software product development | The Weakest Link
post-template-default,single,single-post,postid-593,single-format-standard,ajax_fade,page_not_loaded,,qode_grid_1300,footer_responsive_adv,qode-content-sidebar-responsive,qode-theme-ver-11.1,qode-theme-bridge,wpb-js-composer js-comp-ver-5.1.1,vc_responsive

The Weakest Link

The Weakest Link

Abstract: The software industry has established several advanced and sophisticated methods to help organizations with
their process improvement programs. Those methods use models, metrics or agility indexes and are generally intense to
implement. This article presents an alternative, a simpler method that can get organizations, in particular small settings, to
improve through quick-wins.


The context

The software development process is a value chain. Through a succession of value-adding transformations, the requirements of a project ultimately end-up
in the form of production software. The identification and removal of the weakest of the transformations of the value chain within the organization
constitutes the core of the proposed process improvement method.

The challenge

But let’s start the discussion with a quick look at the current process improvement trends. The industry has developed very advanced and
sophisticated process improvement methods. Some methods are based on capability and maturity measurement (i.e. CMMI®) and will help an
organization benchmark themselves against best-practices and implement well planned improvement strategies. Others, based more on metrics, scorecards or
even near-scientific approaches will measure the organization performance in a purely quantitative way and rationalize an action plan. All those methods have
proven to be very successful and efficient; they however all require a time and budget commitment that many organizations cannot afford.


The solution

The proposed method described in this paper, uses a simple analysis of the software development value chain within the organization, to identify possible
quick-wins. The software development process should be seen as an ecosystem, where various people (roles) use processes and tools to transform raw material
(the project or product vision statement) into a software solution. For examples, a business analyst will transform a vision into requirements, an
analyst will transform use cases into a solution design, a developer will transform a design into source code, a tester will transform non-tested code
into tested code… Even related disciplines such as project management and configuration management are transformation activities, i.e. from an unplanned
set of activities to a planned and resourced development cycle.

Now that the concept of transformation activity is defined, we must look closer into the software ecosystem. By doing so we can notice that at least three of its
properties should be highly considered for the sake of process improvement:
– Firstly, it is important to acknowledge that that each transformation activity is adding value to the previous transformation activities in the value
chain. The performance of the overall ecosystem is really the sum of all the adding-value transformations, each transformation contributing its own piece
to the overall result.
– Secondly, the software development value chain is such that the transformations can only deliver high value-add if the previous transformation
activities in the value chain were also of quality. For example, an architect creating a terrific architecture is of little value if the requirements he was
working from were incorrect.
– Thirdly, the value chain is such that any transformation activity in the ecosystem could jeopardize the overall effort, if done poorly. If the developers
of the organization could not code even to save their lives, then the end-result would still be substandard.



So the conclusion is that to improve the efficiency of the overall ecosystem, one must eliminate the transformation activities that are not adding the
right value: the weakest links.
The idea behind the proposed process improvement strategy is to improve, one by one, through natural selection, the transformation activities that are
not bringing added-value.
By doing so, we enable a better overall result at the end of the value chain. The method consists therefore of focusing on one transformation activity at
a time (the weakest link) and improving it until it is not the weakest link anymore. As soon as the weakest link is not the weakest link anymore, then
it means a next weakest link should be identified and corrected, and so on, making the overall value chain stronger every time.
Transformation activities bear more weight on the final result when they are earlier in the value chain. So when two transformation activities are
found equally weak, the one happening earlier in the value chain should be tackled first.
The method offers several advantaged over more “industrial” process improvement methods:
–  It only requires a small investment at a time, only what is needed to identify and correct the weakest link.
–  It enables the quick-win effect: Each win boosts the morale of the team as the impact has immediate value, for the smallest possible effort.
–  It is less disruptive of the overall ecosystem.
–  The return of investment is the highest at the beginning, delivering immediate boosts in quality, time-to-market and company bottom-line.
–  It can be implemented using Agile project management, iteratively and incrementally