Product Enhancements​

1. Key Driver Analyses
   1. Introducing Confidence buckets (High/Medium/Low) for each segment so you can easily tell whether a segment’s impact on the metric is due to noise or has statistical relevance.
   2. The time comparison selector on the KDA page now allows you to reset the date range to default
   3. The minimum subgroup size now shows up as a top-level configuration item for better visibility
2. Trend Detection
   1. We now also show trends for any 2 and 3 segment combinations! This makes it easier to find those higher order segment combinations that might be driving your metrics.
   2. You can also set the maximum number of segment combinations you want Sisu to automatically analyze in the “Configure Analysis” menu
   3. The effects of monthly seasonality are now accounted for in the anomaly detection algorithm!
3. Dashboards
   1. Resizable Dashboard Waterfall tile: Waterfall charts are one of Sisu’s most differentiated visualizations describing metric change - they can now be resized and better fit into dashboards.
   2. Named dashboard delete modals: When the user is asked to confirm deletion of a dashboard, the modal question now includes the dashboard name, re-enforcing to the user what they’re deleting.
4. API access for get/set Key Driver Analyses Filters
   1. Through a POST request to api/v1/analyses/{analysis_id}/filters users may now overwrite filters in a previously defined KDA/trend
   2. Through a GET request to api/v1/analyses/{analysis_id}/filters users may now view the defined filters on a KDA/trend
   3. This will allow for programmatic bulk creation of similar analyses with different segment filters in conjunction with the duplicate KDA API operation (in-progress)
5. Metric setup
   1. We are working towards making Metric definitions more editable without impacting downstream analyses. As a first step you can now change metric calculation type and the metric dimension!
   2. During metric setup, the input for adding a time dimension now validates that the selected time dimension is in the correct format!
   3. Null Handling in Metrics: When Null or NaN value is present in weighted average or numeric rate metrics, analysis and exploration would return different data for the same metric. This feature fixes the discrepancy by excluding null by default from the above weighted metrics. The user can decide whether they want to include null values in the calculation.
6. Renaming Initiative
   1. Renamed several keywords found in the app to align with industry lingo
      1. Subgroup is now Segment
      2. Fact is now Segment
      3. Column is now Dimension
      4. Slope is now Trend

Bugs Fixed​

1. Core Engine
   • Fixed sampling v1 row count statistic
   • Mitigate email errors from empty data-frame run
2. Dashboard/Exploration Tiles — Fix exploration chart tile popover getting clipped
3. Analysis page — filter and compare fields were using display name instead of actual column name
4. Trend Detection — Users seeing slow trend creation times due to a custom call retrieving start/end trend dates should see now a max of 3 seconds before the new trend is created

Enhancements​

• Trends & Anomaly Detection

  • Charts from trend analyses can now be pinned to Sisu dashboards. This is for the overall trend as well as subgroup trends. This allows the user to create Dashboards that mix together exploratory analyses, key driver analyses, and now trend detection analyses to understand the metric and its change.
  • Added spark lines (i.e. small charts) to the subgroup table so that you can easily visualize how a segment is behaving even before clicking through to the detail page
  • Trend subgroup table is now sortable. You can now sort by trend, impact, subgroup size or even the date of last trend change. This helps you quickly find the subgroups that are driving your metrics.
  • Filtered leading zeros for time series anomaly detection, so charts do not display times before the dataset begins
  • The Trend chart when pinned to a Dashboard can now be resized and can be updated without refreshing the tile.
  • Added a “stop” option when running an analysis
  • Improved error messaging
  • The number of anomalies detected for each segment in the last 30 days is now included in the table of subgroups. This allows you to quickly check how many anomalies a subgroup has had and click through to see more details.

• Key Driver Analyses

  • We now show the user the selected dimensions at the top of the KDA settings
  • Added a “stop” option when running an analysis
  • Improved speed to Group and Ungroup subgroups in the Segment table
  • Natural Language View now displays an indicator for Top Driver
  • Now hide Time Period Option for Time Compare KDAs
  • Updated the custom bin size in KDA dimension settings to support decimals
  • Added a y-axis to the KDA overview charts

• API (pysisu)

  • Trend Detection results are now available in tabular pysisu format
  • Added documentation for dimensions endpoints in the API documentation.
  • Our python library, pysisu, now includes methods to retrieve metrics, datasets, data sources, and dimensions. Further, analyses now have the metric ID associated with the analysis to provide a more complete picture

