January 2022 Release Notes

Aunalytics is excited to announce the first release of 2022 to our clients. This release will provide clients with model and site enhancement information along with any fixes to existing functionality we have included.

Daybreak

Income Series Smart Feature Improvements

This month, we've developed an improved model for generating the Income Series Smart Feature™ for our Daybreak™ for Financial Services data model. This Income Series Smart Feature was designed to analyze a customer's deposit transactions over a period of time in order to identify which deposits most likely represent a recurring source of revenue such as wages or business income.

For our clients, the new model delivers enhanced accuracy at identifying sources of recurring income. First, the algorithm that groups individual transactions into an income series has been improved to better identify members of a series when the description field of transactions in that series may be more irregular. This improvement in the classification of transactions will enable the model to capture recurring deposits that may have less consistency in their descriptions. Second, the criterion for determining acceptable variance in the amounts and regularity of a series have been adjusted to better the recognition of atypical income sources such as gig work, seasonal employment, or drops in revenue due to downturns in business or temporary shutdowns.

For our internal business, this improved model delivers faster performance and decreased costs to build this feature for client datamarts. In one benchmark, the time to build decreased from over nineteen minutes to about a minute and a half, an over 1,300% decrease in build time! This impressive efficiency gain was achieved by moving most of the computational workload into Apache Drill via SQL operations rather than downloading the raw dataset and computing it via the model's own code. Performing the calculations this way also decreased the memory requirements for the model by almost 75% (36GB to 8.5GB).

Together, these improvements yield enhanced value to our clients all while improving our own profitability and efficiency to delivering greater value at a lower cost to our business.

Aunsight™ Golden Record

Small Improvements

This month, the Aunsight™ Golden Record platform team has released a number of feature enhancements and fixes including:

  • Transactional Workflows (TXWF) API Improvements:
    • The job control API now has the ability to trigger other downstream workflows when a job has completed successfully.
    • Jobs can now be configured to Performing periodic purging of the workflow tables.
  • User Experience:
    • AuGR will now prevent automatic logout due to inactivity when a user is running a test query.
    • Users can now preview data from a Golden Record input or TXWF table in a full-screen interface, and they can also specify the number of rows they would like to view in the preview window.
  • AuGR Agent
    • AuGR now supports targeting an individual agent for deployment of an over-the-air hotfix in order to better support clients using the on-premises agent.