What do the "Characterizations" in a CMMI Appraisal Mean?

 Dear Appraiser,

I keep hearing about "Characterizations" in the CMMI, like FM, LM, PM, and DM.  What do these means?  ~ Mick Filipini

Want to pass your CMMI Appraisal?  Here's what not to do!

Hey Mick - good question!

When we conduct Benchmark appraisals we need a way to "score" each practice based on the two types of objective evidence - artifacts and affirmations.  Each appraisal looks at both in order to determine each practice's strength.  This is done by examining "artifacts" (documents, databases, videos, etc), and "affirmations," within the context of the organization's goals, and the Intent and Value statement of the Practice Area.

Fully Met (FM) = There is plenty of documentary and spoken/written evidence to convince the Appraisal Team that that practice is healthy and complete, given the context of the organization being appraised.  The team has no indication that there are weaknesses, and all believe the practice is being executed well.  This is like getting an "A" on a paper.

Largely Met (LM) = There is also plenty of evidence in both categories, but the team feels there is some improvement to be made, and they will tell you exactly what that is.  Examples include: Not ALL projects were as strong as others, one piece of evidence wasn't as strong as the others, or maybe the method chosen to perform a piece of work might have been better by applying a different method.  I see a lot of appraisals that use "planning poker" in a setting where that may not make sense.  This is like getting a "B" on a college paper.  You can have some of these and still be rated ML2/3/4/5, but there are caveats.

Partially Met (PM) = There is some evidence the practice is being performed effectively, but there are major holes, and they need to make improvements.  For instance, maybe they are performed estimating using "Wide Band Delphi" but they didn't execute it the way it is intended.  This is like getting a D on a college paper.  You generally can't have any of these, although there are scenarios where you could pass if a project has one of these babies.

Doesn't Meet (DM) = There is little evidence that the practice is being performance.  This is like getting an F.  The presence of a single DM usually means a fail, but like PM, there are exceptions.

In a perfect appraisal, these are the only four characterizations that we would see, but there are two more!

Not Yet (NY) = The project hasn't reached the point in the project where they should be performing the practice.  This is usually either a sampling error (why was that project selected for that Practice Area) OR a misunderstanding of the intention of the practice.  We see this with Validation and Verification, where an organization claims they haven't reached testing, not realizing that these practices should also apply to earlier actions on the project, like planning or requirements.  The final outcome of an NY depends on what other projects in the sample have scored, and the Appraisal team's view as to what affect that has on the overall organization.

Not Rated (NR) = When the appraisal team can't reach a consensus on the characterization, thereby forcing the end of the appraisal without rating the organization at a maturity level.  This is usually a failure of the appraisal team, or dare I say it, a failure of the Lead Appraiser to facilitate the team's consensus.  Once in a while, we have an Appraisal team member that refuses to compromise also - but that's rare.

So that's the system - here's to you all getting FMs and LMs!

For further guidance, including training resources and instructional materials, please refer to www.broadswordsolutions.com or explore instructional content on CMMI, Agile methods, CMMI Appraisals, CMMI Training, and more via CMMI-TV.