Quick Wins

These five improvements can each be completed in a single sprint and will immediately improve your team's DORA metrics and developer experience. Start with the one that addresses your biggest pain point today.

Quick Win 1: Measure Deployment Frequency

Time: 2–4 hours · Impact: Baseline DORA data, shows where you are today

You cannot improve what you cannot measure. DevLake ingests deployment events from ArgoCD and calculates deployment frequency automatically.

# Check DevLake is deployed
kubectl get pods -n devlake

# Open the DORA dashboard in Grafana
make argocd-status ENVIRONMENT=dev

If deployments are not being recorded, follow the DevLake setup guide.

Quick Win 2: Add a Basic CI Pipeline

Time: 1 day · Impact: Catch bugs before production, reduce CFR

If your service does not yet have automated tests running on every PR, add a GitHub Actions workflow or a Jenkinsfile that runs lint and unit tests.

# .github/workflows/ci.yaml (minimal example)
on: [push, pull_request]
jobs:
  test:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - run: pip install -r requirements.txt
      - run: pytest tests/unit --tb=short

See GitHub Actions Workflows for the full Fawkes CI template.

Quick Win 3: Containerise One Service

Time: 2–4 hours · Impact: Enables consistent deployments, reduces "works on my machine"

Pick your simplest service and add a Dockerfile. Use a multi-stage build to keep the image small:

FROM python:3.12-slim AS builder
WORKDIR /app
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

FROM python:3.12-slim
WORKDIR /app
COPY --from=builder /usr/local/lib/python3.12/site-packages /usr/local/lib/python3.12/site-packages
COPY . .
USER nobody
CMD ["python", "-m", "uvicorn", "app.main:app", "--host", "0.0.0.0"]

Quick Win 4: Add a Health Check and Basic Alert

Time: 2–3 hours · Impact: Detect failures before users report them

Add a /health endpoint to your service and a Prometheus alert that fires when it stops returning 200:

# PrometheusRule
- alert: ServiceDown
  expr: up{job="my-service"} == 0
  for: 1m
  annotations:
    summary: "{{ $labels.job }} is not responding"

Then connect the alert to your team's Mattermost channel via Alertmanager.

Quick Win 5: First GitOps Deployment

Time: 1 day · Impact: Eliminates manual deployment steps, makes deployments auditable

Move your service from kubectl apply to ArgoCD-managed GitOps in one sprint:

1. Package your deployment as a Helm chart (use the Backstage template as a starting point)
2. Add an ArgoCD Application manifest to platform/apps/
3. Commit → ArgoCD syncs → deployment is automated forever after

Follow Onboard a Service to ArgoCD.

Next Steps

After completing these quick wins, revisit your capability assessment and choose your implementation path for the next quarter.

See Also

• Tutorials
• Getting Started