There was a post on Public Sector PM asking what is the role of the Project Manager on Agile Software Development projects. This is a continuing question that has been asked before. And as before usually without a context or a domain. This means if you ask people who work on agile projects "what's the role of a Project Manager on your project?" you'll get the predictable answers from their personal experience based on their own context and domain.. But it turns out there is more to the question and the answer than normally talked about.
What's missing here is the Real question.
How do we increase the probability of success for a project?
By this I mean, how can the stakeholders have a sense that the project has some chance of success. The measures of that chance is the measure of the Probability of Project Success (PoPS). This PoPS paradigm is found in the Department of Defense approach to Program Management. While it may seem a stretch for commercial domains to take this approach, ask a simple question from the point of view of the decision maker - a person or persons whose money you are spending. Many times lots of money.
When will you be done? What will I get for my money? Does the thing you're building me provide the capabilities I need to increase the business value of our firm?
This last question is where folks like Scott Ambler would like us to go with Earned Value. But that's not the role of EV, so "business value," will have to stay where it should be - a measure of physical percent complete compared to the Planned Budget for the work (Budgeted Cost for Work Scheduled). And the strategy Measures of Effectiveness (MoE), used as the starting point for measuring "business value" - just like the strategy folks tell us to do, from Balanced Scorecard to IT Governance.
So here's a few questions that need to be answered before deciding if project managers have a role on an agile project:
- Do we know what "done" looks like in some form that is measureable to the decision makers. Counting stories or even measuring Earned Value is not that measure.
- What is our plan to get to "done"is some way connected to the business strategy. This means describing the order of the business capabilities that will appear for use by the business.
- Do we have enough time, money, staff, and other resources to get to "done?"
- What impediments will we encounter along the way to "done?"
- How do we know we are making real physical progress toward "done?"This is called risk management.
If you've followed this Blog, these should be familiar by now. They are the 5 Immutable Principles of project management. In this case the real question is who answers these questions? Does the agile software development team answer them? Does the product owner? Can the product owner on a Scrum Team answer them for the corporation.
Here are some bigger questions?
- Why do we need this project?
- How does this project connect to our strategy for our business (assuming we have one)?
- What are the units of measure in making this connection to the strategy?
- How will we know we're producing value for the business?
This last question seems to be missing an answer from those claiming EV is not the answer. What are units of measure of Business Value? They can be KPI (Key Performance Indicators), MoE's (Measures of Effectiveness), or maybe RO (Real Options), and maybe some balance sheet values.
But let's start with the pre-conditions for success here. Vincent Barabba suggests in a Russel Ackoff Conference in 1999 that for management to successfully pursue the creation of value for the enterprise they must:
- Have an unambiguous sense of direction permeates the organization. The mission of the enterprise is known and understood by everyone – it is the universal premise behind all decisions and tasks, and it is focused on finding better ways to gain, develop, and -- most importantly -- keep them.
- Possess strategic and operational plans reinforce each other. There are no downstream disconnects between activities.
- The decision-makers must understand how their roles contribute to the total enterprise, and their accountability is clear – all the arrows are aligned. · Fourth, there are no simplistic ideas about how customers or competitors will respond to the actions of the enterprise: planning and execution recognize the full complexity and uncertainty of the market.
So who plays this role? Who answers any and all of the questions here? Software developers? Scrum Masters? Product Owner?
These questions and their answers are called...
G O V E R N A N C E
Project success in many domains (enterprise and especially when you're spending significant amounts of some one else's money) is guided by Governance. A good place to look is IT Governance: How Top Performers Manage IT Decision Rights for Superior Results, Peter Weill and Jeanne W. Ross, Harvard Business School Press. Many want us to believe that Agile is the solution, and it is likely the solution to developing software. But we need higher level processes to allow Agile to succeed. Governance is one.