Contents
Quebec’s digital shift is in full swing: the Gallant report has just recommended creating a “centralized entity specialized in the digital transformation of the Quebec state” (opens in a new tab) (in French).
Yet, as budgets explode, large public digital transformation projects are increasingly synonymous with failure.
SAAQclic, for example, was supposed to make life easier for drivers.
Instead, the project has cost $955 million (opens in a new tab) to date, more than double its initial budget of $458 million, while causing massive service disruptions.
SAAQclic, Santé Québec, Phoenix: rightly or wrongly, these names have become media shorthand for “derailment,” “chaos” and “confusion.”
Who is to blame?
Are these public failures caused by technical breakdowns?
Were they simply too ambitious technologically?
Across all these projects, the same patterns repeat: spectacular announcements, budgets in the billions, tight timelines… then commissions of inquiry and damning audits.
The repetition of the same mistakes across several different projects suggests that the fundamental problem isn’t technological, but lies in the way these projects are governed.
The report on Phoenix (opens in a new tab), for instance, points less to software defects than to governance gaps: risks played down, experts sidelined and a decision to deploy a system that wasn’t ready.
In the spirit of Turbulences, where the aim is less to react in the moment than to understand in depth, I wanted to take the time to think through these issues collectively since last December.
To that end, I reached out to Antoine Desjardins (opens in a new tab), Chief Operating Officer at Tink (opens in a new tab), a leader I respect for his rare ability to combine strategic vision and concrete execution in complex digital environments.
Together, we compared our perspectives over several conversations, enriched by discussions with various industry experts and a rigorous review of the literature on the governance of large IT projects, here and elsewhere.
What follows is therefore both analytical and experiential.
It draws on empirical research, particularly in large-project management, but it is also shaped by our combined experience in the digital industry, with public and private organizations.
It doesn’t claim to offer a miracle solution.
Rather, it proposes a framework for moving beyond a simplistic reading of “technology fiascos” and shifting the debate to what we see as the real issue: the maturity of digital governance and leadership.
From solution to mission: realigning projects with their purpose
In his book [How Big Things Get Done](https://www.amazon.ca/How-Big-Things-Get-Done/dp/077109843X), Bent Flyvbjerg quotes Frank Gehry, one of the most rigorous architects in the world, who systematically starts his projects with a simple question: “Why are you doing this project?”
And as long as the client can’t answer clearly, the architectural work doesn’t move forward.
Public digital services would benefit from adopting this discipline.
Without a why that is shared and understood by everyone, megaprojects are built on a strategic misunderstanding: digital is treated as a solution in itself, rather than as an instrument serving a mission.
The challenge, then, is not only to choose our technologies better, but to invest more in defining the issues.
Why is this discipline so hard to establish?
Because political pressure naturally pushes toward visible action: announcing a platform, a portal, a solution.
Rarely toward patiently defining a problem.
To get at the root of this dysfunction, government technology expert Waldo Jaquith (opens in a new tab) offers an illuminating analogy: the “peanut butter sandwich.”
If a child asks an adult for a sandwich, they will get it without a hitch.
But if the government tries to order that same sandwich, it will write a three-hundred-page call for tenders describing the viscosity of the peanut butter and the angle at which the bread is cut, without even having the in-house expertise needed to write those three hundred pages.
The result?
A supplier who follows the instructions to the letter, charges a fortune, and ends up delivering peanut butter spread directly on the bag of bread.
The contract is honoured, but the result is inedible.
This metaphor captures the heart of the problem: we try to manage digital complexity through contractual bureaucracy.
We pile up technical requirements (the “how”) before we have even crystallized the mission (the “why”).
Yet this is precisely where fiascos are prevented: as long as the why isn’t stable, everything else becomes unstable.
The challenge, then, is not only to choose our technologies better, but to invest more in defining the issues.
Ultimately, in digital, strategic precision isn’t a luxury; it’s insurance against drift.
