Clarity Project Management is all about delivering valuable information management solutions to the client and their organisation. The disciplines we cover require expertise in both waterfall and agile project management methodologies.
Project Management for BI and DW
It is a common notion that Business Intelligence and Data Warehouse implementation projects are highly expensive, do not always end as per the project schedule and produce less or no benefits on completion. You can argue that this is the typical case of any IT project! However, these risks are often mitigated by strong project management alone.
BI/DW projects, just like any other software application development, rely heavily on the requirements, design, effort and schedule and effective execution of the project per the defined plan. This then raises the question of why an experienced project manager, when handling a BI/DW project, fails? The reason is simple. Though a BI/DW project is similar to that of software development projects, there are some specialisations in a BI/DW implementation that are mostly ignored or missed leading to cost/effort overrun and slippage in delivery time. Below we will look at those specialisations along with some best practices that can be tailored per the individual project needs to successfully manage a BI/DW project.
Project Management – The key to success of any project
As with any software development project, a BI/DW project too needs to follow a methodology that will help in planning, monitoring and controlling the project to successfully achieve the project goals within the planned effort, schedule and cost. A project management activity is one, which will be present right from the start and last till the end of the project. Let us look at some of the key areas of a BI/DW implementation that a project manager has to concentrate all through the projects life cycle.
Any project to become successful must have all the scope of a project defined and documented. There has to be a sign off on the scope with all the key stakeholders. Scope creep will always lead to rework, additional effort and cost and ultimately delay in the end deliverable.
Project Cost and Time
It is essential that the costing for the project is defined at the start and monitored closely. This has a direct impact on the benefits, more so in a BI/DW project as the organizations top management and business decisions are arrived at based on the value added by a BI/DW solution. Similarly it is vital that the project is executed per the project plan. Tracking this will help the project manager to provide regular status reports to all stakeholders.
Project Risks and Mitigation
Data is the heart of any BI/DW project. Hence it can be easily stated that most of the risks arising in the project are from this data. A well thought out risks that can arise from the project with a proper mitigation plan will help the project stay on track, both in terms of cost and time.
Any project should deliver a quality product as the output. Here again the quality expectations must be defined at the start of the project as part of the requirements and specifications. Many unsuccessful projects’ history will show that this quality step has often been neglected.
The reason that project benefits has been listed as the last item is that this is the most difficult part of any BI/DW project, as they tend to be 'soft' benefits such as improves access to data and information, single version of the truth etc. Though it can be difficult to measure direct benefits of the BI/DW solution, any other measureable like effort and resources needed to generate monthly reports can be used. It is sometimes useful to try quantify opportunity cost i.e what is the cost of not having ready access to the information. In a modern competitive environment, where products and services are very similar, it has been said that one of the last true differentiators between organisations is the information they collect and how they use it. Expanding on this theme can be a great way to sell the BI/DW story.
Tailored Project Management for a Business Intelligence and Data Warehouse project
It might be difficult for a project manager to use the standard project management methodologies for a BI/DW project. There is some amount of tailoring that one must make in order to adapt to the needs of a BI/DW project and successfully complete it. Some tailoring aspects are discussed below:
Multiple Technologies and skills
The essence of a BI/DW project is the variety of technologies, tools and skills that are involved. All the stakeholders will have different technical and functional skills. These resources must come together to make each project a success. There may be communication gaps, technological gaps that a project manager must understand and collaborate with these two different sets of individuals and create clear and proper roles and responsibilities to make the project a success.
As BI/DW projects often have both data (DW) and reporting (BI) focuses, it is vital to ensure the leads of each focus area are effective communicators and understand the drivers and importance of the other.
Dependencies on External factors
Typically a BI/DW solution means the coming together of different applications, specifically third party applications, for which the development team and the project manager do not have any control on. The same is also true for the data that comes from disparate sources. Hence the project manager should establish an efficient issue tracking system to handle these issues and be in control of the project.
The key to successfully being a project manager in a BI/DW project is to understand that the methodology that is used should be agile for BI and DW design and to a certain extent a waterfall methodology for the DI phase. The reason being, is that it will be highly impossible to freeze the requirements at the time of project kick off. When an agile methodology is followed the project manager will understand that requirement change will be inherent.
Formulating a strong change management process will help the project manager keep track of changes and implement the changes in a phased manner.
Follow an Iterative approach
Another key aspect while following an agile methodology is that an iterative approach has to be followed. At regular intervals, users should be exposed to a working "product", such as a report or set of transformed data. This will enable the user community to communicate changes as early as possible and will lead to easier control and project progress assessment.
An iterative approach will also help the testing team (if it exists!) to test and validate all features, which again will help the project manager to execute the project better.
From a user perspective, their interest and motivation will remain high if they are actively involved at regular intervals.
Involve All Stakeholders and Users
With the enormity of a BI/DW project where many business users from varying background will provide requirements and test the solution, it is critical for a project manager to bring together these individuals and involve them in the project right from the start to the project completion.
Though managing a BI/DW project is basically similar to that of any software development project, which requires planning, monitoring and controlling, understanding and adapting the tailored features will deliver the expected project benefits and results.