Service Delivery Index: A Driver for Reliability

Customer-first: Moving from Hero Engineering to Reliability Engineering From the beginning, Slack has always had a strong focus on the customer experience, and customer love is one of our core values. Slack has grown from a small team to thousands of employees over the years and this customer love has always included a focus on…

Slack logo
Slack
•
Matthew McKeen
•11 min read•advanced•
--
•View Original
Chef

Overview

The article discusses the Service Delivery Index (SDI-R) as a crucial metric for driving a culture of reliability within Slack's engineering organization. It emphasizes the importance of measuring service reliability to enhance customer experience and outlines the processes and tools implemented to achieve this goal.

What You'll Learn

1

How to implement the Service Delivery Index to measure service reliability

2

Why proactive reliability management is essential for customer satisfaction

3

When to prioritize reliability over new feature development

Key Questions Answered

What is the Service Delivery Index and how is it calculated?
The Service Delivery Index – Reliability (SDI-R) is a composite metric that measures the success of user jobs and Slack's uptime. It is calculated using successful API calls and content delivery, along with critical user workflows such as sending messages and loading channels.
How does Slack ensure service reliability?
Slack ensures service reliability through a combination of Incident Management processes, Service Ownership, and the Service Delivery Index. These practices help identify and mitigate issues proactively before they impact customers.
What are the key components of the Service Delivery Index?
The key components of the Service Delivery Index include API Availability, Overall Availability, and monitoring of critical user workflows. These metrics are essential for tracking service performance and ensuring customer satisfaction.
What challenges did Slack face while implementing the Service Delivery Index?
Slack faced challenges such as ensuring that not all API requests are treated equally and the delayed nature of SDI-R reporting. They addressed these by breaking down SDI-R for larger organizations and implementing service-specific alerting.

Key Statistics & Figures

Service Delivery Index – Reliability (SDI-R)
5.15
This is the current SDI-R score reported in the article.
Uptime SLA
99.99%
Slack maintains this availability SLA in customer agreements.

Key Actionable Insights

1
Implementing the Service Delivery Index can help unify your team's understanding of reliability metrics.
By establishing a common metric like SDI-R, all teams can align their efforts towards improving service reliability, which ultimately enhances customer satisfaction.
2
Proactively managing reliability issues can prevent outages and improve customer experience.
Using SDI-R as an early warning system allows teams to address potential problems before they escalate, ensuring a smoother user experience.
3
Regularly review and adjust your reliability metrics to adapt to changing customer needs.
As Slack has learned, continuous iteration on metrics is vital to maintain relevance and effectiveness in measuring service reliability.

Common Pitfalls

1
Failing to recognize that not all API requests have the same impact on service reliability.
This can lead to overlooking critical issues affecting major customers. It's important to prioritize and weight API requests based on their significance to ensure comprehensive monitoring.
2
Delays in reporting service reliability metrics can create a disconnect between actual performance and perceived reliability.
To avoid this, organizations should implement real-time monitoring and alerting systems that provide immediate feedback on service performance.

Related Concepts

Incident Management
Service Ownership
Reliability Engineering