Finding Constraints in a Project Organization

After hearing an excellent interview on the prepcast, thanks Cornelius, with Allan Elder of No Limits Leadership I was inspired to see if maybe we could find "Herbie's" in our project based organization. And of course by fixing them create more throughput. This isn´t the approach that Allan talks about which is focused on the improvement in each project but he does mention that the technique can be used in almost any setting. So I suppose this should work aswell.

Following the scientific method my hypothesis is if we can find where the most projects are waiting around then we will locate constraints.

So I went to the task of identifying all of our project processes against knowledge area thinking that we have a constrained capacity in each knowledge area.

Now this is pretty basic so this is the first step of this socratic method. So I suppose so we will have to see how the story develops. You can find my spreadsheet here.

Risk Profiling the Project Stakeholders

I am really enjoying listening to Cornelius Fichtner PMP Prepcast and one of episodes I listened to today about risk had a special focus on the evaluation of your team members and stakeholders risk profiles. As I was listening to this I was thinking that it sounded pretty similar to some of those standard personality tests that we do when evaluating employees. There are quite a few on line test for your risk profile like this one. What Cornelius and Janice Y. Preston talked about was, how it is important to be aware of the risk profiles or "attitudes" in project.
From my experience this is so very true.
If there are team members who are risk averse the project will not achieve much advancement and little synthesis since using new technology or techniques is obviously a higher risk. On the other hand having too many enthusiastic and "wild" project members can lead the project down a way "out of scope" direction. Here the consensus is that it is good to have a risk mix. In a way, having people from several risk profiles can get you closer to that risk neutral ideal.

As a side thought it might be interesting to do a research study correlating risk profiles of the project members with project outcome and/or performance.

On Contracts, Stress, Peace and Dag Hammarskjöld´s Study

Dag Hammarskjöld´s Study


I put the picture of Dag Hammarskjöld´s Study here because it represents the serenity required to create good things.

In my experience the creation of a contract requires this kind of scandinavian peace and thought.

What got me thinking about this was Cornelius Fichtner PMP Prepcast section on Procurement management and I thought I would relate it to our typical vague scope or poorly defined project. Obviously there is a large amount of risk for the seller and even some risk for the buyer on fixed price contracts. But inspite of this most software development projects, many of them vauge in scope, tend be contracted at fixed price.


I find the norm forIT maintenance projects to be time and material contracts where fixed hourly rate is specified which includes overhead.

I am not sure why there is absolutely no interest in cost plus contracts in IT projects in Spain.


Another contract point which I push for is provision for project cancellation at kill points where the earlier the project is cancelled the smaller the percentage of the time and material that will be paid. This is nice because it motivates everyone to do a good job but in the end if things dont work the buyer can walk away at an early stage/gate.


In an earlier episode of the prepcast Cornelius talks about how when he worked in Switzerland his company put a special provision in contracts to assure that signoffs where implicit in the contractual relationship. I think this is a great idea and should be done more. Contracts are a great way to formally get the parties to agree on things. Since money is involved there will be a signature and in many cases as a provider you may never get your client to sign another thing again. After all why should they, if it isn't in the contract, might be their thinking even if it isnt project management best practice.


So make it count. Maybe to finalize a good contract and scope of work what one needs is a good yoga session or a look at a medow in southern Sweden - Not the stress of negociations. And negociations are always stressfull.

Earned Value Example in Spanish and English

As if there where not enough of the those terms in English!! If you add the Spanish it is an alphabet soup of values.

Anyway remember that the values are for a certain point in time of your project (the beginning of the month ie. when you do the report). And that PV is what your budget budgeted for ie. what your baseline is.

The translations:

English	English	Español	Español
PV	BCWS Budgeted Cost for Work Scheduled	Costo Presupuestado del Trabajo Planificado	CPTP
EV	BCWP Budgeted Cost for Work Performed	Costo Presupuestado del Trabajo Realizado	CPTR
AC (ACWP)	ACWP Actual Cost for Work Performed	Costo Actual del Trabajo Realizado	CATR
EV-AC => CV		Variacion de Costo	VC
CV/EV			VC/CPTR
EV-PV => SV		Variacion del Programa	VP
SV/PV			VP/CPTP
EV/AC	Cost performance index	Productividad del Costo Actual PCA	CPTR/CATR
EV/PV	Schedule performance index	Efectividad sobre la Planificación Realizada EPR	CPTR/CPTP

Quality Plans for Projects with a Vague Scope

In many projects the scope documents are very vague. Interestingly these "high risk" projects tend to avoid making usefull scope and requirements documents.

And without a reqirements document one can't deliver quality via a Project Management process oriented approach since there is nothing to measure the results of the project against.

