What is OwlDQ
Crowdsourcing
“People that have never written SQL are now helping with data quality”
Rule Coverage
“Did in 20 days what took 2 years with our legacy tool”
Audit & Identify Gaps
“Audited our existing checks and could not imagine the gaps we uncovered.”
Automate Repeatable Processes
“Owl cut 60% of our manual workloads”
Technology Limitations
“We now scan files and Kafka, avoiding downstream issues”
Getting standard
“No more piecemeal reports. Files, Warehouse, Lake. All metrics in one, transparent place.”
Building Reports, Visuals, Workflows
“This takes the place of 3 tools”
Prioritized Efforts
“Easy to see top priorities for improvement”
Another expensive project missing deadlines?
Tired of wasting an afternoon unwinding ETL / ingestion jobs?
Know the dread of another fire drill?
Is it crazy to think your time can be better spent than wading through data issues?
Top 10 Bank
Reduced 60% of their manual Data quality workload + $1.7M cost savings
Top 3 Healthcare Organization
Saved 2000 hours during a cloud migration requirement
Top Insurance Organization
Satisfied Regulatory Second Line Controls in a 4-weeks (what took 2 years using their previous tool)
Overwhelmed with tickets
Business users find issues first
Touchy pipelines break with minor updates
Too busy responding to fire drills to implement new projects
Implementing Checks
Autodiscovery
Generates SQL validations, parameters & thresholds
Rule suggestions
Taking Inventory
Bulk Profiling & Metadata Collection
Data Mapping with Column Identification
Map Column Fingerprints, Cross-Table Matches & PII Checks
Consolidating Systems
No more closed-systems or confusing scripts
Macro & micro views for measuring effectiveness over time
Global management Across Sources / Platforms / Environments
Enabling More Users
Self-Service, Easy to use Rule Editor
Pre-Built Analytics and Charts
Extensible APIs, Open Architecture
Every feature, visual, and component within Owl is intended to make the analysis and implementation of data checks easier.
"An unexpected ETL update occurred during a migration that changed our up-to-date-payments indicator from TRUE/FALSE to 1/0. Needless to say, we were very surprised when invoices were not sent. The rework and reconciliation were super painful. An enormous amount of my time was wasted." |
"One of our 200+ reference data feeds introduced a pipe (|) into a position field. The field was defined as VARCHAR so this was allowed. Our models went crazy and we thought we breached risk limits. We ended up selling out of positions (losing millions). Only to uncover the root cause much later that week." |
"We pull data from many APIs. One platform accounts for 10% of enrichment activities (i.e. how we monetize our data). Our auth token accidentally had a daily quota imposed, yet job control said green light (successful connection). We still loaded some rows (1k), just not entire payloads. This was super nuanced. We literally lost ~10% revenue that month." |
"When we introduced new meters, they were hooked up and sending valid readings. They were valid values within valid ranges. Turned out their default setting was rounding the actual values and we were losing precision. Devastating, considering the amount of precision required with blood values." |
OwlDQ offers a low-hanging fruit opportunity to extend your data quality toolset.
The fact that data quality is a consistent pain point suggests it's important to many business-critical functions and legacy products not getting the job done.
