You've successfully subscribed to Thematic
Great! Next, complete checkout for full access to Thematic
Welcome back! You've successfully signed in.
Success! Your account is fully activated, you now have access to all content.
Success! Your billing info is updated.
Billing info update failed.

How to analyze Zendesk tickets & Intercom chats for customer insights

Wouldn’t it be awesome if you could automatically analyze your support conversations to gain deep, actionable insights?

Thematic has just rolled out two new integrations with Zendesk and Intercom. With just one click, you can now send your customer conversations to Thematic for automated tagging and analysis.

Read on to learn how to use our powerful analytics for deeper customer insights.

What is chat analytics?

Chat analytics is the process of finding useful data points in chat conversations and support tickets. For example, it’s useful to know the main reasons your customers are looking for support. Or what kind of requests take the most time for the agents to resolve. These insights help Product Managers to prioritize issues that matter. They also reveal how companies can improve customer experience.

Support conversations have significant advantages over customer feedback collected through other channels:

  • Customers find surveys tedious and often give low-quality responses. In support tickets, customers provide accurate information to get the help they need.

  • Feedback in support tickets and chats are more likely to be unbiased compared to reviews. Usually only extremely satisfied or extremely dissatisfied customers leave reviews. But most of us have had to reach out to support at some point, regardless of how much we love the company.

  • Support ticket formats means customers can freely describe their experience and pain points. This generates richer data than rigid surveys with set questions.

That said, for best insights, you want to gather all customer feedback and analyze it in a consistent way.

Support conversations contain qualitative data about the most pressing areas of friction. But you need the right software to turn this data into insights.

Ultimately, deep customer insights will trigger the right actions and result in positive change.

How support chat data is displayed in Thematic
How chat data is displayed in Thematic

Challenges of traditional chat analytics

Support tickets and chat conversations can be tricky to analyze. There are two main approaches that companies use:

1. Manual tagging or categorization

Typically manual tagging is done by support agents who already deal with large daily volumes of chat and support tickets. Tagging takes away the time they could have spent helping more customers.  

Customer support agents may only tag a support ticket with one or two general topics. Other issues mentioned by the customer may be omitted and nuances are lost. Manual tagging is also subjective as agents may interpret the content of tickets differently.

Finally, a typical company has hundreds of potential tags and more emerge as new products and services are released. Agents need to be trained to apply the right tags and manage an ontology of all possible tags. This is typically the job of a librarian - not a mean feat!

2. Data scientists write scripts to analyze support conversations

You might be lucky to have a data scientist or an analyst available to crunch support conversations. Ideally, they would use off-the-shelf text analytics to analyze data and set up a regular and useful report, but it’s still a challenging task!

A chat conversation or a ticket thread contains a lot of text that’s irrelevant to the issue. For example, automated chatbot scripts, email signatures, or the canned responses agents might be using. All of this needs to be edited out before analysis can begin, which is a topic for a PhD in itself.

Many off-the-shelf text analytics solutions are a black box. Tweaking how the data is analyzed requires many iterations. Bringing in human knowledge requires labelling training data. Whenever new features are released (which in some companies happens daily), more data needs labelling and testing is required.

Let’s take a look at the current analytical capabilities of both Zendesk & Intercom.

Thematic overcomes the traditional challenges of chat analytics - Thematic removes reused text and pre-canned responses. It can detect intent, products mentioned in conversation, and find themes in the chat text.
Thematic overcomes the traditional challenges of chat analytics - Thematic removes reused text and pre-canned responses. It can detect intent, products mentioned in conversation, and find themes in chat text.

Zendesk’s current support tickets analytics capabilities

Zendesk is a customer service platform that pulls your customers’ interactions across channels into a dashboard. These interactions become tickets which can be automatically assigned to customer service agents.

Zendesk Explore is the platform’s analytics and reporting tool. Zendesk users can track quantitative metrics like agent performance and ticket volumes. You can slice your data using a range of pre-set metrics. These include ticket attributes or agent responses and performance.

Limitations of Zendesk’s analytics capabilities

Zendesk Explore is great for managing a support team. However, it is limited when it comes to finding product-focused insights in support tickets:

  • Focuses only on quantitative data such as volume of tickets or first reply rates.

  • Customer service agents can manually tag each conversation with the appropriate topic. These tags are thus limited in scope and subject to human error.

  • Zendesk also offers automatic ticket tagging. The software scans incoming tickets and compares the text to tags that have already been used, adding the top three matches to the ticket. But this system only works if you already have a big bank of tags you’re created manually, requiring a lot of initial effort to set up.

  • No text analytics capabilities to give deeper insights into your customer experience.

  • No sentiment analytics capabilities to understand the feelings driving your customer conversations.