So there are a few approaches to this problem:

1) As the project advances the requirements documents could be updated in a cyclical manner. This will make sure that as the project goes along more defined requirements will provide a better yardstick with to measure the final quality. The problem with this appoach is that it opens up the door to scope creep. But on the other hand that is not so strange since what we really are doing is New Product Devlopment :-)

2) Split the project into at least two major fases. For example at the project mid point, freeze the requirements document to assure that the final product can have an effective quality measured against these requirements document. One way to do this in R&D project where one really has very little idea of the final project results is to schedule a first prototype and then freeze the requirements after the evaluation of that prototype. The second prototype is then made to specification and change control mechanisms should be used to control scope creep.

From the business point of view option 2 provides us with a great moment to carried out a review of this final requirements document as a stage gate to decide if one should kill the project.

After all if the requirements have been watered down too much because things didn't go so well in the first part of the project this would be an excellent reason to cancel the project. Forgeting about those sunk costs takes courage but better at project mid point than at the project end.

Grand Unified WBS Theory: Far from QED

Probably figuring out the unified theory of the universe is simpler than navigating through the different ideas people have about Work Breakdown Structures (WBS).

There is the product focused deliverable flavour of WBS. Then you have the organizational functional flavour. And of course the all time favorite process/flow flavour.

What is normal is to have work packages in any of the combinations of the above with process/deliverable being popular aswell as process/funcional/deliverable which for example we are using in the Cinespace project.

• WP1.- Hardware and Software System Design (Process)


• WP2.- Scenario Definition (Process)


• WP3.- Personalise rich media serving technologies (Deliverable)


• WP4.- Geo-Position estimation technologies (Deliverable)


• WP5.- Wireless media streaming technologies (Deliverable)


• WP6.- Development of a Wireless HMD (Deliverable)


• WP7.- Collaborative Rich Media Experiencing (Deliverable)


• WP8.- Test and Integration (Process)


• WP9.- Assessment and Evaluation (Process)


• WP10.- Dissemination and Exploitation (Functional)


• WP11.- Project Management (Functional)

To add confusion the authority on WBS (PMI) classfies a Workpackage as "the lowest level in the WBS, and is the point at which the cost and scedule for the work can be reliably estimated."

The truth is that much of this is semantics. Of course the whole point of making Project Management a dispiline is to share common language so maybe it is unfortunate that the word work package has become much more common than sub project or phase but the word package is a cool word.

In any case I have seen projects with work packages which are more tens of thousands of hours, which is a pretty big package. Even in Cinespace the work packages are several thousand hours of work and completed by complex teams producing several deliverables.

Are the Cinespace WPs Work Packages in the PMI sense. I would have to say, no. They are the first level of WBS decomposition. "Work Package" really does sound better than "first level of decomposition" which has a kind of negative conotation.

OK. Fine - but doesn´t Microsoft Project actually number all of my tasks
etc and put the WBS numbers on there and even allow me to export this to Visio
and publish to my project web? Why all the fuss?

It depends. If you are working on a project where you have a PMO assigned project manager (think "big important project") then your project will probably have several development sub-projects/projects going on with inegration phases etc. and many of these sub-projects will have their own processes. At the top level you will probably mix functional, process and deliverable Work Packages, woops, I mean first level decompositions to create your WBS. This is natural since MSF is only a partial reflection of a complete top-level project.

Product Management vs. Project Management

Wait a minute. Who said anything about a product? Just
finish the project please.

The above is the all to often sad story of project management. The users are forgotten once the gears are set in motion and the project becomes a self-feeding frenzy with too little user feedback.

So this is where the product manager and product development thinking comes in. To balance the rush to finish on-time and on budget critical junctions are made where rethinking and evaluations can sometimes totally refocus or cancel projects.Thats right. At certain points in the creation of a product gates will actually cancel projects if the results are not meeting the needs (did somebody say Experience, Performance, Basic) of the user.But this spells disaster for the project leader - a canceled project?

Actually there is a way out. By considering the design and front end phases of product development as a project and by considering a prototype as a project. In this way we end the projects at the gates.

And instead of 1 huge failed project we have 3 successful projects and lots of information for the product manager to use for his next product suggestion.

Here is a link to a great posting about this topic.

Agile and PMO interaction

Here is a great post on PMO and Agile team interaction.

Adding the more formal product products into the product backlog and then getting them bumped up in priority based on business value seems like a great exercise!!

How Formal Project Management Thinking leads to Design Projects

Some project management material seems to indicate that all the basic project management processes must be used in every project.

But in fact one of the key roles of the project manager is the tailoring of the available best practices so that the most efficent and effective processes are used. In addition some of those processes will most likely be used itteratively.

Agile methodologies are the extream version of this tailoring.

The PMI view is that what we normally call phases in software methodologies such as Scope, Design, Construction and Deployment can in large projects sometimes be subdivided into individual projects.

This is a bit of culture clash from the SW methodology mindset where we consider the all the phases as one project. After all would it really make sense to think of the stakeholders of the Design phase as different from the Construction phase?

