Delivery Management

Software Delivery Becomes Measurable and Manageable

Delivery Management is the discipline of optimising the flow of software changes from idea to customer. Instead of gut feeling, we rely on hard data (the DORA metrics) to make the performance of the IT organisation objectively measurable and steerable.

It is not about "typing faster" but about reducing friction in the processes and systematically increasing the quality of delivery.

Anti-Patterns: Flying Blind

• Management by Story Points: Story Points are estimating tools for teams, not a comparison metric for management reporting.
• Fear of Releases: Since releases are manual and error-prone, they are carried out infrequently, which massively increases the risk per release.
• Unclear Priorities: Teams work on too many things at once, leading to high WIP (Work in Progress) and slow lead times.

The DORA Metrics

Research by DORA (DevOps Research and Assessment) currently describes five metrics for software-delivery performance. DORA explicitly warns against setting broad metric goals without context; the bands below are benchmark orientation, not universal targets:

1. Deployment Frequency: How often do we successfully deliver code into production? (Benchmark top tier: daily to several times daily).
2. Lead Time for Changes: How long does it take from the first commit to productive operation? (Benchmark top tier: less than one day).
3. Change Failure Rate: What percentage of releases cause problems that must be corrected immediately? (Benchmark top tier: low single digits up to 15%).
4. Failed Deployment Recovery Time: How long does it take to restore service after a failed deployment? (Benchmark top tier: under one hour).
5. Rework Rate: What share of work goes into unplanned rework? A high value points to quality or process problems.

Operational Levers

• Continuous Delivery: Automation of the entire release chain.
• Small Batch Sizes: Small changes reduce risk and accelerate feedback.
• WIP Limits: Do fewer things at once to complete them faster.

FAQ

Are these metrics not too technical for board reporting?

No, they are the direct indicator of the company's adaptability. A high Deployment Frequency means the company can react quickly to market changes.

Doesn't speed inevitably lead to lower quality?

The data shows the opposite: High-Performer are both faster and more stable. Automation enables both at the same time.

References

• Puppet / DORA State of DevOps Report. The annual study on the performance of IT organisations. (2023). www.puppet.com/resources/history-of-devops-reports
• DORA / Google Cloud DORA Metrics. The official documentation of the metrics. (2019). dora.dev
• IT Revolution Accelerate. The book that summarises the scientific research behind DORA. (2018). itrevolution.com/product/accelerate/