Enterpret
Product Updates

Launched Custom Cohorts for more flexible analysis

Easily segment and analyze customer feedback without the headache of manual tagging or complex queries.

• Effortless Segmentation – Quickly categorize large datasets without manual tagging.
• Flexible Metadata Mapping – Define metadata keys just like in Amplitude & Mixpanel cohorts.
• Enhanced Filtering – Use cohort tags to filter and analyze data more efficiently.
• Saves Time – No need for complex queries—just upload, map, and filter.

Common use cases for Custom Cohorts in Enterpret:

• Win/Loss Analysis
• Churn Risk Analysis
• Beta Tester Feedback
• Feature Adoption Tracking
• VIP Customer Monitoring

Enabled SCIM and Data Export to be Self-Serve

SCIM and Data Export is now fully self-serve via Workspace Settings, ensuring a more secure and efficient process and better support for enterprise organizations

Shipped improvements to Wisdom models

We continue to ship improvements to Wisdom's underlying models. Here's what's new:

• ‍No more empty plans: The new Wisdom models always generates a plan—even when the previous version couldn't. No more dead ends!‍
• Understands quotes in context: Wisdom now recognizes quotes in context—it knows which tool to use and how to frame the question when handling quoted text.‍
• No more streaming: Analysis showed that streaming the first step takes nearly the same time as generating the full plan in 90% of cases. So, we’ve stopped streaming. The result? We can now introduce validators on the generated plan—making it stronger and more reliable.‍

Launched the ability for Wisdom to recognize and apply Custom Filters

‍Previously, Wisdom couldn’t detect or apply Custom Filters mentioned in queries. Now, it can identify and use them—making responses more precise and relevant!

Example: Your team has a cohort of defined users

Note: Custom Filters are user-defined filters. Instead of manually specifying the same filters repeatedly, users can create, save, and reuse these filters for more efficient querying.

Shipped a new Homepage for quicker speed to insights

We've introduced a new homepage experience to help all users get the most valuable insights more quickly. Users will see

• Quick access to the Wisdom prompts
• Create a Dashboard from Templates
• Auto-generated dashboard of your organizations more critical insights along with Charts

Streamlined the Wisdom Experience

Shipped new streamlined buttons for more top prompts and easier to find Wisdom Prompt Library.

Delivered a more intuitive flow for Unified Field creation & management

We shipped a usability improvement to make Unified Fields management more intuitive and reduce context switching. Instead of having to navigate to a separate management page users are now able to directly create and edit Unified Fields from: 

• Integrations page
• Metadata table

Continued improvement to Wisdom response times

We continue to make significant optimization in Wisdom's context retrieval system that has resulted in substantial latency improvements.

• Improved Performance: Significantly faster response times across all percentiles (e.g., P50: 28% faster, P99: 31% faster), with noticeable speed improvements for complex queries.
• Enhanced Consistency: More predictable performance across different question types, ensuring reliable and steady response times.
• Maintained Quality: Same high-quality responses you expect from Wisdom, with strong context quality preserved and minimal recall impact (54.4% vs. baseline 55%).

Bug Fixes and Improvements

• Streamlined the experience to remove a widget from a Dashboard
• Released an update to the saved artifact searches to be sorted by the most recent
• Unveiled optimizations for Workplace Setting UX
• New UI for main navigation menu

Shipped improvements to our Query models

We’ve made significant enhancements to our Query Models for sharper subquery generation, smarter keyword handling, and better performance.

• Sharper Subquery Generation:  No more filtering on generic terms or organization names—say goodbye to irrelevant results like “Feedback from customers” or “Feedback from users.”
• Smarter Keyword Handling: The new model handles specifics more effectively. Example: “What is the total number of support tickets in the last year?” The new model is capbabe of zeroing in on the specificity of support tickets and timeframe.‍
• Query Performance Boost: Enjoy a 2% gain in accuracy without any compromise on overall performance.

Launched Taxonomy Estimator for improved Label processing

We’re excited to roll out our new Taxonomy Estimator for label processing! Previously, whenever a label was edited, merged, or newly created, we reprocessed and re-tagged feedback records based on updated information within an hour (though we occasionally missed this target).

Now, with our latest update:

• Accurate Time Estimates: The Taxonomy Estimator provides iterative time estimates for label processing, giving you a more accurate idea of when your updates will be completed.
• Improved Reliability: If label processing gets stuck or fails, the system automatically restarts to ensure smoother and more reliable operations.

This update sets realistic expectations and reduces frustrations around label processing delays, ultimately enhancing user experience.

‍

Create Anomaly and Compare Charts in seconds with Wisdom

Wisdom is our co pilot for customer insights that delivers deep customer insights in a simple, intuitive chat interface. Ask any question and get answers to customer questions without the complexity of an analytics tool. Here's three ways you can use this update: ‍

• ‍Spot Anomaly spikes & dips: Have questions about sudden changes? Now you can easily ask Wisdom to identify and highlight spikes and dips in your data!‍
• Data comparisons within same time period: Compare different reasons all within the same time frame. Spot trends and gain insights effortlessly.‍
• Compare across different time periods: Dive deep by comparing reasons and keywords across different periods. Discover how patterns evolve over time.

More transparent answers in Wisdom

The latest enhancement in Wisdom makes every interaction more transparent and reliable. Answers produced by Wisdom will now provides more context and understanding in responses in way that makes it easier to consume.

• ‍User Quotes Integration: Wisdom now includes more user quotes to make responses clearer and more insightful.‍
• Accurate Feedback Citations: Feedback citations are more accurate, representing user expectations effectively.‍
• New Hierarchical Structure: Answers now feature a three-level hierarchy for better organization and ease of skimming.‍
• Refined UX: We've updated the UX for improved readability of quotes, ensuring users are not overwhelmed by multiple entries

Use Wisdom to generate filters in the Query Builder

‍

Anywhere you can search, you'll now be able to use Wisdom to auto-generate Filters.

Try

• Create a filter for Tracked Keyword [ x ]
• Create a filter for praise
• Create a filter for feedback from the App Store.

Wisdom Performance Update: 50% Faster

We continue to make investments in Wisdom's performance - we're now seeing Queries run ~50% faster!

Shipped "contains" filter in Query Builder

Our highly requested "contains" filter operator is now available in QueryBuilder! Users can perform substring matching on metadata field values directly within QueryBuilder, unlocking greater flexibility and precision in your searches.‍

Improved Comparisons usability on Quantify pages‍

We've made the following enhancements to improve the experience of comparing data sets when using Quantify: 

• The comparison period (Duration B) now automatically adjusts to maintain a consistent relative gap when updating the global date filter.
• No more manual fixes or error messages—just smooth and seamless comparisons every time.

Added sorting by Metadata fields on Quantify

You can now sort metadata fields by labels, offering a more intuitive experience when you're looking to organize and analyze your data efficiently.