• Administrative

  • All members can now see organization settings. Admins only continue to have edit access to these settings
  • Now you can duplicate dashboards if you want to create several similar dashboards.
  • Substantial data ingest speed improvements for Databricks
  • Admins can now bulk upload users to be invited to an org in the “user and permissions” section.
  • If you move an analysis to a new project, the new project is hyperlinked to help navigating there.
  • Right Click on Global Search to open in a new tab
  • Users can now duplicate dashboards with 1-click from the Dashboards page.
  • Metric setup: Now the user only sees time stamp columns when selecting the time dimension. These can be DATE or DATETIME

• [Invite only Beta] Predictions

  • Model quality stats are now shown above predictions table
  • You can now download results in to a CSV file!
  • We now support datasets with more than one primary key opening up predictions for more datasets

Bugs Fixed​

• Key Driver Analysis
  • Fixed bug in row sampling feature where sampling rates were not being saved consistently before running the analyses
  • Fixed layout bug where fact table filter would get cut off by bottom of screen
  • Fixed the “Christmas tree view” alphabetical sort
  • Fixed small issues where clicking run or “save & run” on analyses that did not actually run.
• Explorations
  • Allow users to remove failed Exploratory Analyses tiles from Dashboards
  • Improved date truncations in the explore analysis to avoid double nesting truncation
• Trend & Anomaly Detection
  • Fixed issues for trend detection with AWS Athena, Redshift, and Big Query. Now these warehouses can run trends regardless of time column type
• Administrative
  • Fixed UI error when changing password / requesting invite in the login page
  • Improved Data Library display rendering
  • Properly match dbt connections to data sources
  • Fixed “Copy URL” link - in navigation bar of analyses, dashboards, etc
  • Fixed issues in the metric change grain setup
  • Fixed edit metric from Analysis list view
  • Fixes in bulk delete, can scroll list of all selected items to delete to recheck and confirm.
  • [SQL Editor] Fixed renaming a query
  • [SQL Editor] Improved side bar minimization
  • Dashboards fixed the text tile formatting bar

Breaking Changes​

• Updates to our Amazon Athena driver. If you encounter any issues, please contact the Sisu Customer Success team

New Features​

1. [Beta] Anomaly detection

   • We have launched a beta version of anomaly detection for business metrics! Using anomaly detection, customers can quickly identify unexpected increases or decreases in a metric.
   • Anomalies are determined by first calculating the range of expected values for a given metric for each day (or week). The metric value is flagged anomalous for data points where the metric value was outside of the range of expected values.
   • Anomaly Detection is included as part of our Trend Detection analyses. No special configurations needed.
   • Multiple anomalies in a row may mark the beginning of a new trend that may take a few more data points (days or weeks) to be confirmed as a true trend change vs. just an anomaly.
   • To learn more, see the following video: https://www.loom.com/embed/ba79c41d4b4b4d869cddca0b364ba6d6

   

2. [Beta] Unified Workflows for Key Driver Analyses

   • Customers using the “High Accuracy” Top-Driver model can now see statistically-selected “top driver” subgroups alongside all other subgroups, without needing to run two separate analyses to do so. i.e. a unified view of analyses!

   • Subgroups identified as statistically significant Top Drivers are highlighted with a blue indicator, where-as all other subgroups (with lower statistical significance, but some impact to the metric) are shown without this indicator.

   • This feature is presently in Beta. If you would like to give it a try, reach out to your Sisu Customer Success contact.

     

     

3. Global Search — We are introducing Global Search, a simple search bar that lets you search across projects, metrics, dashboards, and analysis from anywhere in the product. Just click on the “search” icon on the left navigation, or press “cmd/ctrl + k” to initiate, and navigate to what you are looking for in one search. To learn more, see the following video: https://www.loom.com/share/3cad750bf655404fbd24ea7e4db90d51

Enhancements​

1. Trend Detection
   • We now automatically default the time filter window to the available date range in your underlying dataset.
   • Any subgroups with a size (i.e., number of rows) greater than 80% are now included in the evaluation of trends. Previously we only looked at those below 80%. This change ensures that even large / obvious subgroups are present in your results for you to evaluate trends and anomalies.
   • For Sum/Count metrics, when clicking on “why did this trend change happen” we now set the data-range for the Key Driver (”Why”) Analyses to be of equal number of days on each side of the trend change date. This ensures that the resulting analyses has even days for each comparison period and is thus more accurate is helping you identify the drivers of the trend change. (Note this issue does not exist for Average/Rate type metrics)
2. Product Experience enhancements
   • Introducing Right Click to Open: Project list, analysis list, and dashboard lists now have a standard right click so you can right click and open the analysis or project or in a new window or a tab, without leaving the original page you are on.
   • Introducing Bulk Delete: You can now delete multiple analysis together and keep your projects clean with the new “bulk delete” feature.
   • Improved Sorting: Project lists sort in the same default order across the app so that you can view the same list and order everywhere.
   • Improved error messages for various data types in the application to give you more context on errors.
   • We have centralized the “metric deletion” to the “metric library” only, so that you do not accidentally delete a metric by mistake from other pages.
   • 90% faster load times of all pages that have a list of tables, queries, or dimensions in them from your data-warehouse!
3. Metric Units & Decimal Places
   • You can now set the units and decimal places for any Metric on the Metric setup page. The units show up on Key Driver Analyses, Trend Detection, and Dashboards, making it easier to consume and understand analyses results.
4. Key Driver Analyses APIs
   • Analyses results now include “impact” for each subgroup in the API response for Key Driver Analyses. With this update, all statistics in the “Top Drivers” table are now included in the API response
5. Updates to pysisu
   • Our python library, pysisu, has been updated. Now you can run analysis with the run() method, allowing customers to embed KDA executions directly within a data pipeline. Impact score has also been added to KDA results to complete the export of all fact table information
6. Predictions [Beta, Invite only]
   • Improved predictions table formatting based on the type of metric the predictions are for.