Intercom's current support tickets analytics capabilities

Intercom connects customers with agents via a live chat. Users build chatbots to automatically route customer conversations and streamline customer support processes.

Earlier this year Intercom introduced new features like suggested topics which uses machine learning to analyse customer conversations and identify new contexts where existing topics reoccur. Users can combine this with the customer topics feature to get a visual representation of the key themes their customers are talking about.

Despite these recent updates, Intercom’s analytics capabilities still have their limitations.

Limitations of Intercom’s analytics capabilities

  • Agents can use pre-canned responses with tags already pre-attached to these. However, this will only cover a portion of the conversations.

  • Otherwise agents must manually add or create new tags to add to chat conversations which can result in limited and subjective tagging.

  • Suggested topics are limited to existing topics that have already been added to the platform. For best results you need to have a wide range of topics that have already been manually tagged.

  • Custom Reports allows users to visualize chat metrics like response times, conversation volumes, and top weekly issue types filtered by tags. But users who need stronger analytical capabilities rely on integrations with platforms like Google Analytics or Analytics for Intercom

  • No sentiment analysis capabilities to understand the feelings driving your customer conversations.

How to analyze your Zendesk or Intercom tickets with Thematic

Thematic is an AI-powered feedback analysis solution. It works by applying both thematic analysis and sentiment analysis to your data. This enables you to capture deep customer insights without laborious manual coding.

Thematic identifies the themes mentioned in a piece of customer feedback. The software analyzes word and sentence structures using NLP (Natural language processing) to automatically discover themes in feedback. No predefined taxonomies. No training required.

Themes found in Zendesk ticket analysis and Intercom chat analysis

Thematic automatically discovers themes such as “unable to upgrade” and “upgrade to Gold”, so that you can track this issue.

Once you address the issue, trends analysis in Thematic will be able to show the declining volume over time.

Thematic’s Zendesk and Intercom integrations

Thematic has built-in integrations with both Zendesk and Intercom. You can seamlessly transfer all your chat data into Thematic for thematic and sentiment analysis.  This saves you time and helps you focus your efforts where you can make the most impact.

How Thematic finds insights in your chat data
How Thematic finds insights in your chat data

Benefits of using Thematic for chat analytics

Thematic works seamlessly with your Zendesk or Intercom data to generate actionable insights:

  • Thematic automatically and accurately tags chat conversations and support tickets with themes. This gives you an overview of the key reasons why customers contact customer support.

  • These are broken down into sub-themes. Thematic makes it easy to analyze all sources of friction for your customers in detail.

  • Thematic’ automatically filters out pre-canned chatbot scripts and response templates. This increases the accuracy of your chat analytics.

Here are some of the things that the deep insights from your support conversations can be used for:

  • Discover the drivers behind the negative or positive support conversations. Maybe certain bugs are causing the most distress for customers. Or maybe your communication is unclear or inconsistent and this is generating a lot of angry support conversations.

  • Identify emerging trends as they are occurring, not later. Such as a recently broken feature in your software that you were unaware of.

  • Inform your product roadmap or corporate strategy.

  • Reduce the amount of support conversations occurring by resolving underlying issues identified in Thematic’s analysis

Conclusion: Thematic makes rich insights from chat analytics easy

Here at Thematic we’re always working on new features to help you improve your customer experience analytics. Users can now combine Zendesk and Intercom with the analytical power of Thematic. This makes it easier than ever to get actionable insights from your chat data.

If you’d like to see Thematic in action, you can reach out by booking a demo, and we will demo the analysis of your own data.

Ready to scale customer insights from feedback?

Our experts will show you how Thematic works, how to discover pain points and track the ROI of decisions. To access your free trial, book a personal demo today.

Recent posts

How to theme qualitative data using thematic analysis software
How to theme qualitative data using thematic analysis software
Members Public

Become a qualitative theming pro! Creating a perfect code frame is hard, but thematic analysis software makes the process much easier.

How to super-charge your Qualtrics setup with Thematic & Power BI
How to super-charge your Qualtrics setup with Thematic & Power BI
Members Public

Qualtrics is one of the most well-known and powerful Customer Feedback Management platforms. But even so, it has limitations. We recently hosted a live panel where data analysts from two well-known brands shared their experiences with Qualtrics, and how they extended this platform’s capabilities. Below, we’ll share the

Customer Experience
How we use our own platform and Chrome extension to centralize & analyze feedback
How we use our own platform and Chrome extension to centralize & analyze feedback
Members Public

Customer feedback doesn't have all the answers. But it has critical insights for strategy and prioritization. Thematic is a B2B SaaS company. We aren't swimming in feedback. Every piece of feedback counts. Collecting and analyzing this feedback requires a different approach. We receive feedback from many places: * our in-product NPS

Feedback Analysis