Presenters

Source

Beyond Self-Service: ING’s Journey to Developer Productivity Through “Toil Hunt” 🚀

For years, organizations have chased the promise of DevOps and Platform Engineering – the holy grail of faster software delivery and happier developers. But what happens when self-service platforms and automated pipelines still leave developers feeling frustrated and bogged down? ING, a global financial institution, discovered this firsthand. They’re sharing their journey, and it’s a powerful lesson in what really drives developer productivity. 💡

The Problem: Autonomy Can Breed Cognitive Load 🧠

ING initially embraced Agile and DevOps with gusto, granting teams significant autonomy and investing heavily in self-service platforms. While this initially boosted velocity, it quickly created a new challenge: increased cognitive load for developers. The problem wasn’t a lack of tools; it was the cross-platform journey developers had to navigate – a complex chain of tools and processes that introduced friction at every step.

As one ING representative put it: “It’s not about self-service and automated platforms; it’s about the cross-platform journey that matters most.”

Introducing “Toil Hunt”: A Data-Driven Approach to Eliminating Friction 🛠️

To tackle this, ING developed a unique initiative called “Toil Hunt.” It’s a structured, data-driven program designed to identify, categorize, and eliminate these friction points – the “toils” that bog down developers. This isn’t just about fixing production incidents (the traditional SRE definition of toil); it’s about any friction in the developer workflow, from onboarding challenges to inefficient processes.

How Does “Toil Hunt” Work? 🎯

The “Toil Hunt” program follows a clear process:

  1. Identify Tech Streams: Map out the specific technology paths applications take (e.g., container platform, VMs, databases).
  2. Developer Involvement: Engage developers – the users of the platforms – to actively report these friction points. Their insights are invaluable.
  3. Data Collection: Systematically gather data on these pain points, including:
    • Size: Estimated effort to resolve (days, sprints, months)
    • Frequency: How often the toil occurs (daily, sprint, monthly)
  4. Prioritization: Analyze the collected data to prioritize areas for improvement based on impact and effort.
  5. Action & Iteration: Implement solutions, measure their impact, and iterate on the process.

Key Learnings & Results ✨

  • 40% Productivity Gain: A pilot program in Belgium captured around 100 toils and resulted in a measurable 40% productivity gain.
  • Shifting the Definition of “Done”: Platform owners now define “done” not just by onboarding teams, but by measuring actual platform consumption. Are teams truly using the platforms?
  • Tooling is a Means, Not an End: Focus on solving the underlying problems, not just deploying the latest and greatest tools.
  • Expanding the scope of DevOps: The definition of “toil” extends far beyond production incident troubleshooting to include friction points in the entire development lifecycle.

Applying “Toil Hunt” to Your Organization 📡

You don’t need to be a global financial institution to benefit from ING’s experience. Here’s how you can implement a similar program:

  • Assess Your Current State: Honestly evaluate your DevOps maturity, focusing on cross-platform integration and developer experience.
  • Define “Toil” for Your Context: Adapt ING’s broader definition of “toil” to fit your specific challenges.
  • Implement a “Toil Hunt” Program:
    • Start Small: Pilot the program within a single team or department.
    • Establish a Reporting Mechanism: Create a simple and accessible way for developers to report friction (e.g., a dedicated Slack channel, a simple form).
    • Analyze and Prioritize: Regularly review reported issues and prioritize improvements.
  • Focus on Platform Engineering Principles: Invest in building a Platform Engineering team or function to orchestrate the developer journey.
  • Measure Impact: Track key metrics such as developer satisfaction, cycle time, and error rates to measure the effectiveness of your efforts.

ING’s “Toil Hunt” program is a powerful reminder that true developer productivity isn’t just about automation; it’s about understanding and eliminating the frictions that hold them back. By embracing a data-driven approach and actively involving developers in the improvement process, organizations can unlock significant gains in efficiency and innovation. 💾

Appendix