The failed Agile Transformation.

The failed Agile Transformation.

… This is a story of what happens when companies introduce “agile” but do NOT introduce engineering practices (CI/CD & TDD)…

John, the CEO heard that Agile is THE way to deliver faster AND be responsive to changes in business needs. Everyone’s Agile these days, so he wanted to join the trend.

It must be utopia, right? John knew his business people had plenty of ideas they wanted to push to the market. John knew the huge competitive advantage to be able to release more frequently to the market. 

So John told everyone, “We want to be fast. So we’re going to become Agile”. 

And then, the Agile Transformation began. 

Scrum Teams formed.

Certified Scrum Masters.

Brand new JIRA boards.

… (But NO change in engineering practices – no CI/CD & no TDD – the team would still continue doing manual deployment and manual testing)…


The team did deployments at end of each sprint, so that the QA Engineer could do manual testing. During the first sprint, deployment + testing + bug fixing + re-deployment + re-testing took 1hr. However, a few sprints afterwards, as more and more features piled up, the total time increased – going up to 4 hours, then it took 1 day, then later 2 days, eventually 3 days… Regression bugs started piling up. As features piled up, more and more bugs. So in order to make the system ready before the sprint demo, the whole team had to do a “feature freeze” 3 days before the sprint demo.

News came up to management that the team was “unproductive” for 30% of the sprint. So then management told the team since development & QA takes up so much time, that it’s a waste to do it every sprint. Retrospectives were also dropped, since the same problems were repeated each time. The team moved to doing feature sprints, followed by QA / bug-fixing sprints, followed by UAT sprints, and finally the Production Release sprint.

Delivery cycles became longer, time-to-market slower. Endless bugs. Unhappy customers. Couldn’t keep with business demands, couldn’t deliver quality. So, ultimately, their customers left.

Epilogue: At the end of this, everyone concluded that Scrum doesn’t work in practice, that it’s snake oil. So then later they tried to improve the way they do Scrum, the way they use JIRA… but engineering practices (the root cause) – where are they?