Drive Engagement
Let’s say that you’ve pulled the last 12 months of Facebook posts in an effort to analyze which posts drive the most engagement. Obviously each post has a myriad of attributes (the text, the images, hashtags, mentions, etc.), as well as the performance metrics. Just imagine trying to analyze this manually.
Instead, you could plug this data into Watson and immediately start looking for correlations. Here’s a quick video from one of our favorite data geeks, Christopher Penn.
Identify New Topics with Unstructured Data Analysis
Watson Content Analytics can be a great complement to your content marketing platform as it allows you to look at unstructured data. That’s the key part of what you’re missing, and it can guide your content topic creation.
Content planning is no easy job. It’s always agile as you never know what will happen to change your customer’s priorities completely. That’s especially true now with the COVID-19 pandemic. So how can this unstructured data help you create more relevant content?
There are crucial insights that are “hidden” without a tool like Watson; it’s in your emails, chats, call center logs, customer feedback, and more. Watson allows you to analyze all this unstructured content so you can see trends and patterns.
With this approach, you have a more in-depth view of what matters to customers right now. For example, you could analyze customer service transcripts to find the most common complaints, then develop content that answers these concerns.
Advanced content analytics could be a big piece of your content strategy. They provide you a comprehensive view, taking into account structured and unstructured data. Structured analytics offers information on the what, where, and when, while unstructured analytics delivers the why and how.
Looking for more great content marketing content? Subscribe to the DivvyHQ blog today.