Unlocking the potential of public-sector IT projects

Digital transformation has been an often elusive goal for many public-sector organizations. However, the COVID-19 pandemic has both underscored its importance and diminished many barriers to digitalization, creating new momentum for governments to invest in their digital futures. In many countries, public-sector IT transformations are now progressing swiftly and are supported by an unprecedented amount of government funding. Germany allocated more than €3 billion to its Online Access Act and the modernization of public registers, France plans to spend €1.5 billion on modernizing its public information systems, and the European Union has budgeted €143.4 billion for “single market, innovation, and digital” initiatives.¹Corona-Folgen bekämpfen, Wohlstand sichern, Zukunftsfähigkeit stärken, Ergebnis Koalitionsausschuss, June 3, 2020; “France Relance,” Ministère de l’Économie, des Finances et de la Relance, 2020; “Recovery plan for Europe,” European Commission, 2020.

The task of many public-sector organizations now is to ensure that their digital transformation efforts yield tangible, value-creating outcomes for the public. The success of these IT projects will be measured not only on the realization of their expected benefits but also on whether these benefits are delivered on time and within budget.

McKinsey and Oxford Global Projects have collaborated on an extensive research program to better understand the outcomes and success factors of both private- and public-sector IT projects. The early research, conducted in 2011–12, suggested that half of all large IT projects—defined as those with initial price tags exceeding $15 million—exceeded their budgets due to broad challenges in delivery.²For further details, see Michael Bloch, Sven Blumberg, and Jürgen Laartz, “Delivering large-scale IT projects on time, on budget, and on value,” McKinsey, October 1, 2012; and Alexander Budzier and Bent Flyvbjerg, “Why your IT project may be riskier than you think,” Harvard Business Review, September 2011.

In the decade since the original results were published, the database has grown to include additional IT projects. Using this expanded set of data, our latest analysis yields the following key insights:

• Only one in 200 projects delivered the intended benefits on time and within budget. On average, IT projects overall—regardless of size—exceeded their budgets by 75 percent, overran their schedules by 46 percent, and generated 39 percent less value than predicted.
• Public-sector IT projects are more likely to miss their marks compared with private-sector projects—and at a greater cost. More than 80 percent of public-sector IT projects overran their schedules, while only half of private-sector projects experienced delays. Cost overruns were nearly three times higher on average for public-sector organizations than for their private-sector peers.
• Across both the public and private sectors, four performance levers—strategy, technology, governance, and adoption—have an impact on the three metrics of IT project success: staying within budget, completing the project on time, and delivering the project’s intended benefits.

This article provides new, data-backed insights on how public-sector IT projects can improve their odds of delivering expected value on time and within budget. Below, we discuss the disparity in outcomes between private- and public-sector IT projects, analyze the impact of the four key levers of performance improvement, and present a set of clear and pointed questions to help public-sector organizations assess where they stand today and chart a course for more successful IT projects in the future.

The shortfalls of large-scale IT projects

Last year, McKinsey and Oxford Global Projects analyzed data from more than 6,000 public- and private-sector IT projects completed between 2001 and 2017 (see sidebar “About the research”). Across all three performance metrics—cost, timeline, and benefits—public-sector IT projects lagged behind their private-sector counterparts.

Only one in 200 IT projects delivers the intended benefits on time and within budget

Traditional measures of IT project success are delivering on budget, on schedule, and with the intended benefits. Measured against these individual factors, IT projects across sectors appear to perform relatively well: 59 percent of projects are completed within budget, 47 percent on time, and 44 percent with the intended benefits.³Our research compared the actual cost, actual time, and actual benefits against the estimates made for the final investment decision (also called the final business case). Costs include all costs from an owner’s perspective. Schedule is defined as the duration between the final investment decision and the go-live date. Benefits are measured as hard cashable benefits (for example, cost savings).

However, these figures obscure a less rosy picture. Projects that fail to meet their planned budget, time frame, or projected cashable benefits often do so by a considerable amount: on average, costs balloon by 75 percent, projects are delayed by 46 percent, and benefits shrink by 39 percent.