Where actually this does make sense is in high risk SW projects common in New Product Development (NPD). In these cases creation of a prototypes and the evaluation of achitectures really should be turned into specific projects (I like to run these projects in my group:-) ).

To me this more formal project management view actually answers some of the short comings in the way many IT projects are run and sold. Customers are only sold one project. We should sell them several. The first one being the "Fuzzy front end projects" - that is usability, design and architecture projects.

Some interesting thoughts about evaluating online PMP training offers

Consider the following. MIT is making all of their courseware available online. So people could do all those courses online with the $100 laptop and then take an exam at a small training center in a small village in India for $30 or they could take the exam at MIT for a total cost of $180,000. Who would know more?

Here is an interesting debate about online PMP training.

Interesting debate here. For someone who is evaluating PMP course material it is really difficult to know the value of specific courses. One never knows if online reviews are real. But I tend to agree with John, just because the training is less expensive doesn't relate to its quality.

For example we have:

Cornelius Fichtner PMP Prepcast podcast training which offers 35 PDUs for $49

PMCampus´s 35 PDU online course for $199 (registered PMI Education Provider)

PMTrainingOnline´s $400 for a PMP preparation course (Is a registered PMI Education Provider through Skillsoft)

Looking at the different online training sites I get the feeling that what this is really about is that the basic material can´t get much better than some of the excellent PMP preparation books like Rita Mulcahy´s.

So maybe I will check the prices on tickets to Mumbai, always wanted to visit India ;-)

By the way I was very happy to Gilbert Stang´s picture on the MIT site. His text book on linear algebra is one of my favorite books fo all time.

Reviewing past Projects for PMP

Its actually a pretty interesting exercise to go back and review the projects one has worked on in the last six years for the PMP application.

What becomes clear is that there are a lot more project initializing phases that closure phases :-) And that at some point one is actively involved in far too many projects than can be healthy to maintain your sanity.

A good way to visualize this data is to create an excel spreadsheet with all your projects, names, start dates, end dates etc. which you will need for the PMP application.
Then subtract the start and end dates and divide by the hours spent on initiating, planning, monitoring and controlling and closing processes. You will need to reduce by some constant to account for holidays, weekends etc. assuming you do those things. And voila you have the average hours worked per day.

If you have the data in horizontal use the special copy function of excel to transform the data. Organize the columns in MS Project and just past the data right in. The customize the bars in the Gannt chart and you will have something like the image above.
Since I am very much hands on in my management style as I did this exercise I saw the gaps where I remember that I was coding or developing some specific algorithms. I also saw periods where 10 hours days where the norm.

Another interesting thing is to color code the projects according to their success in a kind of heat map. I guess most of us have times in a careers when we are in the "flow" !!

View Multiple Baselines in MS Project

Since we work with projects from the very early product development stages many projects sufer fairly radical changes in scope. So in many cases it can be intersting to track the changes via baselines. If you are using project there is a little know view called multiple baslines available in the "more views" pop up in the view menu.

Here is nice article on this topic.

Web2project Open Source PM System

This project management system looks promising. Looking forward to the beta. The demo features make it look very complete. To bad it wasnt done in Django or Ruby on Rails.

Looks like a release version is imminent.

Project+ Certification

A good introduction to doing the PMP certification before actually getting the PMP certification is the Project+ certification which offers a certificate in project management but doesnt require as many hours of experience.

Here is an excellent pdf with information about both the PMP and teh Project+ certifications.

And more information about the Certification from CompTIA.

In Zaragoza the FORUM ARAGON S.A.L and RANDOM FORMACION SL offer the Project+ certification exam.

Project Management vs. Project Leadership

There is a great post here about some ideas on the differences. Also make sure to read the first comment it is seems like it is in dissagreement but actually agrees with the fundamental concept the article that one must lead to manage effectively.

In the article Brian Sullivan says,"...management is a reactive tool to whatever situations happen to crop up. When problems develop, these executives respond. When they pursue action, it's on familiar terrain or through time-tested strategies. "

And in the response Ben Simonton comments, "But for managing the resource named "people" we have no well-defined body of knowledge."

Then continues to add, "I learned that leadership is the only way to manage people because people follow the value standards reflected by what they experience in the workplace. Leadership and following are opposite sides of a coin called values. What workers experience is the boss' leadership and is what they will follow. If the norm is to give orders to workers and since orders are considered demeaning and disrespectful by the everyone, this leads people to demean and disrespect their customers, their work, their fellow workers and their bosses."

A little bit Bitter but good Content on the PMP Credential

Here is a nice post about the process of taking the PMP exam and getting things working. Some recomended books and more.

A Interview with PMP Candidate in Spanish

Nothing to specific about this interview but generally positive.

QFD, House of Quality and Software Development

Here is an interesing article about using traditional manufacturing product design tools in software engineering. Six Sigma in general and QFD in specific are almost never covered in basic software project management programs.