ベストプラクティス

6 ways to combine multiple types of data to improve your software experience

Published May 5, 2025
When you know what users do, say, feel, and experience, you can build software they never want to leave.

Building great software experiences requires data—but it’s about more than just tracking clicks or collecting one-off survey responses. It demands a holistic understanding of how users think, feel, and behave.

Relying solely on quantitative data like usage metrics might reveal what users are doing, but it won’t uncover why they’re doing it. On the flip side, qualitative feedback from surveys or user interviews offers rich insights into user sentiment and motivation, but lacks the scale needed to validate patterns across a broad audience. The same goes for sentiment and visual data: each offers a unique lens, but none tells the whole story on its own.

In order to truly elevate their software experience, teams need to connect the dots across all of these data types. The most valuable insights come from the intersections of quantitative (what users are doing), qualitative (what users are saying), sentiment (how users feel), and visual (what users experience) data.

Here are six practical use cases for combining multiple types of data—and how to use the Pendo platform to generate these insights.

1. Detecting silent user frustration

The best product teams take a proactive approach to preventing churn. With the right data, you can understand whether and how seemingly successful workflows are actually causing pain or frustration for users. This way, you can prioritize updates, redesigns, or in-app guidance in order to alleviate frustration before it turns into churn.

Find the insight with Pendo: Use a Path report to identify users who are completing tasks (quantitative) but leave negative feedback (sentiment) and show signs of struggle in session replays (visual), like excessive clicks.

2. Spotting hidden feature discovery bottlenecks

Just because you’ve built a great feature doesn’t mean your customers are using it—or even know about it. By combining multiple types of data, you can get to the bottom of why a feature has less-than-expected usage and then immediately take action to address it. For example, you might find you need to redesign a page’s layout or add guided tours to surface key features to users.

Find the insight with Pendo: Use a Funnel report to identify drop-offs at a certain step (quantitative), and then confirm via session replay that users never saw the next step’s CTA (visual). Pair this with survey responses or free-text feedback where users complain about the feature’s “missing functionality” (qualitative).

3. Validating feature requests with real behavior

Your customers likely have a lot of opinions about and ideas for improving your software, and it can be difficult to determine which ones are worth prioritizing. With a breadth of data sources, teams can use desire and usage signals to identify the feature requests that make the most sense for the business—and deprioritize requests from the vocal minority.

Find the insight with Pendo: Validate whether a commonly requested feature in feedback submissions (qualitative) makes sense given other data types. For instance, if session replays (visual) and usage data (quantitative) show low engagement in related areas and sentiment is neutral or negative toward alternatives.

4. Measuring emotional impact of new releases

Even when your adoption numbers are looking good, users could still be having a negative experience with a new feature or workflow. This makes it even more crucial to measure the emotional impact of a new feature. If you see signs of emotional dissatisfaction, you can take action by adjusting onboarding messaging or tutorials to better meet users’ needs.

Find the insight with Pendo: Look for signs of dissatisfaction after a feature launch. For example, if feature adoption data (quantitative) shows strong adoption, but sentiment from surveys has decreased and session replays (visual) show hesitations in the form of confused hover patterns or retries.

5. Finding silent champions and beta testers

There are likely plenty of users who love your product in practice, but don’t necessarily tell you that in surveys or other feedback forums. By looking for signals across multiple data sources, you’ll be able to identify users who are a good fit for early access beta groups, ambassador programs, and other advocacy activities.

Find the insight with Pendo: Identify users who engage with your product heavily (quantitative), leave positive feedback (qualitative), show high satisfaction or are NPS promoters (sentiment), and smoothly complete tasks and flows (visual).

6. Proving the ROI of UX improvements

More than ever, product teams need to connect their work to business outcomes, many of which hinge on customer satisfaction. After making changes to your software or iterating on a specific feature, you can build airtight, data-backed narratives for leadership that prove these investments paid off. 

Find the insight with Pendo: After a UX redesign, look for faster task completion (quantitative), an increase in sentiment, an uptick in positive feedback (qualitative), and visual replays that show more direct, confident interactions (visual).

 

With Pendo, your quantitative, qualitative, sentiment, and visual data all live in a single platform—helping you uncover powerful insights, faster. Get a demo to see it in action.