Product Owner in 1 minute
Februari 7, 2017

Any learnings based on 13.000 teams research?

The findings in this document were extracted by looking at nonattributable data from more than 160,000 projects, 50,000 agile teams, and 13,000 active teams using the CA Agile Central application lifecycle management (ALM) platform.

The Impact of Agile. Quantified. by CA technologies

Performance analysis

Based on time in process (or time to market): The amount of time that a work item spends in process.
Based on defect density: The count of defects divided by man days.
Based on throughput/team size: The count of user stories and defects completed in a given time period.
Based on throughput variability: The standard deviation of throughput for a given team over three monthly periods divided by the average of the throughput for those same three months.

Stable teams = 60% better productivity

  • 60% better Productivity
  • 40% better Predictability.
  • 60% better Responsiveness.

Dedicated people deliver almost double

  • Productivity: 95% dedicated people teams deliver almost double compared to teams with less than 50% dedicated people
  • Quality and predictability: drops with 30%+ in teams with less than 50% dedicated people. More dedicated teams gain only slightly in quality and predictability
  • Responsiveness: hardly any difference, only a very minor correlation.

Estimating = 250% better quality

Teams doing Full Scrum have 250 percent better Quality (=less bugs reported) than teams doing no estimations.

Leightweight Scrum = better Productivity, Predictability and Responsiveness

  • Lightweight Scrum =estimate stories, no tasks
  • No estimates = not estimating teams
  • Full Scrum = teams that estimate stories (in points) and tasks (in hours)
  • Hour-oriented = teams that estimate tasks only, in hours.

Lower WIP = higher troughput and quality

lowest WiP teams have four times better Quality than teams with the highest WiP

Warning: WiP lower than 2/person = 34% Productivity loss

If your WiP is already high, then by all means drive it lower. However, if your WiP is already low, consider your economic model before you decide to drive it lower.

Teams of 7+-2 = most balanced performance.

  • Set up teams of 7, plus or minus 2 people, for the most balanced performance.
  • If you are doing well with larger teams, there’s no evidence that you need to change.

team 1–3p = less predictability & quality, more productability

  • 40 percent less Predictability.
  • 17 percent lower Quality.
  • But 17 percent more Productivity

2-week iterations= more productivity, predictability and responsiveness

  • 14 percent more Productivity.
  • 8 percent more Predictability.
  • 26 percent more Responsiveness.
  • But Quality was 5 percent lower.

no testers = best productivity & almost same quality

  • The best Productivity.
  • Almost as good Quality.
  • But much wider variation in Quality.

1 testers/3 developer = +20% quality, -12% productivity

But they had 12-percent less Productivity and 15 percent less Responsiveness.

retrospectives = 24% more responsiveness & 42% more quality

Teams that strongly agree that they have sprint retrospectives have 24 percent more Responsiveness and have 42 percent higher Quality with less variability

Comments on Any learnings based on 13.000 teams research?