Top Contributor Badge

This article is written by Jonathan Craft, founding director of Policy Ready.

  • The problem: Governments are risk averse and lessons from failures are not well shared.
  • Why it matters: Failure is an essential part of digital ways of working.
  • The solution: Failing smarter through improved sharing of lessons learned and working in the open will help advance digital government.

You have probably heard the term “fail fast”: Build actual things and test them in the real world so that inevitable failures lead to learning and improved iterations. Normalising this type of purposeful and controlled failure in government is no easy feat. It can be far more comfortable to strike another committee, write another briefing note, or consult with key stakeholders.

• Want to write for us? Take a look at Apolitical’s guide for contributors

Failing is only valuable, however, if it leads to learning. Failing smart requires lessons to be generated, internalised, and appropriately shared to avoid repeating the same mistakes.

We need smarter failing in the digital government space

A new two-part case study for the Policy Ready and the IPAC digital government series takes a hard look at a project that suffered repeated failures and the lessons it offers. The Application Rationalization (AR) project was launched in 2011 to modernise the various Public Service Commission of Canada’s (PSC) information technology infrastructure for its testing applications to ensure their continued viability and improve their functionality. There were real concerns the systems were not sustainable, limited the effectiveness of the PSC, and were costing too much to maintain.

For digital government to be successful, there will be lots of failures. Let’s work to ensure that they can be smart ones.

After a challenging start marked by poor project definition and scoping, the next decade involved successive budget overruns, staffing and capacity issues, and product development and delivery delays. Midway through the project, major staffing changes occurred with agile scrum methods and new project management techniques adopted. Interviews and document analysis revealed initial aims and objectives were ultimately watered down, deliverables simplified, and functionalities and features were ultimately abandoned.

The first case study highlights how crucial early project budgeting, scoping, and specification work is and the serious costs associated with starting projects without appropriately resourced and dedicated teams. It demonstrates the difficulties that arise when product development work is undertaken in outdated IT environments with teams working with inadequate tools.

Avoiding Faux-gile traps

The project also exemplifies what we term “faux-gile”. Where agile methods are only partially implemented and within a “waterfall” framework, which results in staff frustrations, frictions between teams and units, and poor outcomes, as neither approach is truly being applied. The PSC case helps shine a light on the hard work, resources, processes, and executive support required to move from buzzwords to applying agile approaches in a traditional public administration environment.

The case also unpacks several positive learnings including the ability of PSC staff to deliver and build operational applications and products despite significant resource and tool inadequacies, and with shifting project goals. It also demonstrates how executive leadership can reorganise, resource, and bring strategic coherence to a project and deliver results.

Part II of the PSC study will be released for open access soon. It focuses on the governance and performance management challenges that undermined the project but also ended up being key avenues to get the project to the finish line. More on that soon...

It takes courage to pull the curtains back and reveal where things went wrong and how course corrections were made. Kudos to the Public Service Commission for doing so with the hopes that their digital government journey will help others confront similar challenges. For digital government to be successful, there will be lots of failures. Let’s work to ensure that they can be smart ones. — Jonathan Craft

We want to hear what you think about this article. Pitch an article to us or send your comments to discussion@apolitical.co

(Picture credit: Unsplash)