Release Notes - 2021.11.10

Welcome to the latest One Model product release update. This article provides an overview of the product innovations and improvements to be delivered on 10 November 2021. You can see the full 2021 release schedule here:

User Experience

  • In this release we have added a few tweaks to how gridlines are displayed in the new Storyboard Table we launched in our last product release. Some customers were seeing gridlines in unexpected places, but this would vary between users depending on their screen resolution and browser zoom level etc. So, we have added some improvements to this and as a by-product we now show small divider lines between highlighted cells within the table, but we think this actually looks better and hope you do too! (ref 9586)

  • We added a small enhancement to the Storyboard Library that helps limit how many rows are shown when you have Storyboards shared with lots of roles. In those cases there is a “Show All” button to display all of the values for a given Storyboard. (9966)


OneAI Innovations

Local Area Unemployment Data Augmentation

  • An important part of the value of OneAI is the ability to augment customer data to create more powerful predictive models. Data augmentation can be in the form of logic applied to existing data and generating attributes that don’t naturally appear in the source system, e.g. number of managers an employee has had over the last 3 years. In addition to this, we are able to augment data based on external data and for a couple of years have had the ability to include commute time in predictive models. In this release we are adding Local Area Unemployment data.

  • The Local Area Unemployment data augmentation is initially based on data sourced from the US Bureau of Labor Statistics (BLS) and over time and based on customer demand we will look to extend to other countries.

  • You can configure the Local Area Unemployment augmentation just like you would configure the Commute Time augmentation.

  • NB: To get these options to show in your instance you will need to contact the One Model team. This is because we require written authorization as this feature may share some data with a third party organization. In this case we are simply sending lists of Zip codes, or MSAs and retrieving the relevant data from BLS, so no personally identifying data is included, but we are applying an abundance of caution regarding the use of these services. (ref 9696)

Improvements to Data Pipeline Processing

  • Our Workday API connector has the ability to extract all historical data, or extract data on an incremental basis. As of this release it is now possible to extract data from Workday for a specific time period. This is particularly useful for customers with a lot of data in their Workday instance, allowing them to build data models etc. in One Model or update a specific period of data without the need to extract all data at once. (ref 7207)

  • We've improved the stability of processing XML files in our File Data Source by keeping the data from previous runs until data is available to load for a new run. Historically, the old data was deleted before any processing was run on the new data - which could cause Processing Scripts to error if they were started during that time. (ref 9091)

  • We improved our file matching algorithm after finding a case where a single file could match two different file inputs in a file Data Source. (ref 9374)

  • Emojis are omnipresent and now you can also include emojis in your One Model data sources and use these in Explore and Storyboards. One of our customers was using classic emoji symbols as part of an employee survey, so we enhanced our platform and they are now able to build data visualizations and Storyboards using these symbols to mirror how employees are providing their survey feedback. 😎 (ref 10026)

  • We've updated the default Maintenance Window configuration for Workday data sources to 11pm Friday, Pacific Time. This is to match Workday's new weekly maintenance period. (ref 9971)

  • Added a feature where Files in a single Data Source can be marked as Required or Not Required. This will allow us to group files in data sources and therefore process them together, even if only a subset of those files are always sent from a specific source system. (ref 9576)

Performance Improvements & Bugs Squashed

  • Fixed an intermittent error that could cause a data file to be lost from an API Run, causing an incomplete data load. (ref 9770)

  • Some customers had recently started seeing errors when running Processing Scripts where the "finalization would take more than 5 minutes". This was caused by Redshift Locks not being cleared correctly causing delays during this finalization process, and has since been fixed. (ref 9484)

  • Fixed an issue with our new Snowflake Data Destination where loading tables with similar names (such as employee, and employee_event) could cause a mismatching columns error. (ref 9647)

  • We've worked around some timeout exceptions coming from Oracle by adding retry logic to the connector when these timeouts occur. (ref 9164)

  • We fixed an issue where certain future dated transactions in Workday may cause our Workday connector to error. We will now ignore this data when initially queried, and will re-query it when it's not future dated, and therefore can be queried successfully from Workday (ref 10033)

  • We have fixed some issues impacting the automated role assignment for new users uploaded via the user file upload process. (ref 10013)

  • We fixed a number of issues with how Storyboard drill-through tables were being generated from charts or tables. There was no security issue as data was always limited to the user’s security permissions, but we weren’t always correctly applying all of the filters applied in the tile. This has been corrected. (9998 & 10034)

  • We fixed a small UI bug with how the settings menu was displayed in the Storyboard Library. The bottom border of the menu box was missing, but has now been found and put back in its proper place. (9868)



Was this article helpful?

0 out of 0 found this helpful



Please sign in to leave a comment.