Technology Risk Management in Software Development

1
Technology Risk Management inSoftware Development

• Risk management practiced by computer software
  systems development organizations in Chicago
• Preliminary survey results

Research by L. Charles Andersen, Sr.
2
The Motivation

• Computer technology, including both software and
  hardware, is changing rapidly
• Businesses must move to new technologies when
  older technologies become obsolete
• Employing new technology in any project
  interjects certain inherent risks
• Managing these risks helps lead to success

3
Survey Universe

• This survey studies practices currently used in
  industry
• The companies surveyed are listed in Crain's
  Chicago Business as operating in the Financial
  Sector
• These companies are headquartered in Greater
  Chicagoland

4
Survey Scope

• The survey limited responses to projects using
  some New Technology
• New Technology any language, platform,
  methodology or paradigm that is new to any one or
  more of
• the industry
• the company
• the project team

5
Respondents Characteristics

• The companies sizes ranged from under 500 to
  over 1000 employees
• The project team sizes ranged from under 5 to
  over 90 members
• The project sizes ranged from 12 man-weeks to 120
  man-years

6
Preliminary Results - Project Tracking

• One third of the respondents did not perform
  estimating or time tracking as part of the
  project effort
• Those who estimated and tracked results, averaged
  completion at 123 of the original time estimate
• Three tenths of the projects came in under the
  estimate, using an average of 97.1 of the
  estimated time
• Seven tenths of the projects came in over the
  estimate, using an average of 133 of the
  estimated time

7
Preliminary Results Team Makeup

• 87 used senior managers, those with five years
  or more experience
• 20 used intermediate managers, those with one to
  five years experience
• 80 used senior development teams
• 20 used intermediate development teams

8
Preliminary Results Project Success

• 100 claim moderate to total success
• 64 were delivered within the projected timeframe
• 55 required working overtime to deliver on time
• 72 were delivered within the projected budget
• 27 were delivered within budget without
  requiring overtime

9
Preliminary Results Project Success

• Less than one-tenth delivered on-time,
  on-budget, full-featured, without overtime
• Those were not considered bug-free, but were
  reported as having a moderate number of bugs

10
Interim Conclusions

• New Technology success is significantly lower
  than published success rates for software
  projects in general
• Results reports tend to subjectivity rather than
  objective reality

11
Interim Conclusions(continued)

• Management experience alone does not solve the
  problem
• Team experience alone does not solve the problem
• Overtime is seen as an acceptable solution to
  slipping schedules