Text Analytics

Unstructured feedback allows your customers to talk about issues that you might not have even known to ask about. But how can you quickly make sense of all this data, across thousands of responses? Text Analytics.

The “why” behind the score.

About 80% of business-relevant information originates in unstructured form – primarily text. Feedback contained in places like social media posts, online reviews, emails, and survey verbatims contain insights that go beyond a score or 5-star rating. They provide the texture and insight a business needs to understand “Why” a customer likes or dislikes any part of the customer experience.

Text Analytics allows companies to uncover countless issues, opinions, and opportunities that would traditionally be buried in your data. You’ll know about emerging trends before they balloon into giant problems and capitalize on opportunities you would otherwise miss. Text Analytics helps businesses:

  • Uncover the reasons behind feedback scores
  • Analyze the entire customer experience journey
  • Shorten survey length
  • Broaden feedback capture beyond surveys
  • Identify emerging trends
  • Prioritize investment and allocate resources effectively

A “best fit” Text Analytics engine.

Medallia’s award-winning native Text Analytics engine works across 39 languages to parse insights hidden in text and present them in a format that’s both easy to understand and act upon for teams across global organizations. Our approach, however, is not “one size fits all.” After the Medallia Natural Language Processing engine parses the text that comes into our Text Analytics engine in real-time, Medallia then takes a “best fit” approach in which our engine employs multiple rules based and statistical techniques that are best for getting to the insights business users need to resolve customer problems. Some of those approaches and their uses include:

  • Topic analysis to categorize customer comments into business-relevant topics. There are two general topic analysis approaches:
    • Rule-based – This approach relies on a set of rules to identify the topics under which the system categorizes a comment.
    • Machine learning techniques- These techniques include two specific statistical approaches: 1) Clustering – which is a learning algorithm that automatically groups similar data points together; 2) Classification – which uses predefined categories to automatically classify and categorize data points.
  • Sentiment analysis to discover the positive and negative drivers for business issues. There are two general sentiment analysis approaches:
    • Dictionary/rule based – This approach relies on a predefined sentiment dictionary consisting of positive and negative words with assigned optional sentiment scores based on the dictionary builder’s intuition.
    • Supervised machine learning – This approach relies on a large set of training examples—sentences annotated with sentiment scores as judged by a human. The learning algorithm can then pick up both explicit and implicit sentiment cues, as long as they are present in the training data.

Text Analytics precision and recall that you can bet your business decisions on.

Precision is the proportion of comments that were correctly categorized into a given topic. For example, if a topic analysis system identifies 100 references to the topic “staff attitude” and 90 of the identifications are correct, then precision for this topic is 90%.

Recall measures the completeness of your Text Analytics system. If there are actually 120 true references to a topic “staff attitude,” for example, then the system recall for this topic is 75% (90/120.)

To pursue high precision topic identification and high recall – to get the most insights out of unstructured feedback – Medallia combines the rule-based approach (because it yields the highest precision) with a machine-learning topic tagging system. This combines the advantages of both clustering and classification, leading to high precision and recall.

Regardless of technique though, to ensure that our identified topics are business relevant and actionable, our analysts typically take our clients through a consultative exercise that maps the topics to business processes and customer journeys.

Text Analytics within your customer feedback system.

Medallia Text Analytics is available right inside the Medallia system, so insights hidden in text feedback are readily available and easy to share with people across your organization. There’s nothing new to learn, no new software to install, no integrations to worry about. Medallia integrates scale-based quantitative data including CSAT metrics like Net Promoter Scores®, Top Box or Customer Effort Scores with Text Analytics, so every employee can now easily see the “Why” behind the score.

Understanding the Key Drivers using Text Analytics

Medallia Text Analytics generates intuitive, easy-to-use reports that users can understand with little to no formal training. This means users from the frontline to the C-suite can analyze and act on the root cause of customer issues.

Learn more about Medallia Text Analytics by downloading the brochure.