Coalesce 2024: Don't panic: What to do when your data breaks
This talk isn't about data quality tests. Why? Well, because there’s no shortage of tools and processes for testing your data, monitoring your data, and alerting your team when the pipeline breaks.
This talk isn't about data quality tests. Why? Well, because there’s no shortage of tools and processes for testing your data, monitoring your data, and alerting your team when the pipeline breaks. Instead, this is a talk about what happens after the alarm goes *off*.
Data incidents can quickly snowball into much more than simply fixing the issue. Juggling comms, diagnosing the issue, ownership of the failure, your DMs lighting up…we’ve all been there and understand just how stressful it can be.
In this session, Matilda covers a number of tool and product agnostic approaches teams can adopt to improve how they manage data incidents. These are practical steps any data team can follow to improve how they’re resolving, communicating, and learning from their incidents, with the goal of providing more resilient data systems.
Speaker: Matilda Hultgren, Data Analyst - Product, @ incident.io