While most projects manage to stay within budget, the share of IT projects that deliver both on budget and on time falls to one in 14. And the share of projects that meet all three measures—budget, schedule, and benefits—drops to just one in 200.

Public-sector IT projects are more likely to exceed budgets and timelines

The data show a clear difference between public- and private-sector IT project risks in terms of cost and schedule overruns (Exhibit 1). Eight out of ten public-sector IT projects take longer than expected, compared with just more than half of projects in the private sector. Cost overruns occur in nearly one in two public-sector projects, as opposed to about one in three in the private sector. Notably, the average IT project cost overrun is nearly three times higher for public-sector organizations than for their private-sector peers.

1

Public-sector IT projects face a higher risk of exceeding initial cost and time estimates due to a number of factors, including a more complex stakeholder landscape; less effective risk management; difficulties in recruiting, retaining, and developing talent; slower governance processes; and less pressure to deliver minimum viable products.⁴Munmun Baishya, Ankur Ghia, Naufal Khan, and Gayatri Shenai, “Scaling digital in the public sector: Building blocks for the future,” McKinsey, November 2020; Tera Allas and Richard Dobbs, “Mitigating risk in IT megaprojects,” McKinsey, January 26, 2018; Anusha Dhasarathy, Mike Joyce, and Mark McMillan, “The next chapter: Driving technology leadership in the public sector,” McKinsey, February 17, 2021. The success rate of public-sector IT projects is thus extremely low, lagging significantly behind that of private-sector IT projects. Furthermore, virtually none of the public-sector IT projects in the database measured the benefits of the project after completion, highlighting a much larger governance challenge (see sidebar “Managing benefits targets in public-sector IT projects”).

Extreme cost overrun is higher for public-sector IT projects

IT-project planning is always uncertain. On average, public-sector projects take 3.9 years to complete and private-sector projects take 2.4 years. Planning four years into the future for public-sector projects can never be perfect, and a certain amount of cost overrun is expected and tolerated by many governments. However, extremely large cost overruns create significant consequences for both the organization and the people involved.

Indeed, the analysis shows that large cost overruns are more frequent in the public sector. Fourteen percent of public-sector projects exceeded their budget by at least 100 percent, compared with only 7 percent of private-sector projects. Additionally, 4 percent of public-sector projects exceeded their budget by 400 percent, while only 1 percent of private-sector projects did so.

These disparities may be explained in part by the differences in how private- and public-sector organizations deal with projects that are over budget. While most private-sector companies have procedures for terminating failing projects, this is not necessarily the case in the public sector. Our analysis suggests that many public-sector organizations lack explicit protocols with criteria outlining when to defund projects that are over budget.

Public-sector IT projects with longer timelines have the biggest average cost overruns

For public-sector projects, our research found that a project’s estimated duration is associated with an exponential increase in cost overruns (Exhibit 2). However, the opposite trend appears to be true in the private sector. Data on long-term IT projects in the private sector are sparse because the private sector tends to avoid these types of projects, but the analysis suggests that private-sector IT projects with long time frames may actually have lower budget overruns, though the correlation is not statistically significant.

2

One explanation for this difference may be the fundamental differences between the aims of the private and public sectors. In the private sector, organizations pursue primarily economic objectives, whereas public-sector projects also have policy goals. As policy objectives tend to be more volatile—especially in the case of a change in government—projects in the public sector face challenges in aligning stakeholders on strategy and prioritization, especially over long periods of time. Unfortunately, changing objectives often conflict with a project’s original goals and may alter previous efforts. Long-term projects in the public sector are therefore less likely to achieve their objectives. Additionally, public-sector organizations may have a culture that is more resistant to change, which can lead to a lower adoption of new IT technology and more expensive change management approaches.

Four levers to drive performance

Drawing on insights from practical experience and former research, we have identified four potential levers of IT project performance that apply regardless of sector: clear strategy, technology mastery, effective governance, and user adoption. When it comes to implementing these levers, there are no magic bullets or simple solutions. Dozens of factors contribute to the success of technology transformations, many of which are dependent on the objectives, the organization, and the context of the project.⁵Katya Defossez, Mark McMillan, and Hrishika Vuppala, “Managing large technology programs in the digital era,” McKinsey, November 25, 2020. In this article, we emphasize best practices for public-sector projects in particular.

