How to Adapt Governance for Agile Projects

How to adapt governance for agile projects

By Christopher Worsley

If you’re working in a structured project environment with a project office, the chances are that you are using a right-size governance approach.

What does that mean?  Essentially, the level of management attention and oversight varies appropriately, depending upon the characteristics of the project, such as size and complexity, or the level and significance of the impact of the project on the organisation. 

In the example below, projects are classified for governance purposes into three types based upon size and complexity.

Governance on Agile projects

Project categorisation

Type ‘1’ projects demand formal ratification of key deliverables such as the business case and project initiation document.  They will not be allowed to continue (or at least that’s the idea) until there is real evidence that the legitimate governance stakeholders have given their authorisation to proceed stage by stage.

Type ‘3’ projects?  Well, they typically take only a few weeks, a few staff, and not a lot of money, and have a very limited impact on the organisation’s strategy.  They often simply require a sign-off as an agreement to operationalise and close the project.  It’s not unusual for work to continue while the sign-off is being negotiated.

Ultimately, the choice of governance affects the way the project is controlled, monitored, and the way decision-making is managed.

Governance for Agile projects

It all sounds good, doesn’t it?  Except that it simply isn’t working.

Right-size governance is failing on so many levels

Twenty years ago, most of us recognised that adopting strict life cycles and gateway processes may well reduce risk, but at a huge cost to the agility of delivery.  It didn’t matter how many times PRINCE2 theoreticians told us it was ‘just a framework’ and its strictures must be adapted to the project and organizational context, there were methodologist practitioners who were determined to implement a rigid, formulaic system.  This was the era when the joke was:

What’s the difference between a methodologist and a terrorist?

“You can negotiate with a terrorist!”

Right-size governance was introduced to proceduralise the judgements about which governance techniques to apply and when.  It makes logical sense.  In line with Pareto’s law, you concentrate most management attention on the top 20% of projects.  But like so many well-intentioned ideas, it had not factored in the Machiavellian behaviour of organizations and their project stakeholders.  In a review comparing the actual governance approach taken by projects against the approach suggested by project size and complexity evaluation, we found over 25% of projects were not in line with the right-size governance recommendations.  Here are just some of the examples we come across:

  • Under-inflation: When projects are misdescribed as ‘simple’ to ensure low levels of governance oversight.  “I know I said it was large, but actually it’s quite straightforward – I’m sure it’s a type ‘3’.”
  • Over-inflation: When project classifications are confused with project ‘status’.  Yes, it does happen!   “This is definitely a type ‘1’ project.  Look how important it is!” Perhaps we should read here – “…look how important I am!”
  • Process override: When there are clear indications of a project of being one type, but alternative governance approaches are mandated, often by a powerful stakeholder.  “I don’t need all this, and I’m not prepared to pay for it.”

Right-size governance so rarely deals with change

In the organisations we work with, the classification of projects for governance purposes is part of the project initiation process.  It must be because fundamental decisions to questions are dependent upon understanding the nature of the project. Questions such as:

  • Who will be involved? 
  • What level of project manager skill is necessary? 
  • How should we register the project?

This works well where the level of simplicity and complexity is obvious, but for those in between, it can be more difficult to predict in their early stages.  We don’t know what we don’t know yet.  Selecting and implementing the governance for the project at this stage is a problem.

The PMOs we work with report that it is often these projects that get into trouble simply because the management oversight is just wrong.  While there may be good intentions to review the project categorisation at stage gates, in reality, what happens is that the project drifts into a governance black-hole with nobody prepared to expose the existing governance regime to challenge.

The PMO’s role

Clearly, with some projects and programmes lasting over several years, the governance approach must be reviewed.  In these circumstances a PMO can add real value, monitoring the risks associated with projects in the wrong governance state and highlighting the need for change to occur.  We suspect, however, that many PMOs are subject to the ‘magpie effect’ – they become overly focused on large projects and programmes.  Strange really, because these are the ones we assign our most experienced (and costliest) project and programme managers to – exactly because of the known risks.  It seems an exercise in project manager disempowerment for the PMO to pitch in as well.  Rather it is those middle level projects where changes in context are most likely, and where the skills and experience of the managers involved may be more suspect, that the PMO should focus.

Adaptive governance

Reviewing your governance approach is one thing – adapting it is something else.  Indeed, one might consider that the whole idea of adapting governance is an oxymoron.  After all, the purpose of governance is to give predictable approaches, based upon best practices, to reduce the risks associated with the management of projects.  Adapting governance – well it sounds like the sort of can that is best left unopened!  

Yet if yours is a complex project environment where the organisational context of projects is varied and varying, or indeed if you are working on Agile projects, adapting governance approaches is exactly what you are expected to do. 

As governance is about reducing management risk, it has to remain alert to the sources of management risk, and the first and possibly most important is where and by whom are management decisions being made.  So often demanding and dangerous stakeholders are involved or included in the decision-making, and yet good practice means that only the decision-makers should be limited to those who have a legitimate right – which means the decisions are made at the right place by the right people.

A second, and in some ways, more subtle point about adapting governance to better suit changing circumstances was made by Cohn, an early Agile theorist.  He pointed out that project governance – far from eschewing change – should welcome it and see change as a positive consequence of having learned something and avoiding the mistake of doing something not wanted.  A far cry from the rigid, predictive governance strait jackets of yesteryear that saw the role of keeping to the script and frustrating change.

Agile governance

Agile projects

So in the Agile world, governance matters but it’s a much more consultative process.  It’s not just about whether the project is complex, but what the clients’ attitudes and desires are around the way the project should be conducted.  If the client is open to exploration and the requirements are difficult to define then the Agile space is a great place to be.  But if the client needs and demands predictability in the delivery then it probably isn’t.

Bringing those legitimate stakeholders much closer into the project – moving from a negotiative relationship to a collaborative relationship – is key to shortening decision making time.  Scrum practices such as the product owner is a good example of attempts to do this.  But, as the use of these practices increases, there is a very real danger that (as per the role of the project sponsor) the business will become project-weary and circumvent the Scrum mandate, allowing projects to run without a genuine product owner in place.  You may even know instances of that happening in your own organization right now!

Governance practices must diversify and become change competent

As project management disciplines and approaches extend into more diverse areas, as the product development processes projects encompass become more sophisticated, and the demands made by stakeholders increase, project governance must respond – it too has to diversify without losing its role of providing senior managers – the investors in projects – with the confidence they need to implement their organisation’s strategy.

About the Author

Christopher Worsley has over 40 years’ experience in project and programme management.  He is a visiting lecturer on the University of Cape town MSc in project management and is the author of the following books with his wife, Louise Worsley. 

Christopher will be talking on the importance of adaptive planning practices at the Project Management South Africa (PMSA) monthly meet-up on the 28thApril 2020. 

Adaptive project planning

The lost art of planning projects

print