Uncover the “missing middle” of customer experience feedback with text analytics | 6 components your TA solution needs

Bennett Gamel | Dec 17, 2020 Bennett Gamel 12/17/20

Analysis of open-ended feedback is critical in surfacing what customers are saying and why they are saying it. And an efficient text analytics solution is a crucial component to gleaning those insights you need to take informed action on customer feedback.

But not all solutions are created equal. You need the ability to dive into every level of analysis—not just the extremes of high-level themes and the granularity of individual comments, but every layer/level of complexity in between.

That “missing middle” is just as important and actionable, but these mid-level insights in open-ended feedback often get lost. The right text analytics solution uncovers these often-overlooked insights by providing these 6 essential functionalities:

1. An intuitive dashboard with real-time, high-impact alerts

With today’s heightened health concerns, operational risks present a significant threat to multi-unit businesses serving a high volume of customers. An AI-powered text analytics platform mines customers’ unstructured feedback and alerts you to time-sensitive comments so you can take immediate action when high-impact, low-frequency issues arise.

This includes food safety. There are few things that can damage a brand’s reputation more than foodborne illness. Real-time alerts bring to light these high-impact, low-frequency events—identifying potential food safety issues before they hit headlines or become systemic issues.

2. Language support for linguistic rules + statistical analysis

The ability to apply natural language processing (NLP) is the best way to capture accurate sentiment within customer responses—both at the whole-comment level as well as granularly by phrase, product, or category. This type of analysis provides users a breakdown with the best indication of the overall customer experience. But to provide the best accuracy in sentiment, these machine-learning algorithms must support variables such as the language of your industry or your geographic presence (e.g., how customers express overall positive sentiment in the U.S. may vary greatly from those in the UK).

Sentiment values can also benefit from adaptive leaning, informing the accuracy of machine-learning algorithms to continuously fine-tune sentiment accuracy and power AI-based alerts (i.e., "This chicken is sick!" vs. "The chicken made me sick.")

3. Domain-specific ontologies that allow for distinction between sub-industries


Unlike the quantitative data you’re collecting through your CX program, findings from unstructured data don’t always fit into a neat little box. More often, they’re spread across the wide array of products, services, and initiatives you’re trying to track—which means you need to ensure they align with how you’re tracking them.

That’s where custom entity ontologies come in. Entity ontologies essentially serve as brand-specific encyclopedias—ensuring the lingo consumers use lines up with your own terminology. With customizable comment groupings, users can:

  • Tag + group comments according to brand-specific product hierarchies
  • Isolate categories + subcategories of interest for deeper analysis
  • Refine areas of focus to keep customer satisfaction trending upward

But many text analytics technologies either can’t handle the volume and complexity of the data being processed or they’re so convoluted it’s difficult to know where to start. More than ever, brands need robust, intuitive tools designed with end users in mind—especially when it comes to input as complicated as the unstructured data found in customer comments.

4. Data correlation across your experience management platform

No one is closer to the customer experience than your front-line employees. By pairing customer experience (CX) data with employee experience (EX) data, brands are able to show higher employee engagement equates to better performance with customers.

For the most intuitive integrated results, revisit your EX touchpoints with an eye toward where and how they overlap with your customer experience metrics. The more synchronized your experience management (XM) strategy is, the easier it will be to find meaningful correlations in the open-ended data.

You’ll also receive critical insights about the customer experience from the unique perspective of your front-line employees (VoCE). Being able to identify and address employee-customer disconnects around your service culture will make it easier to commit teams to improving the customer experience and focusing efforts where they’ll have the biggest impact.

5. Third-party data processing

Your experience management program requires a cross-channel approach that provides a variety of solicited + unsolicited feedback options. Here are a few text analytics must-haves:

  • Speech-to-text: Use machine intelligence to convert recorded conversations to text, making it easier to spot emerging themes and sync up qualitative insights
  • Call center support: Collect feedback at the point of contact—whether it’s phone calls, emails, or chat sessions—so you can see how individual agents, full teams, and even entire centers are performing in real time
  • Video feedback: With video feedback technology, you’ll have the ability to search themes, explore sentiment, and stitch together showreels—driving empathy in your organization and enabling informed action

Feedback data from multiple channels is most powerful when you can combine it and see how one source of data impacts another. With an open API architecture, your XM program can integrate related data into one spot—providing a holistic view of experience feedback and revealing more actionable insights.


6. Ability to contextualize open-ended feedback with industry benchmarks

In addition to being a source of on-demand insights, your text analytics solution should also help answer complex research questions that impact your business long-term. With customized industry libraries and text benchmarks—populated with hundreds of millions of comments—you’ll have a deeper, more contextualized understanding of how customers perceive your brand relative to competitors, providing insights like:

  • How often customers talk about the most important measures for your brand
  • Frequency of employee mentions + how that impacts satisfaction
  • The percentage of customers talking negatively about your staff
  • The categories where customers think you’re better—or worse—than the rest

Mine customer comments for insights across all levels

While quantitative customer feedback data is invaluable, it’s often the qualitative insights from open-ended comments that help you add context to scores and answer questions you hadn’t thought to ask.

The 6 text analytics capabilities listed above help you get to every level of insight, but specifically provide that missing middle. SMG has honed our text analytics visuals/dashboards to guide users to where they need to focus, sorting comment data by categories + subcategories—proving suspected hypotheses in customer feedback and surfacing new data and insights that were previously unknown.

In short, our text analytics technology helps brands turn open-ended feedback into next-level insights—with top-tier accuracy and powerful, multi-source reporting. To learn more, reach out and we’ll schedule some time to demonstrate our capabilities in more detail.

Bennett Gamel | VP, Product Management
Customer Experience Update