ALM - market hype or reality?

David Piper, Lamri

Photo of David Piper David Piper, managing consultant of Lamri, determines whether ALM really can reduce the complexities involved in delivering application projects successfully.

The IT industry is preparing itself for yet another three-letter acronym (TLA) to explode onto the scene. The new TLA is ALM - application lifecycle management. More evolution than revolution, ALM brings together tools and principles that have been of increasing importance to the IT industry over the last several years.

The scope of ALM varies depending on which vendor's offering is reviewed. However a recent Butler Group report[1] gives an all-encompassing definition of ALM based on current offerings and likely future developments.

The key aspects include: portfolio management; project management; requirements, change and configuration management; analysis, design, code, build, test and deployment; software performance and software asset management.

Integration of information across traditional project-based activities and roles is the key with which ALM tries to unlock the complexities of delivering application projects successfully. Project office, project management, change management, analysis, development, test, deployment and operations are all potentially impacted by the scope of ALM.

Despite the importance of integration, the tool vendors seem strangely reticent to challenge existing market leaders, so that when it comes to project management, for example, most simply defer to the market leader by offering loose integration with Microsoft Project. Issues with this type of integration, including the synchronisation of changing information on both sides of the bridge, are well known.

Some areas of integration are undoubtedly new and important. For example the integration of software performance management can help to close the loop between the development and maintenance cycles and the operation of the software.

Ensuring that software faults are rectified and performance targets are met through corrective action can prolong the life of the delivered applications. Integration of these actions through a common set of information and tools can help organisations to monitor the continuing costs of applications and to plan application replacement as operational costs become excessive.

A key question for ALM vendors is whether these benefits actually arise from the tool solutions being offered or from the organisational and process maturity, which the effective use of such tools inevitably requires.

It is very unlikely that an organisation that does not have a significant level of organisational and process maturity will gain the benefits of these tools - a fool with a tool is still a fool. Process capability and maturity are critical areas missing from ALM at the moment; it is on this basis that ALM is more packaging than reality today.

If ALM was to be strengthened in these areas how might this be achieved? There are lessons that can be learnt from the application and use of process improvement models. There are a plethora of models available (15504, TickIT, CMMI etc.) but in the scope that ALM is applied the Capability Maturity Model Integration (CMMI) is probably the most used.

Through its defined levels of capability and maturity, its model of process areas and practices and the SCAMPI appraisal process, CMMI supports organisations in evaluating their capability or maturity and in taking appropriately targeted, effective steps towards improvement. Data reported by the SEI[2] makes it clear that process maturity has a very significant impact on project delivery consistency and speed - the very benefits targeted by ALM.

At level 3, the CMMI introduces the concept of the organisation's set of standard processes (OSSP), which define how the IT organisation works. Of course the focus of CMMI is on the process areas traditionally associated with software and systems engineering, but there is nothing to stop an organisation defining processes for other areas, including service management, which are covered by other standards.

As the appraisal part of CMMI makes clear, simply to define an OSSP is insufficient to guarantee maturity. The organisation must ensure the processes are applied as business as usual. Project managers use the OSSP to tailor a project-specific process that meets their needs; project plans are derived from the tailored process.

The very direct linkage between process and plan allows feedback to be provided on the performed processes; the organisation can accurately target its improvement efforts.

Process management is a vital omission from the ALM marketplace as it is currently defined. Integration of tools that support process definition, process tailoring and process-based planning will lift ALM to the next level and make real its current potential.

Process management tools, which are already available in the marketplace, will integrate with the project planning and portfolio management disciplines already covered by ALM. Taking our prompt from CMMI, perhaps in the next generation, Application Lifecycle Management will turn into Organisation Lifecycle Management. OLM, here we come...

For further information please contact David Piper on email: david.piper@lamri.com

References

1. Butler Group (Sep 2005) Application Lifecycle Management – Aligning Software Development with Business Needs, ISBN 1-904650-26-0.
2. SEI (Oct 2003) Demonstrating the Impact and Benefits of CMMI®: An Update and Preliminary Results, CMU/SEI-2003-SR-009.