How to Choose Pilot Projects for Agile and DevOps Transformations

Agile and DevOps Pilot Project Selection

The author draws a parallel between the TV show "Band's Summer" and the challenges of choosing pilot projects for Agile and DevOps transformations, emphasizing that small wins are essential for scaling success.

Four Key Factors Influencing Pilot Selection (Mike Cohn)

1. Duration – Projects should be of moderate length, avoiding the extremes of too short (questioning Agile’s relevance) or too long (delaying feedback). 2. Size – The pilot should be small enough for a single team to complete, eliminating multi‑team coordination issues. 3. Importance – Choose projects that are important to the business but not mission‑critical, ensuring motivation without excessive risk. 4. Business Owner Involvement – Highly engaged sponsors help drive the process and mitigate concerns about impact on delivery schedules.

Additional Considerations

Beyond the four factors, the author suggests: the team should be co‑located; prioritize new product development over maintenance projects; and favor pure software projects before tackling hardware‑integrated initiatives.

Four Characteristics of Micro‑services

1. Small – Fine‑grained services focused on a single responsibility, owned by a “two‑pizza” team. 2. Own Process – Each service runs in its own process. 3. Lightweight – Uses lightweight communication mechanisms. 4. Independent Deployability – Services can be compiled, deployed, and run independently without binary or ordering dependencies.

Further Factors for Pilot Selection

Consider team proximity (avoid distributed teams unless common), select new product development over maintenance, and prioritize software‑only projects to ensure visible, rapid results.

Discussion of New Bands in "Band's Summer 2"

The article briefly mentions emerging bands such as Mandarin, 福禄寿, and 傻白, highlighting their fresh energy and the notion that innovation thrives when there is “nothing to lose.”

Conclusion

Successful pilots require typical team size, reasonable project duration, and clear business value that can serve as a model for broader adoption; incremental scaling from 0→1 to 1→3 and beyond should be pursued with disciplined, repeatable practices.