100 times more frequent deployments: How did we create a high performance team?

This team went from two deployments a month to over 200. See how they ditched pull requests for trunk-based development and pair programming to do it.

#1about 3 minutes

Diagnosing a low-performing team's deployment process

The team faced infrequent deployments, multi-day code freezes, a large monolith, and many manual checks, which created waiting times and a high cost of failure.

#2about 1 minute

Using DORA metrics to define team performance goals

Four key DORA metrics—deployment frequency, lead time for change, change failure rate, and time to restore service—provide a framework for measuring software delivery performance.

#3about 3 minutes

Building a foundation of trust and shared principles

Adopting the Prime Directive, aligning on agile principles, and committing to practices like end-to-end testing and tech huddles creates a unified way of working.

#4about 4 minutes

Adopting trunk-based development and continuous delivery

Shifting to trunk-based development, supported by pair programming and extensive automated testing, enables continuous integration and delivery by eliminating PR wait times.

#5about 2 minutes

Navigating resistance and achieving 100x deployment frequency

Despite initial resistance to changing established workflows, the new practices resulted in over 200 deployments per month and reduced lead time to just 15 minutes.

#6about 1 minute

Focusing on people is key to technical success

The success of a high-performance team depends more on team commitment and continuous knowledge sharing than on individual skills or simply mandating new practices.

#7about 8 minutes

Q&A: Motivating teams for trunk-based development

The discussion covers motivating developers by highlighting the pains of PRs, using pair programming for different seniority levels, and leveraging feature toggles to manage releases.