Clearly define the project strategy

Strategy involves defining a clear project vision, strategic pillars, value drivers, goals, and metrics. A compelling project strategy helps orient the organization toward a successful IT-driven “North Star.” However, it is effective only when shared and aligned across all relevant stakeholders and used to generate long-term sponsorship and momentum. An IT project’s success requires a close partnership between technology and business leaders. For example, nearly three in four respondents at top-performing companies reported that their chief information officer or chief technology officer is highly involved in shaping the enterprise-wide business strategy and agenda—compared with only one in three respondents from all other companies.⁶“Prioritizing technology transformations to win,” McKinsey, March 24, 2022. Within these organizations, business and IT leaders are also more likely to cocreate the IT and digital strategies in tandem with the business strategy, which ensures alignment.

For private institutions, decision-making stakeholders are generally internal, while public institutions are obligated to consider public opinion.

Master the technology

Mastering the technology is vital to ensuring the success of IT projects. It requires high-quality customer input, clear design with as few unknowns as possible, clearly defined required platforms and architecture, robust performance management of technology partners, and the right balance between innovative and tested and proven solutions.

The most successful tech transformations focus on a higher number of initiatives and take aggressive action to adjust course when needed. While many organizations are unwilling or unable to make difficult decisions when problems arise (as they inevitably do), those that invest the time and resources to pinpoint the root cause and intervene as necessary are more likely to succeed—even if it means forfeiting sunk costs and starting over.⁷“Managing large technology programs in the digital era,” November 25, 2020.

The public sector may be at a disadvantage when it comes to the technology lever. To truly master technology and define a future state, an organization must understand the status quo of the existing IT architecture. The public sector tends to be more reliant on legacy architecture and infrastructure, which can pose a significant starting-point challenge for developing and implementing new technology.⁸“The next chapter,” February 2021. Replacing this dated infrastructure with more commercial off-the-shelf software may therefore help to derisk public-sector IT projects.

Implement effective governance

According to the results of our analysis, governance is the most effective lever. The data show a clear link between effective governance and reductions in cost and schedule overruns. This is not surprising because governance is the centerpiece of a successful transition from planning to execution. There are three cornerstones of effective governance:

• Stakeholder involvement. Effective governance works upward to involve stakeholders and sponsors and downward to steer project progress and course-correct when necessary.
• Talent planning and capability building. Access to the right talent on both the IT and business sides is crucial to achieving project success. People-focused initiatives create the most value for technology transformations. Conversely, people—rather than technological—issues (such as skill gaps, cultural differences, and talent sourcing issues) are the most commonly cited obstacles that prevent large-scale IT programs from realizing their objectives.⁹“Prioritizing technology transformations,” March 2022. Because hiring and assignment processes in the public sector can be complex and time-consuming, it is even more important to plan ahead and bring in people with relevant skills and experience as early as possible.
• User and expert involvement. Users and experts are most effective when they are engaged in an efficient and targeted manner. Rather than engaging experts continuously, successful project leaders plan ahead to determine which critical stages benefit from intensive expert involvement.

Drive user adoption

User adoption is critical to the successful execution of an IT project. In the public sector, driving adoption often involves complex bureaucratic and long-established processes, whereas the private sector is generally less constrained by historical precedent. To increase the adoption of IT solutions and processes, public institutions can start by understanding the operational requirements—which calls for the involvement of business departments early on—and then introducing organizational change. Best practices for change management include communicating with transparency, transferring ownership of the project to the business, involving frontline management, cocreating and delivering on-the-job training, and building trust with the employees whose work will change in the new operating model.

Critical factors for success

While a few organizations have already invested heavily in effectively managing their IT projects, many are still in the early stages of such efforts. At critical inflection points, public-sector organizations may benefit from honestly assessing their IT project work against key success metrics and the four performance improvement levers (see sidebar “A diagnostic framework for public-sector IT projects”).

As IT transformations gain momentum, public-sector organizations can leverage four performance levers to improve their odds of success. Understanding the impact of these levers will allow public-sector organizations to set more realistic goals for budgets, schedules, and benefits and to unlock the full value of their IT projects.