Recent much publicised IT project debacles in the public sector have given rise to debate on how expensive failures can be avoided in future. According to Professor Joe Peppard, academic director of UCD Smurfit Executive Development, the boards of those organisations often have the odds stacked against them when it comes to oversight of large technology investments. Indeed, it is often the case that they just don’t know what they don’t know.
Peppard has been considering this topic for some time and has co-written a paper on it for publication in the California Management Review later this year. He is also programme director of the Smurfit Executive Development Diploma in Digital Innovation & Transformation as well as of the Diploma in Artificial Intelligence & Digital Analytics.
Part of the problem is the language used to describe IT projects; he points out. “There are difficulties with the vocabulary and there is inconsistency in the words used and the meanings ascribed to them. Consider something as simple as a database. We now have data lakes, data warehouses, data sets and more. That causes confusion. Why are we using so many different labels? We talk about platforms, but what is a platform?”
He notes the sharp contrast with finance where consistency is key. “There are lots of globally accepted definitions and international standards to ensure that everyone is talking about the same thing,” he notes. “We don’t have a common language for technology.”
READ MORE
The abstract nature of technology projects is another issue. “An IT project is not like a construction project,” he points out. “If a public body is building a new wing for its headquarters they get a two-dimensional visual representation in the plans, they can also get a three-dimensional representation through a physical model, and it is possible to use virtual and augmented reality to do a virtual walk through of the proposed building. With an IT project there’s nothing to see.”
Those language issues and their abstract nature can make IT projects quite amorphous. “I can appreciate the challenges boards of directors face,” says Peppard. “If a project manager comes to the board and says the project is halfway through, there is nothing to see. At least on a construction site there is visible progress, there is a physical reference point.”
This means that boards need to work with advisers and leadership teams to create a shared mental model of what the project is all about from the very outset, he advises.
The next step is to devise a reporting and verification mechanism. “We don’t have the same reporting mechanism in technology projects as we do for finance,” he notes. “There is no profit and loss account, no balance sheet, no cash flow statement. We don’t have the equivalent in IT and that’s a weakness. Also, it’s not a case of the board saying to the CEO and CFO that they don’t trust them, but they still have external auditors to verify that the accounts are true and accurate.”
In the case of IT projects, the reports presented to the board are often coloured by misplaced optimism. Traffic light systems tend to be used to describe process with lots of greens, very few oranges, and almost no red at all in evidence. That is not necessarily an attempt to mislead, rather a natural bias towards positivity.
If the culture is one of only reporting progress, and if the leadership team is not very tech savvy, they will assume the project is on track and report that to the board. The board will also assume everything is alight, but they have no equivalent of audits to verify that assumption.
‘The board and leadership team have to agree upfront what is going to be reported on and how. And they have to be willing to pivot where necessary’
In these circumstances, boards need to become more knowledgeable in relation to technology. “If you look at training for boards, the Smurfit Diploma in Corporate Governance last year for the first time introduced a module on the governance of technology,” says Peppard. “No other board training programme does that at present. Boards have to be familiar with directives and regulations like NIS2, DORA and GDPR. Accountability for compliance resides with the board. Boards also need to be about to stand over the operational resilience of the organisation. I don’t know of any company today that could exist without IT its systems; businesses are fundamentally dependent on them.”
Ultimately, it comes back to the common mental model and a willingness to call a halt when necessary. “It’s about everyone seeing the project in the same way,” he explains. “Where they have the same picture of what the finished project is going to look like and what it’s going to achieve. Everyone needs to understand the benefits and business outcomes and how they are going to be achieved.”
Boards also have to ask if the organisation is ready to undertake the project, he adds. “Have we done this before? Is the organisation set up to succeed? There also has to be an agreed reporting scorecard for the project. The board and leadership team have to agree upfront what is going to be reported on and how. And they have to be willing to pivot where necessary.”
He advises boards to be wary of the sunk cost fallacy where continued expenditure is justified by past investment in the project. This is also known as throwing good money after bad.
While success cannot be guaranteed, the odds against it can be reduced by arriving at a shared mental model, employing clear, consistent language to describe the project, ensuring that the organisation is ready for the project, and having clear sight of how its outcomes and benefits are going to be achieved.
Agreement on a reporting scorecard is also required. “This is a difficult area but there is a need to have the same clarity as we have with finance,” he says. “That’s a challenge but one I’m working on at the moment.”
Finally, boards must be willing to hit the off switch before costs run completely out of control.
For more information on taking a course at UCD Michael Smurfit Graduate Business School, visit smurfitschool.ie
