Blowing Hot and Cold on Project Based Solutions

In May 2007 R “Ray” Wang of Forrester published a report called Introducing Project-Based Solutions and a few months ago a page about Project Based Solutions (PBS) appeared on Wikipedia written by Tony Humphries from IFS. I’ve met Tony, and he has a lot of experience and knowledge. Having aroused my curiosity and also noting a small posse of ex-Artemis colleagues had joined the PBS Group in Linked In , I took a quick peek at PBS and wanted to share some thoughts.

A Quick Look-see

At the end of this blog, I’ve listed the small number of resources I’ve looked at so far, but must advise that I’ve hardly scraped the surface with Ray’s blog. I liked the first three articles I read on it, though they had nothing to do with Project-Based Solutions.

Wikipedia articles are pretty hard to write because you need to be objective, but I imagine most people that write Wikipedia articles are passionate about the subject. Particularly when it’s an idea. This article captures what it needs to and gave me a good introduction to PBS.

Project based solutions (PBS) is an enterprise wide computer software business solution to manage and coordinate all resources, information and functions of an organisation from a shared data source; where data can be attributed to and managed through individual projects.

The bit about shared data source rang alarm bells with me at first, as well as a line a couple of sentences after about the use of a local area network. I think this is partly due to what I think a shared data resource is and what the author thinks it is. I am also assuming the local area network could be a wide area network if it wanted to be. A shared data resource in my mind, is where all the applications dip into exactly the same database and I don’t see this happening anytime soon. Primavera P6 has it’s database, Oracle E-Business suite has its, Active Risk Manager has their’s Deltek Cobra theirs and so on. I do believe that through the use of APIs and even reading direct from databases there is the potential to develop services that appear as if they share the same data, and in the scheme of things, this is a step in the right direction.

I do see multiple applications from different vendors holding their own data and by using APIs and a Service Oriented Architecture it is possible, nay desirable, to have just one way to read and write data from any other application. Even making one application the holder of the master data that the others have to grab. In this scenario when a project is created in one of the applications, then calls are made to a service that ensures all the other applications have the very same project created in them. We have already seen how technology like Oracle Fusion Middleware can support this type of behaviour. I just don’t see one big database on a LAN that gets distributed around the world.

It has been possible to mash up data from different applications and make it appear as if there is one datasource. An example of this can be found with Deltek wInsight and the use of CS Glue. This combination can provide seemingly seemless access to view data that originated in Deltek Cobra and data that exists in Primavera P6 as if they were in the same database. CA Clarity also has a similar capability where columns in reports can come from different database applications.

It’s A Model But Is It Looking Good?

I like the names of each PBS sub-category: Project Critical, Project Centric, Project Enabled, Project Potential and Project Never. Though I’d like them more if Project Never was called Project Agnostic, but it hardly matters. I’m sure that if and when PBS is developed, the model will start to include the processes and types of applications expected or required to be part of a particular model. I have never been convinced with the idea that the whole of any organisation is ran as a set of projects as PBS implies. Where does the Receptionist and other similar roles fit into this notion? I do believe a project inside a software application can be used to budget for the Receptionist and record actual costs, and if Earned Value is important to be applied, then the Receptionist will belong to a Work Package with a Level Of Effort Earned Value Type.

I have seen where the majority of an organisation’s business is ran through projects, but not all of it. We’ve used the mechanism of a project to budget, book time and collect actual costs for the work of the Accounts department, but it doesn’t turn it into a project, even if that’s what the thing in Oracle Projects or Primavera P6 is called. It’s a clever “trick” to deal with an exception in an otherwise project world.

The Emperor’s New Clothes?

Those of us that spend time implementing Project and Programme Management frameworks know we agonise over the point where one system hands off to another, and contemplating how we deal with Manufacturing or Maintenance in a project-based organisation. Many years ago we were further challenged by the lack of integration capability in the tools. That’s changing for a lot of the products and lately we have seen functionally rich and faster performing APIs and web services. Coupled with technology like Oracle AIA, it is helping to make the technology a lot easier to harness for the enterprise.

The improvements in technology is a relatively small part of the equation. The greatest challenge remains the people. I’ve lost count of the number of times the Finance Department has insisted on a standard WBS to be used by everyone, even though it isn’t anything like a WBS should be? Then there is the Project Manager that is frustrated by the rejection of their perfect project plan, unable to see it is not viable due to funding issues or strategic priority? I’ve carefully examined project schedules that are difficult to follow because they are split up according to the department doing the work rather than the work that’s being done? Not since the last century have I come across a project schedule with one start and one finish. These are not issues that can be cured by applications. These are issues that are only cured by education and the experience of those people making decisions and following poor practice. I believe that organisations would get better results if they put the people issues ahead of the technological challenges of implementing ideas such as project based solutions.

Despite my preference for People before Process, and Process before Technology, I find the idea of PBS quite alluring, but am less interested in the technological challenges than I am in the implementation challenges for organisations. I have read Trends 2008: Project-Based Solutions and highly recommend the recommendations – if that makes sense. It is fairly sound, which further makes me believe what I have read so far about PBS just hasn’t been written with people like me in mind. That, or it is easier to address technology issues than people issues.

Conclusion

The idea of PBS as I understand it has many strong points. The tools we use to run businesses being able to talk to each other through services. Using a tool that excels at scheduling to schedule and one that excels at change control or another at risk management to do their job. Then knit it all together through services makes for a desirable and flexible application architecture, but will there be enough people capable of using it to improve organisational performance?

References

a. Wikipedia article on Project Based Solutions by Tony Humphries at http://en.wikipedia.org/wiki/Project_based_solutions
b. Trends 2008: Project-Based Solutions by R “Ray” Wang and and Andy Salunga which I got from the Oracle web site at http://www.oracle.com/corporate/analyst/reports/ent_apps/ppm/forrester-trends-2008-pbs.pdf
c. _R “Ray” Wang’s blog at http://blog.softwareinsider.org/