Service provider takeaway: Business intelligence consultants have a few options for getting various team members
and systems involved in a BI project to work cooperatively.
In my career in the business intelligence (BI) consulting space, I've sat in on presentations of project management (PM), collaboration and system development lifecycle (SDLC) tools and have observed an interesting behavior -- as the software presentation progresses, the different constituencies in the audience will show interest only in areas pertaining to their defined roles.
That behavior is no surprise: A common BI project scenario has the PM/PMO (project management office) focused on the project plan. Collaboration tools, such as SharePoint, are used to store and access project documents (like work plans and requirement documents) in Word-format ETL mappings within Excel spreadsheets.
The project assets and deliverables are created by many different project team members, and the resulting collaboration repository usually lacks intelligent integration of the assets and deliverables. In addition, the SDLC tools have little to no direct tie to the project plan and resources assigned to complete the development. Most importantly, the various tools have no connection to the linkage of the system development function to either the upstream analysis that determined the need for the function or the downstream result of the design/build in the testing and user acceptance processes.
In the world of business intelligence consulting projects, where the need to satisfy business objectives is essential, it always struck me as odd that a more holistic approach to BI project delivery is not used more. Seamless integration of PM, collaboration and SDLC is critical to project success -- not just the ability to track and manage tasks, but also to ensure that the overall BI program is achieved and can be leveraged for other BI projects. In an optimal project delivery environment, the integration of the three should be easily accessible, display the association across the project continuum, provide flexible search filters for project elements, supply the capability to report on any filtered criteria and make all project assets reusable.
The traditional project approach raises several questions about the ability to achieve BI consulting project success:
- If a project plan shows that a task is 100% complete, but a PM or other project team member has no details on what the task required, how does one know that the task has truly been satisfied?
- How can a project team and PMO share and interact with project team members to easily organize and access related project plans, progress of task completion, project analysis, design documents, testing plans and so on without a singular environment?
- How can you assure that a designer, developer or tester will satisfy a requirement or project objective if the developer has no way to check for completeness and accuracy?
- How does an organization enforce proper workflow promoted in their BI methodology when the "hand-off" is a purely manual function and depends solely on the project team members?
- While master data management (MDM) software is good at storing and providing access to one source of the truth for data definitions contained in operational BI applications, how does an organization assure these accepted and agreed-upon definitions are adhered to during the project development process?
Achieving a project environment that integrates the entire project ecosystem is a significant challenge for most organizations. While some PM, collaboration and SDLC tools provide linkage among them, few are designed to work seamlessly together. Most organizations manually map project work plans to the actual system design and development processes while using another third-party tool for team collaboration. This loosely integrated solution lacks intelligence and workflow capabilities. Some organizations use a suite of individually purchased SDLC tools that can share information among components, but this only solves part of the problem and is very costly.
So how should you address this problem in your business intelligence consulting projects? There are a few options, each suited for projects of different sizes. The first method, good for smaller projects, is to have team members responsible for cross-functional delivery roles. For example, one individual would oversee PM, architecture and development. This forces the team to collaborate and understand the impact of decisions across the project lifecycle.
On large projects, this approach is not always viable because of the scope and complexity of the work. In that case, you should recommend a modified version of the first option: Break down the project plan into manageable sections, then use the cross-functional role assignment detail above and have a higher-level group of experts overseeing the entire program.
It's important to note, however, that these aren't perfect solutions. They require organizations to find and retain individuals experienced and talented enough to wear multiple hats -- a very intensive manual undertaking.
The Holy Grail -- something that's not easily adopted or affordable -- is an application that embeds a best-practice BI methodology within a single integrated PM-collaboration-SDLC tool specific to business intelligence consulting projects. Unfortunately, such an integrated application doesn't yet exist. The individual components do, but you or your customers would have to buy and then modify the packages to share information.
As the BI market continues to mature, those of us on the front lines of project management and delivery can only hope software suppliers will address the need for an integrated collaboration tool.
About the author
John Onder is a partner in the Chicago Business Intelligence Group, with more than 20 years of experience. He has been involved in all facets of information technology strategy, planning and development engagements, as well as management consulting, business strategy and reengineering projects. He is an expert in the field of business intelligence, data warehousing, CRM analytics and decision support systems.