Some of you may be aware that the latest update of CMMI from version 1.2 to version 1.3 is in the pipeline. The current target release date is the 1st November this year, and so far it appears to be on track.
There are 4 main drivers for V1.3;
- Clarify the high maturity practices
- Simplify the generic practices
- Increase the appraisal efficiency
- Improve the commonality across the existing three constellations
These are all very appropriate aims. The high maturity material has sometimes been misconstrued and this can be very expensive to both an implementing organisation and the status of a lead appraiser!
The original changes proposed to the generic practices however were not so good. In very early drafts of the proposed changes it was suggested that the number of generic practices could be significantly reduced. This was feared by many of us as potentially seriously damaging to the very essence of what makes CMMI so unique. The power of the generic practices is in creating an environment in which processes can thrive. Each of the 10 generic practices in Generic Goal 2 provides a component of that environment and together the Generic Practices ensure that good practice is embedded in the everyday experience of working in an organisation. To remove several of these practices would have been like pulling on a loose thread; the integrity of the fabric is damaged and soon the garment is a shadow of its former finery.
Fortunately, these very radical changes no longer appear to be on the agenda. However the generic practices will still change. The most significant change is the proposal to remove the generic goals 4 and 5 and the four practices associated with them. These practices provide a route for implementing the high maturity aspects of the model in the Continuous representation. They are in some ways a mini form of the high maturity process areas (Organisational Process Performance (OPP), Quantitative Project Management (QPM), Causal Analysis and Resolution (CAR) and Organisational Innovation and Deployment (OID)). However, to capture in just 4 practices the full essence of these non-trivial process areas, is very difficult to achieve, and in practice organisations wishing to implement high maturity actually pick up the full process areas themselves. So this is probably a good change. There may be some fears that it will change the models orientation towards the continuous approach but I think this is spurious. Continuous is most frequently useful to the organisations starting their journey, and long before they even look at the high maturity content of the model. Its value as an approach is very strong and will not be damaged by this move at all. This change is still to be agreed within the SEI but it looks like safe one from here.
One of the high maturity changes currently mooted (i.e. still not agreed as well) is the creation of a third process area at ML5. This change would transfer some aspects of what is currently contained in Organisational Process Performance (OPP) to a new process area Organisational Performance Management. I am not too clear at the moment about which aspects are being transferred, and I am interested to see what impact it will have on effectively implementing ML4. So I’m afraid its watch this space on this one at the moment.
Two other changes in the model are worth talking about at this stage. The first of these is that the Integrated Product and Process Development (IPPD) material which is currently an addition or optional in the DEV model, will become part of the expected material in the same way as it already is in Services and Acquisition. This effectively removes its optionality and as many organisations have excluded it from their scope it will increase the size of the re-appraisal challenge. In case you are unware of IPPD it has in large part to do with setting up and maintaing Integrated Teams. Progressively from V1.1 to V1.2 and now to V1.3 this material has changed. Originally it existed as “two and a half process areas”; Organisational Environment for Integration (OEI), Integrated Teaming (IT) plus two specific goals in IPM. All this was optional. In V1.2 this was reduced to a single additional specific goal in OPD and another one in IPM, again optional.
But the additional constellations changed that. In first ACQ and then SVC the IPPD material was subsumed as non-optional specific practices into OPD and IPM. So far this has not caused a problem with either of these models, but in some respects it may be early days. DEV may also be a different kettle of fish. However, I suspect these fears are largely ungrounded. In many organisations that I deal with these days the move to integrate practices across multi-disciplinary teams is frequently an aspiration. Perhaps these practices will be knocking on a door that is partly ajar anyway.
The second change of note is a change to SAM that may cause a sigh of relief in some quarters. When V1.1 migrated to V1.2 the old optional process area Integrated Supplier Management (ISM) disappeared and was replaced by two practices being added to SAM. One of these practices, SP 2.2 – Monitor Selected Supplier Processes, has always had a difficult ride. It requires that an acquiring organisation makes some evaluation of its supplier’s processes. This has frequently been misconstrued as meaning that the organisation has to run some form of supplier quality audits, something that many organisations today either do not do or have no leeway to do under the contracts that they have in place. In practice it did not have to mean using such a heavyweight approach. But nontheless it regularly caused consternation when organisations tried to work out how to achieve this practice. So its removal in V1.3 will probably be met with relief in many areas. I personnally will be sad to see it go. The idea of the practice was to really think about situations in which what the supplier did might have a detrimental effect on your own success or failure. Monitoring a key aspect of the supplier’s performance may give an early warning of potential problems. This does not mean that you have to be auditing their processes in their premises, rather keeping appropriate metrics of the supplier behaviour in some aspect of quality or performance, such as hitting deadlines, may be sufficient to present a useful insight.
These are just a few insights to what is potentially on the horizon for v1.3. By and large I think it is a positive development and continues to place CMMI as a useful model responsive to its users actual needs.
