• Prioritization framework: RICE
    • Reach
    • Impact
    • Confidence
    • Effort
  • Goal is to split their time between new initiatives and KTLO stuff (keep the lights on)

KATHRYN KOEHLER, DIRECTOR OF PRODUCTIVITY ENGINEERING

Over the past year, we’ve adopted formal product management for our central productivity org. Previously, planning was somewhat ad hoc, based on a bottom-up understanding of our customers’ needs gathered by engineering managers and individual contributors through conversations and support interactions.

With our product managers’ guidance, we now have a comprehensive strategy grounded in top customer needs, industry insights, and our overall mission of evolving our infrastructure, platforms, and tooling to support Netflix’s streaming and studio development teams. We prioritize customer issues based on the RICE framework (reach, impact, confidence, and effort), plan on semester boundaries (Q2/Q3 and Q4/Q1), craft objectives that align with our overall org strategy, and incorporate the work that “keeps the lights on” (KTLO). We budget 50 percent of our teams’ capacity for new initiatives and 50 percent for KTLO and other team-based initiatives. We also work with developers to understand the key results that will indicate success or failure—the impact our work has on customers, not just “we shipped the thing.”

We’re new to objectives and key results (OKRs), but we’ve found them useful for aligning our conversations around goals that have measurable customer impact. We also use Nicole Forsgren’s SPACE framework of developer productivity—which measures satisfaction and well-being, performance, activity, communication and collaboration, and efficiency and flow—to evaluate how our platforms and tooling empower Netflix engineers to do their best work.

We’re in the early days of this new approach to planning, but we’re already seeing benefits: Our work more closely aligns with customers’ needs, we’re prioritizing more effectively based on reach and impact, engineering teams are more focused on fewer things, and we’re better able to coordinate timelines and manage dependencies with partner teams.