Bug Fixes​

1. Key Driver Analyses
   • Fixed SQL Server, Big Query and SAP Hana Date Truncations not bucketing days from Monday to Sunday.
   • Gracefully handle connection errors to customer data warehouse for Filter drop-down values
   • Fixed an issue where the “Christmas tree” chart in the Fact Detail Page was broken for percentile bin columns
2. Trend Detection — Fixed handling of timestamps in Redshift
3. SQL Editor — Fixed bug that wasn’t saving new queries in the query editor.
4. Dashboard Filters [Invite only Beta]
   • Fixed an issue where our latest workflow results were not correctly assigned to filtered KDA
   • Fixed exploration filter when there's only one default expression
   • Fixed series and result loading for non-board sliced explorations

New Features​

1. Sisu integration with Looker

   • You can now kick-off a Key Driver Analysis from a Looker Dashboard. This makes it easier to understand “why” your metrics are changing if you use Looker to understand the high-level view of your metric. Sisu automatically creates the metric and analysis based on the measure and dimensions defined in the LookML of the Explore. Video Demo below.

   • Setup Looker integration with Sisu, instructions here

   • Demo Video: Start a Key Driver analyses from Looker:

2. Live Chat for Support

   We now support chat communication inside the Sisu platform for all Support related issues or questions. Click on the help logo (pictured below) to get started and connect with us! We plan to be live 8AM – 4 PM Pacific Time, Monday – Friday with the exclusion of major holidays.

3. New help pages

   We released a new Knowledge Base to provide a cleaner UI and more features to make it easier for you to find the content you need. Check it out here: https://docs.sisudata.com/

4. [Beta] SAP Hana connector

   Sisu now supports data connection through SAP Hana.

Feature Enhancements​

1. API Updates API Key Generator in the UI: Users can now generate API keys directly from the UI. See demo below:

2. API Documentation

   Released API documentation here. This makes it easier for users to understand our APIs and build the scripts/apps they need.

3. [Beta] Consume Trend Detection results

   Just like KDA consumption, customers can now consume Trend Detection results via Sisu APIs. See documentation here. (You will need to select Trend Detection analyses on the right panel)

4. [Beta] “PySisu” SDK to consume Sisu API results

   We just released an initial SDK to show Sisu facts in a clean table view. SDKs make it easier for customers to consume Sisu APIs and build the scripts/apps needed to automate ACTIONS from Sisu insights. You can find it here: https://github.com/sisudata/pysisu

5. [Beta] Scripts for exporting Sisu Facts into Snowflake

   Leveraging the Sisu API and Airflow, we now have scripts to help customers export Sisu facts programmatically into Snowflake (other CDWs can be done on request). Find them on Github: https://github.com/sisudata/sisu-airflow

6. Trend Detection

   • Updated chart tooltip to show trend slopes before and after trends change

     

   • Added explicit error handling when no time dimension is selected

     

   • Improved the trend detection algorithm to truncate the number of facts being returned to the web app based on fact quality (determined by various factors such as slope and impact)

7. Dimension name styling

   Updated fact factor styling in Trends and KDAs so that dimension names are easier to read and percentile range details are more concise

8. Easier Navigation

   • It is now possible to filter by “analysis type” on the project’s analysis-list page

     

   • Analysis list navigation dropdown now shows details of the analysis type making it easier to quickly hop to the analyses you need

     

   • Dashboard list can now be sorted by “last run”.

     

   • When you create a new dashboard, it now lands you at the new dashboard, rather than having to navigate yourself.

9. SQL Editors

   • Removed the unsaved “untitled query” from the list of sql queries.

   • Query Preview - now we freeze row headers on top for the table preview.

