 |
|
  
Mike Harris, DCG
President to present "Understanding and Implementing ITIL" in
London on 2/23/07 at the Fairmont-Savoy Hotel at the ITMPI
SOFTWARE BEST PRACTICES CONFERENCE 2007 hosted by The IT
Metrics and Productivity Institute (www.itmpi.org).
Registration is open for the
free DCG Webinar “ITIL Performance Measures" on 3/13/07, vist
the DCG website to reserve your seat today!
DCG becomes
charter member of SEI Performance Benchmarking Consortium
(PBC).
DCG CMMI Process Library available on-line at
the DCG website.
DCG Industry Data available on-line at
the DCG Website.
DCG particpating in GEIA Industry
Forecast Workgroup for Federal Government Services.
DCG
Podcasts coming in next month's February Newsletter, stay
tuned...
|
|
January 2007 Volume 1
A recent article in the IEEE Software Nov./Dec.
2006 issue, "Are CMM Program Investments Beneficial?"
(Copyright® IEEE Software), caught our eye. Here are some of
our thoughts for our friends in the Software Process
Improvement community.
This article provides the hard data many of you
have been asking for regarding the Return On Investment or ROI
of CMMI. This research sought to consolidate and normalize the
reported results from CMMI program investments. The authors
Daniel Galin and Motti Avrahami, chose 19 papers which
reported on results for more than 400 projects from many
different companies worldwide. I find that it brings
compelling statistics of CMMI model based improvement current
in a no-nonsense fact-based presentation of data.
Some of the improvement measures covered
are:
 Defect density (reduction in
defects)
Fewer regression test cycles &
correction cycles (decreased testing but specific to lowering
the defect rate and test cycle reduction overall)
Fewer software errors post release,
and
More project milestones completed on
time.
This research effort focuses specifically on
CMM(I) “level transformations” or CMMLT meaning initial
company performance pre-CMM(I) and post-CMM(I) level
improvement. From the 19 papers the authors created 99 CMMLT
records, 85% of which are single level improvements. The
implementation times range from two to ten years. These
records cover common performance metrics such as Error
density, productivity, rework, ROI and so forth. See the
actual graph on page 85 (page 5 of the article).
In order to verify the results the authors
performed statistical analysis of the CMMLT records. They
carried out a t-test for each performance metric. Each metric
rejected the null hypothesis: The mean performance improvement
reported for a single CMMLT is zero. So for the rest of us
that means the improvement results are more than just
persuasive, they are compelling and backed by data.
The results are both reaffirming and exciting.
If you are new to the efforts of CMMI based improvement or an
experienced leader in process improvement you will benefit
from reading this research. I recommend putting this powerful
information to use in your organization.
If you would like a copy of the article it is
posted for download at our website at www.davidconsultinggroup.com.
Please browse the updated www.davidconsultinggroup.com website...
-
Visit the new on-line Resource Center - Learn about the new
on-line CMMI Process Library and Industry Data service;
available as corporate or individual subscriptions
- Visit
the updated Publications area for instant downloads of free
information
- Check out the Capabilities area and all areas
of the updated DCG website
|
|
|
