The Process of Improving Engineering Best Practices
In every high-functioning engineering organization, success isn’t just a product of talented developers or cutting-edge tech stacks—it’s rooted in strong engineering practices. These best practices shape how teams write, review, test, deploy, and maintain software. But engineering best practices aren’t static. What worked last year—or even last month—may not serve your team today.
I’ve worked with companies at various stages of growth. Whether you're scaling fast or trying to untangle tech debt, evolving your best practices can unlock better delivery, quality, and team health.
Here’s a detailed step-by-step approach I use when helping organizations level up their engineering practices:
Step 1: Assess Current Practices (with Eyes Wide Open)
Objective: Understand what’s really happening—not just what’s documented.
-
Shadow engineering ceremonies: Observe standups, retros, planning meetings, and code reviews.
-
Interview developers, leads, and PMs: Ask what’s working, what’s not, and what they wish they had.
-
Review current artifacts: CI/CD pipelines, code guidelines, test coverage reports, incident postmortems, etc.
-
Tools to use: Surveys (e.g., Engineering Pulse), heatmaps of DORA metrics, and dev analytics platforms like Code Climate or Velocity.
Tip: Look for pockets of excellence—some teams might be doing things others can learn from.
Step 2: Identify Friction and Failure Points
Objective: Map where time, quality, and trust are being lost.
Common friction points include:
-
Delays in code reviews
-
Long test cycles or flaky automation
-
Lack of clear Definition of Done
-
Inconsistent use of feature flags or branches
-
Misalignment between engineering and product
This step often reveals dysfunctions hidden under surface-level delivery success. Even high-output teams may be masking burnout or rework.
Step 3: Benchmark Against High-Performing Teams
Objective: Set a baseline for what “great” can look like.
Use benchmarks from:
-
DORA / SPACE metrics
-
Industry standards (e.g., trunk-based development, continuous deployment, test pyramid)
-
Internal high-performing teams
Example: A team struggling with long release cycles might benefit from studying how top teams use feature flags and release trains to ship continuously.
Step 4: Co-Create a Modernized Best Practices Playbook
Objective: Align the team on an improved, realistic standard.
This playbook should:
-
Be co-authored with input from ICs, leads, and QA
-
Prioritize clarity over perfection
-
Include why, not just what and how
-
Highlight practices like:
-
Pull request hygiene
-
Automated testing tiers
-
Incident response protocols
-
CI/CD best practices
-
AI-assisted tooling and workflows
-
Make it living—not a PDF that’s forgotten in Confluence.
Step 5: Pilot, Measure, Iterate
Objective: Start small, get feedback, and expand.
-
Run 2–3 team pilots with coaching and support.
-
Track impact using agreed-upon metrics (velocity, bug rates, PR cycle time, etc.)
-
Collect qualitative feedback from the team.
-
Iterate before rolling out org-wide.
Tip: Use internal champions to evangelize success stories.
Step 6: Scale Through Enablement and Automation
Objective: Bake best practices into the developer experience.
-
Codify standards in tooling: linters, CI checks, test runners
-
Offer templates, PR checklists, onboarding guides
-
Pair with AI assistants (like GitHub Copilot, CodeWhisperer, or internal LLMs)
-
Set up a Best Practices Guild or Engineering Excellence Council
Enablement beats enforcement. Let developers want to do the right thing because it's the easy thing.
Step 7: Create a Feedback Loop for Continuous Improvement
Objective: Make best practices adaptive and iterative.
-
Hold quarterly retros on engineering practices
-
Use lightweight engineering satisfaction surveys
-
Encourage grassroots innovation and experimentation
Your best practices should evolve as your product, people, and platform grow.
Final Thoughts: Best Practices Are a Team Sport
As consultants or engineering leaders, our job isn’t to impose a rulebook—it’s to guide, enable, and inspire teams toward sustainable excellence. The best engineering cultures aren’t rigid—they’re adaptive, collaborative, and rooted in continuous learning.
By following a thoughtful, data-informed, and human-centric process, you can elevate engineering best practices—not just for productivity, but for long-term team health and innovation.
