Salesforce.com Increased Productivity by Using Scrum

 

The picture next shows the success Salesforce.com had, when the introduced Scrum in a Big Bang approach within 2 weeks after 3 month. Featurs delivered per team started to increase dramatically and the days till next release dropped.

By implementing Scrum “by the book” as the speakers said today at the Scrum Gathering in Stockholm and adding   some “agile” engineering practices Salesforce.com was able to solve the findings they had in 2006

Findings

  • Features delivered per team decreased over time. from 2001  to 2006
  • Days between major releases increased in this timeframe
From the beginning of Salesforce.com that have been the two basic performance metrics Salesforce.com used to know how good they are. And the performance dropped from 2001 to 2006 dramatically.

Classical problems of waterfall

They said in the presentation that the classic problems of waterfall, traditional methods had been the reason for this bad productivity:

  • lack of visibility
  • Resource bottlenecks – people try to restrict decision making
  • unpredictable release dates
  • lack of responsiveness, lack of team alignment on priorities
  • people blame around themselves.
  • unhappy customers
  • people were starting to leave
  • gradual productivity decline as the team grew

So they wanted to stop the decline in productivity by going in the opposite direction:

“I knew we needed radical change to get us back on track to regular releases and agile delivered.” Parker Harries – Founder and Executive Vice President, Technology 

The plan

So the plan was to do within 3 months a big bang “by the book” implementation of Scrum.

The immediate result was: Findings: The did suddenly the shortet release cycle ever

The did this all not because they wanted to implement Scrum. They had clear business driven goals. Goals:

  • Predictability  – On Time
  • Releases – more often

Since the implementation of Agile:

“Every agile release has been deployaed on time (down to the exact minute)”

Results:

+94% in feature request delivered between 2006 and 2007

+38% of feature request delivered per developer

Now:

Business wants to slow the development down, because they can not handle the output of the development any more. They deliver too much.

See the full presention here: It was incredible (for some people) – and proved everything I tell people in coaching and trainings. Yes – it sounds like a miracle.

Btw – now they started to do Agile not only in R&D but in IT Operations and in Services also.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s