Project lifecycle in its basic form consists of the stages that the project goes through from initiation to closing. It has a collection of logical project activities that are responsible for some aspect of the project.
The chart above has so much information that you may be able to answer few questions on the PMP exam just by studying this chart.
So, take a few minutes and look at each of the lines and interpret how those variables impact the project.
A project is typically divided into four phases –
- Starting the project
- Organizing and Preparing
- Carrying out the project work
- Closing the project
The Project manager does certain work in each of these phases and produces certain project management or project output.
Starting the project
In this phase business case is prepared, high-level requirements, constraints, risks, and assumptions are understood, a feasibility study is typically done and all preparatory work required to get the project off the ground is accomplished. The output would be a Project charter. This is also the document that officially authorizes a project manager to the project and outlines her responsibilities and authority.
Organizing and Preparing
This is where most of the planning – at least high level and a bit of low-level planning for high priority features (if Rolling Wave planning is followed) is done. All other domain knowledge-specific plans called ‘subsidiary plans’ are prepared – these include risk planning, human resource planning, communication planning, procurement planning (if required), and so on.
Certain baselines such as scope, cost, and schedule baselines are prepared. The Project management output document will be a Project management plan.
Carrying out the project work
This is where all the action happens. Project deliverables are prepared, changes to scope, schedule, and cost baselines are managed, communication and stakeholder engagement are handled, procurement activities (if applicable) are conducted and deliverables are accepted by the customer. Vendor procurement can be closed in this phase.
Closing the project
This is where formal project closure happens. As part of this, administrative closure, updates to organizational assets such as ‘lessons learned in this project’, and release of team members are done.
Impact on certain project variables on project lifecycle
Understanding this aspect is very crucial for project success. Please refer to the above graph.
The project cost is lowest at the beginning, and increases as the project progress. This is because the hiring of team, purchase/lease of equipment, procurement payments, defect repair costs add up as work progresses. The highest spending happens in the Construction phase, or the phase where project work is executed. This is depicted by the blue curve.
There are few variables that are highest at the beginning of the project and wane down as time passes – these are the influence of stakeholders, project risk, and uncertainty associated with the project outcome. This is depicted by the orange curve above.
And then there is a variable whose degree steadily increases as the project progresses, this is the cost of changes. In the beginning, when not much work has happened, this is low. Most of the changes (even major) can be made without much cost. This gets progressively harder as the project progresses due to the impact of changes on the already accomplished work. The more the impact, higher is the cost.
Hence, the adoption of Agile project management methods such as Scrum is increasing. This approach favors what is commonly known as “Rolling wave planning” – where detailed planning of known and high priority features are done, and other features are only planned at a higher level.
Types of Project Lifecycle
There are 2 types –
- Predictive lifecycle
- Adaptive lifecycle
Which lifecycle is suitable for the project is something that is decided by the development method used, industry, external dependencies, environmental factors, or even the technology used.
Each of the stages when a project passes through must move over a point called Phase gate. This is a point where project outcomes are compared with planned outcomes and the decision to go ahead to the next phase or not is made.
Phase gate has other names such as Phase review, Control gate, or Kill point.
Interesting names, eh? 🙂
What’s the relationship between these?
We know that a project lifecycle is the broadest on the timescale, and it contains multiple phases.
Now, one or more of the phases may include the development of a product, service, or result (remember the definition of a project – temporary in nature and creates a P, S, or R?) – and this is done by choosing a corresponding development lifecycle.
Types of Development Lifecycle
There are 5 types –
Predictive development lifecycle
Also called waterfall development lifecycle. This is how traditionally all projects have been managed. Here the project scope, schedule, and budget are determined early in the lifecycle. Most of the things are pretty much written in rock.
This approach however results in plenty of rework and thus increased cost.
Iterative development lifecycle
This is an improvement over Predictive approach, where only the scope is determined early on, the rest are determined at the onset of each iteration in which the product, service, or result is created.
In this type of project life cycle, activities are performed in a repetitive fashion in a series of repeated cycles, called Iterations. The product is built in an incremental fashion and delivered at the end of each iteration.
These deliverables are fully tested and ready to be pushed to production. Subsequent iterations may either enhance these deliverables or build something new.
These iterations may run within phases, and these phases can be sequential or parallel. Also, the iterations may run in a parallel or sequential manner. The decision to run them in parallel depends on what is being produced in these iterations and the amount of risk appetite the project wants to have.
This approach is used for projects that are large or complex, where risks are high, the scope is not completely known upfront and is discovered as the project progresses. Research-based projects typically follow this approach.
At the end of each iteration, a ‘usable’ deliverable is created.
Incremental development lifecycle
In an incremental life cycle, the deliverable is produced through a series of iterations that successively add functionality within a predetermined time frame. In the incremental approach, the complete workable deliverable is available only after the final iteration.
Adaptive development lifecycle
Adaptive life cycles are agile, iterative, or incremental. The detailed scope is defined and approved before the start of an iteration. Adaptive life cycles are also referred to as agile or change-driven life cycles.
This is a bit advanced version of the iterative and incremental life cycle where scope unknowns are high. The main difference is that the duration of iterations is very small.
Consider Agile software development practice. The product is released to production over few release cycles. Each release consists of a few smaller (2-4 weeks, typically) iterations. Each release of the product comes out in a span of a couple of months, and then the next release of the product begins. Pretty much the same set of activities are performed in each of these iterations – design, development (construction), testing, and release of deliverables.
The team dynamics here are very different from earlier types of project life cycles. The typical team size is 7 plus or minus 2 (that is 5-9 members). The team is completely responsible for committing to the scope of iteration and delivering it at the end of the iteration. The project manager here dons the role of a facilitator, protector, and enabler for the team to do their work in the most efficient manner.
The team maintains a prioritized list of broken-down features called Product Backlog. A certain number of top features are taken for an iteration from this list and broken down further into smaller tasks. Team members are allowed to pick the tasks they are most comfortable with.
At the end of the iteration, the customer can see the completely tested features in working condition. This however does not mean that customers can accept deliverables; it only indicates that the product should not contain unfinished or unusable features.
There is no reward if the team over-achieves the scope of the iteration, and no penalty if it under-delivered. Eventually, the team finds its ‘velocity’ to deliver the maximum amount of features per iteration by working in a collaborative way and achieving optimized productivity.
Hybrid development lifecycle
In practicality, though the choice of a development lifecycle is governed by many variables, and hence most often than not a hybrid of multiple above development lifecycles from the above list are used in projects.
In the next study note, we shall look at ways to develop a project charter. There may be more to this than you may be thinking…