LevelTen's Experience With Two Project Life Cycles
While there are 4 project life cycles adopted for software and web development, LevelTen primarily works within the Waterfall (also known as Serial) and Iterative cycles. While all project life cycles provide advantages and disadvantages, these two have been widely accepted and adopted by software, IT, and web companies alike.
Both were developed to manage project risks, but one is probably more suited than the other depending on the amount and type of risk involved you your project. To explain, let’s first start by providing a brief overview of the Waterfall and Iterative life cycles. If you would like a very in-depth understanding of life cycles we recommend reading The Pragmatic Programmers, “Manage It!” This book is fantastic for beginners and experts alike.
The Waterfall Life Cycle
Using the Waterfall life cycle, your team is supposed to be able to first obtain every conceivable requirement (Project Planning and Strategy). Based on those requirements the team moves into design to determine the look and feel of the system. Once everyone agrees on the “big picture”, the team starts developing. All pieces are developed, completed and integrated before testing begins.
Serial life cycles take longer because they are supposed to be able to predict how much a project costs, how long it will take to implement features, and how quickly defects can be fixed – things that are inherently unpredictable.
Many business stakeholders prefer waterfall life cycles and fix-bidding projects up front with a very high-level overview of features and services, and for good reason. It’s very reassuring to know exactly how much a project will cost before you spend any money on it. Basic websites which include static, brochure-type pages thrive from this type of development process because there are few technical features with less complex programming requirements. However, if your web site requires functionality including e-commerce, CRM/CMS integration, or any type of community involvement, we highly recommend utilizing an iterative, incremental, or agile life cycle. Trying to use the waterfall approach on these types of projects will most certainly set your project up for failure.
The Iterative Life Cycle
LevelTen uses an iterative life cycle for projects that require more complex planning and development. The incremental life cycle looks a bit like the waterfall approach at the beginning of the project for the requirements and analysis phase, but utilizes chunking features (as seen in illustration above) through the remainder of the project. This chunking helps align and manage expectations by allowing the customer and project team to see working prototypes of pieces of the system as they are created. This early review allows much greater flexibility for development, testing and changing so the customer gets exactly what they want.
Instead of spending 2-3 months in preparation and low-level requirements gathering, then building out all features and testing them before release, we recommend spending one month understanding the feature sets, determining which features are most important to the customer, then developing, testing and releasing features on monthly increments.
While the project’s timeframe and price are not fixed at the beginning of the project, the inherent nature of developing this way has proven to reduce cost and increase the potential of success exponentially.
The following diagram draws
The following diagram draws from the sphere of engineering, showing the "traditional" (also known as "classic") and most straightforward approach to constructing software, the Waterfall Software Life Cycle Model. You can modify this plan to your needs, as the aim is to give you a starting point from which you can properly consider the work ahead. All modern software development models, including RAD and XP, have their roots in this approach.
On the diagram, there is some scale implied in the length of the boxes left to right. Notice that the arrows can feed back up the process. This is important because if things start to go wrong somewhere down the line, you may need to back up one or more steps to get to the reason
Idea
This is literally just the idea, the first inspiration, the vision. This really is the realm of dreams, but realising the dream is where the Waterfall model comes in.
Analysis
This is the stage where the nitty gritty of the idea begins to be explored. What is involved? What will it need to make it happen? How long will it take? And the all important question: is if feasible?
Often an idea will be reconsidered, or at least tempered at this stage, and that's fine. In fact, it's probably desirable, because it means that the concept is being fully explored, and thought is genuinely being given to what can be achieved with the resources available. It's a lucky idea that has unlimited resources!
Design
Feeding from your analysis, you begin to design. This stage will often feed back and forth a lot with the analysis and the idea, because issues can begin to come to light at this stage that perhaps weren't apparent previously. The design should of course be based on the analysis, and for the project to have any hope at all the design must keep in mind the constraints that the analysis has put in place.
Construct
Clearly, this is where the project starts to be built. There is obviously a level of commitment at this stage, and the previous work is relied on. There is still the facility to move back up the waterfall if required, even as far back as rescoping the idea if necessary, but the further down the waterfall a project has gone the more damaging it is to cycle back up to earlier stages.
Test
Test test test! Then test again!
Seriously, it should be very obvious that almost a full half of the project's time scale is taken up with testing. Don't underestimate this, as you can easily end up with a very expensive piece of binary junk. Testing of software, from websites to banking applications, is a huge subject, but it's a simple concept. Just think how frustrated you become when you use software and perhaps lose hours of work because of a bug. And that's just a minor problem. Internet crime is largely able to exist by exploiting bugs and weaknesses in poorly tested software. So, whatever you've made, test it as thoroughly as you can.
Implement
This is when you unleash your project on the world! Of course if the idea has been properly analysed, properly designed, well constructed and fully tested, it will simply work straight off the bat. Sadly, this is rarely the case. So, this is an intense period, where there may be much flurry of new analysis, redesign, reconstruction and renewed testing. Regardless of how well you feel that you've gone through the process, it's best to have a everybody involved on the project alerted and ready for action at this point. If it works out, great, but if you have to react, make sure that you can.
Maintain
You need to keep up with the project. In part this does mean that you should watch for bugs and other errors that are reported to you, and fix them. Yes, even your extensive testing regime may well have missed the odd issue here and there!
Just as importantly, though, is to keep developing the project. New ideas, new improvements, and simply bringing things up to date with the changing world (new currency for example!) are just as important to maintaining a project. Software and websites rarely need to die, as long as they are maintained. Look at everything from Windows to the Yahoo! website - the key to their longevity is that they are constantly renewed and updated. Keep the project fresh, relevant and useful, and it will survive.
Libros Gratis
Project Life Cycles
I found this post to be very informative. Maybe we could make some sort of chart or diagram to show the process of choosing a cycle and the steps involved in each. Then we could put this potential diagram on our brochure site.