     

   • SQL Queries in the side panel are now sorted by ‘Last Edited’

10. Admin Settings

    Member role can be changed to Admin from the UI.

Bug Fixes​

We fixed over 30 bugs recently! Here are a few major ones:

1. SQL Editor

   Better error reporting for SQL queries — Customers get the correct line number in which the error occurs when there is an error in the query.

2. Trend Detection

   • When creating a Key Driver Analysis from Trend Detection Analysis, filters are preserved and applied to the new analyses
   • Dynamic filters and summary periods now populate for all trend detection analysis

3. Dashboards

   Fixed dashboard tile resize handle icon (previously was not rendering)

1. [Beta] Trend Detection
   • We increased the speed of running time series stats by 45x.
   • We now support time period aggregations, by day, week, or month
   • Subgroup sizes are now shown on a per-trend basis rather than overall for Trend Detection.
   • Trend analyses now ignore leading zeros in the data
   • Sisu now gracefully handles datasets without data in the analysis period
   • Added custom time filter to replace history length
2. Key Driver Analyses
   • When there are few facts on the Fact Table, we now automatically “ungroup” the facts so you can see them all. Without this, order 2 and 3 facts would be hidden.
   • Filter dropdown values are now sorted numerically or alphabetically and not by how prevalent they are.
3. Dimension Management
   • Dimensions are now sorted alphabetically in both the key driver analysis and time comparison dimension modal, as well as the metric setup page.
   • You can now quickly filter dimensions by data type so you can easily see and find the dimensions that are strings, numbers, booleans or time-stamps.
4. Boards & Embedding
   • Added “Share” link to copy the URL of a board
   • [Beta] You can now pin the time chart from the fact detail page to a board or embed it in Tableau
5. Improving Analyses Run Speed
   • Key driver analysis → Unweighted categorical compares now runs faster for some datasets
   • Enabled native Snowflake unload to Azure
   • Fixed a bug in the ‘compare top driver’ algorithm which improves workflow times for large datasets
6. Project Page
   • Updated analysis tags to show more explicit types: “General performance”, “Trend”, “Time compare”, “Exploration”
7. Metrics Setup
   • Added “select all” and “deselect all” for dimension lists in metric setup
   • Made “duplicate metric” more discoverable in metric setup

1. [Beta] Trend Detection improvements
   • Time grain configurability: You can now examine trends aggregated by day (existing), week, or month
   • Control trend sensitivity: Some data sets have a lot of noise, and some don't. Now you're able to adjust how sensitive you want our trend detection algorithm to behave.
   • Subgroup size calculation: See the size of each subgroup during a trend and not just over the whole data set period. This should help you paint a better picture of just how impactful this subgroup was during a trend.
   • Zero filtering: Missing data for one day/week/month? Trend detection can now ignore “zeros” to more confidently and accurately determine actual trends and trend changes.
2. [Beta] Row Sampling for Key Driver Analyses: You can now run key driver analyses on a sample of your data (e.g. 10% of rows randomly sampled). This improves analyses run time by 2-10x and allows you to quickly iterate on your dimension setup before running on full data.
3. Dynamic time filters for General Performance and Group Comparisons: You can now dynamically filter on time ranges in all key driver analyses (i.e. not just time comparisons). This means you don't have to update their date selection as time progresses (e.g. "Last month"). This makes it easier to enable Auto-Run and starting Following such analyses
4. Verified metrics: You can now toggle a metric as “verified”. With verified on, you know which metrics are the source of truth and worth using (among duplicates or various versions) in order to save time and trust analysis results.
5. App Navigation:
   • In the project analyses list page, you now clearly see which type of key driver analysis was run (e.g., ”Time comparison”). This makes it easier to know what type of analyses you are navigating to.
   • You can create a metric and analyze from within projects. This interaction was available in the data and metrics tab and is now available in the projects tab as well!
6. SQL editor improvements
   • Improved table name view: You can now (A) resize the side pane; to see the ENTIRE_TABLE_NAME_NO_MATTER_HOW_LONG_IT_IS, (B) hover to see the full name (C) click to copy the table name, and (D) see the custom query name or table name of current data preview.
   • Faster typing: We fixed a bug causing a typing lag. Type as fast as you can!
   • Explore a Query: You can now directly open an Exploratory Analyses to explore the results of a SQL query without needing to create a metric first! This makes it easier to do some data exploration for a table before creating metrics for further analyses.
7. Dashboards & Sharing
   • Pin Key Driver Fact Table to Dashboards: You can now pin the Fact Table from the Key Driver Analysis page to a dashboard by clicking a “pin” button, and selecting your favorite dashboard(s).
   • [Beta] Embed Fact Tables in Tableau: You can now embed any Fact Table into Tableau and share the magic of Sisu with Tableau.