M

Tuesday, July 20th, 2021 7:35 PM

Functional: DQ Capabilities

Please use this INTERNAL discussion thread to ask any functional questions about the Data Quality (DQ) Capabilities. Below are some frequently asked questions and resources to get you started.

Top Links

Workflows

Q: Does Collibra DQ provide workflows?
A: Yes, Collibra DQ has built-in assignment queue and alerts if quality drops below a threshold.

We are working with PS to bundle / offer a DQ workflows package in marketplace in September to facilitate DQ use cases. Please note, this will not be bi-directional DQ<>DGC<>DQ to start with i.e. these are only within DGC.

Logging / Monitoring

Q: Does Collibra DQ Provide Front-End Data Entry Monitoring That Also Recommends Appropriate Data Entry To Salesforce For Example?
A: No, We Do Not Currently Support Front-End Data Monitoring. We Could Scan Against Salesforce, But That Would Be After Entry

Q: Can Services And Applications Be Monitored By Apps Such As Nagios?
A: Yes

Q: Is There A Logs Page?
A: Yes, We Expose Logs On Jobs Page. Users Typically Spend More Time On Jobs Page

Q: Can Logs Be Centralized In Dedicated Server?
A: Yes, Logs Can Be Redirected To Any Location

Alerts

Q: How Do I Get Alerts / Notified Of Issues? (A: Collibra DQ currently offers e-mail alerts based on thresholds or events; we will want to leverage and promote Collibra’s workflows shortly.

Q: Do You Provide Notifications In Case Of Updates From User In Data Model?
A: Data Model Updates Are Handled By Flyway

Dashboard / Reports

Q: View By Attribute, Value And Count?
A: Custom Reports Accessible By Metastore, Need Custom SQL For Beyond TopN

Q: How Customizable Is Dashboard? Are There Widgets?
A: You Will Have Access To Metastore For Custom Reports

Q: Do You Have Capability For Dashboarding And KPI Tracking?
A: Yes, And We Will Further Develop As Part Of Integration Process With Broader Collibra Platform

Q: Can the frequency of the Pulse view be changed to every minute (or different frequencies?)
A: Currently, Only Days / Months

Replay / Behavior

Q: Behavior Lookback, Related To # Of Samples?
A: Yes, # Days Selected Forms Baseline

Q: How Is Replay Different From Behavioral Lookback?
A: Replay Looks At Gathering Actual 10 Days; Behavioral Lookback Function Looks To Adjust Baseline Profile

Cleansing, Remediation, Standardization

Q: Does Collibra DQ perform cleansing?
A: Collibra’s principle is to not change the source data directly but provide the complete context to the data issues as well as provide workflows that enable efficient resolution of the data issues. This holistic approach means users can easily identify and focus on highest-impact issues.

Collibra Data Quality (CDQ) facilitates comprehensive remediation by providing the exception records based on insights that our competition would not capture via manual rule creation and management or incomplete data source coverage. With regards to time-series anomalies, our baseline profiles and adaptive rules continuously monitor and learn from data over time and will alert the dataset owners on data defects.

With regards to integration with remediation tools, our CDQ solution working with the Collibra Data Governance solution can notify the right data owners of errors or abnormalities and offer detail about those errors so the individual can better remediate the issue. Our customers do not just focus on just fixing the records, but remediate the root cause as part of their ETL process. Collibra’s OOTB asset types (eg. Data Quality Rule, Data Quality Metric) and APIs also allow for easy integration with any external DQ tools (eg Address Doctor). We also have pre-built integrations with common DQ tools such as IDQ, Precisely, Ataccama.

1.2K Messages

3 years ago

Can Collibra DQ REST API’s push/put bad records that were found, etc., out of our metadata store to another SQL database, or can you only use our Rest API’s to pull data out of our metadata store for use in a BI app, for example, and not “more”/push that data to another database?

33 Messages

Hi John, yes, I believe you can do a detailed export of the Breaks tab or get the break records via API to any destination. I am looping in @brian.mearns.collibra.com and @leon.kim.collibra.com to confirm.

21 Messages

3 years ago

A feature for Collibra DQ to PUSH the result (via webhook or something similar) is on feature grooming state. We are actively tracking this feature at the moment, but it is not available. No specific release date yet. For now, customers have to PULL the data from CollibraDQ via REST API or direct SQL query to DQ Metastore Storage (postgres)

2 Messages

3 years ago

Will the programmatic approach to Collibra DQ still be available in the long term?

3 years ago

Hi all,
I have a prospect who wants to know details about Collibra DQ capabilities and support. Below are the questions from the customer. Appreciate if someone can help me with these.
Thanks!

  • Does it offer data indexing, grading features?
  • Does it offer rules definitions to manage Data quality aspects : segregate data, profile data, automated notifications upon anomalies identification etc.
  • Does it offer predictive data quality aspects using state-of-art technologies viz. AI/ML etc.?
  • How does the offering help define business rules & standards (KPIs) for key data sets
  • How it integrate with the Data Catalog service (list the various Data elements (Both Technical and Business meta Data)) along with DQ
  • How does it help identify critical data elements (CDE) across domains and areas from Data quality perspective
    • Sales, Marketing, Service delivery, Service Assurance, Finance, Network Engineering/operations etc.
    • Which Key entities supported : Customer, Contract, Contact, Address and Reference Data sets
  • How does the offering help Data Cleansing best practices rollout

33 Messages

Data indexing: We have many ways to organize datasets in our Catalog e.g. business units (https://dq-docs.collibra.com/observation-assignments/business-units) and more feature enhancements to come for additional ways to organize such as business concepts / semantics. Do you / prospect have a specific use case in mind for indexing?
Data grading: Can you elaborate on what you envision here with grading? We can promote datasets to ‘certified’ in our DQ Catalog but I imagine you are after something else, so would love to learn more about the problem.

Predictive data quality: Yes we use AI/ML techniques for autogenerated rules but we try not to engage in discussions around the latest ML technologies because those conversations don’t lend to the heart of our main value proposition which is saving time in building rules, identifying issues, and trusting / using your data.

Business rules / KPIs: Please see the business units documentation for more detail https://dq-docs.collibra.com/observation-assignments/business-units. Please follow up if you have more specific questions, happy to add more color at that point!

Integrate with Collibra Data Catalog: See this thread https://community.collibra.com/articles/knowledge-base/the-page-you-are-looking-for-no-longer-exists/6638e69271654016712feda4

How does it help identify CDEs: This is a roadmap, more to share as we make progress. https://community.collibra.com/articles/knowledge-base/the-page-you-are-looking-for-no-longer-exists/6638e69271654016712feda4

Data cleansing: We have a nice blurb on this topic here: https://community.collibra.com/forum/t/functional-dq-capabilities/868

Loading...