The miracle software syndrome
When the quest for the “Next Shiny Object” fragments the digital state
If digital projects often start badly because their purpose is fuzzy, they then derail because they rush toward a miracle solution.
In public and private organizations alike, there is an almost irresistible temptation: to believe that a new tool can fix structural flaws.
The reflex is appealing; human, even.
When a system ages, we imagine that replacing it with a “modern” solution will let us turn the page.
The problem is that this logic rarely produces transformations (opens in a new tab).
Mostly, it produces replacements.
This is the myth of the Next Shiny Object (opens in a new tab): the belief that a new tool can make up for a lack of vision, architecture or internal coherence.
The phenomenon is made worse, among other things, by public funding mechanisms that encourage what some call “premature solutioning (opens in a new tab).”
To secure funding, you must submit a business case centred on… the technology solution.
That is precisely the trap of miracle software: it promises a fresh start, but amplifies existing challenges instead.
The uniqueness bias
A quick aside: Flyvbjerg’s research, mentioned earlier, helps explain why this illusion persists.
A central bias, which he calls uniqueness, leads leaders to believe that their project is exceptional, that it escapes the statistics and deserves a special solution.
This belief fuels the appeal of custom-built solutions.
Indeed, the Gallant report (opens in a new tab) describes how the SAAQ sought to innovate by using untried processes; yet “testing so many new concepts within a single megaproject could only substantially raise its level of risk,” the commissioner writes.
The empirical literature shows that it is precisely custom-built projects that blow up: they combine technical risk, dependence on suppliers and the impossibility of learning from previous implementations.
Standardized projects, by contrast, make it possible to align business processes with proven practices rather than trying to bend software to an organizational logic that is sometimes obsolete.
To get out of this dead end, the question to ask is no longer: “Which software should we choose?”
It is: “Have we clarified the problem enough to determine whether software is needed at all?”
As long as organizations look for a single, brilliant solution, they will keep stacking tools rather than building a coherent architecture.
Forgotten people: transforming without your real experts is a recipe for failure
The value of insiders: understand before you transform
The Gallant report finds that the Quebec state is excessively dependent on external consultants on the technical side, and recommends remedying “the lack of government expertise” (opens in a new tab) as quickly as possible.
Research confirms that successful transformations don’t rest on external consultants, but on internal employees who intimately understand real interactions with citizens.
A telling example comes from Santa Clara County (opens in a new tab).
After years of failure with consultants imposing standardized solutions, the organization entrusted the overhaul of its permit process to internal employees.
The result: a 33% reduction in processing time, not thanks to a new tool, but thanks to a fine-grained understanding of citizens’ journeys.
“An external partner can bring method and experience, but it will never replace the organizational memory of 20-plus years of system evolution, often poorly documented. When you sideline internal experts, you deprive yourself of an essential ingredient of the solution,” says Antoine Desjardins (opens in a new tab).
In public megaprojects, this mistake is structural.
Correcting it should be a priority.
Replacing the false metric of “on time, on budget”
Delivering on schedule does nothing to guarantee value creation
In large digital transformation projects, staying on budget and on schedule often serves as the main performance indicator.
Yet a project can be delivered “by the book” while completely failing to create value.
The Phoenix system is the most striking example: rolled out in line with the official milestones, it nonetheless had more than a hundred known defects and triggered a national pay crisis.
The schedule was optimized rather than the quality, sacrificing the experience of employees and citizens alike.
Value creation as the new compass?
To break out of the flawed “on time, on budget” paradigm, organizations need to measure the value created rather than compliance.
More relevant indicators exist:
- User adoption (rate of active users, frequency, etc.)
- User experience (NPS, employee satisfaction, reduced friction, etc.)
- Smooth journeys and reduced friction
- Impact on business indicators
- Reduction of operational debt
- Error rates (number of major incidents, recurring fixes, etc.)
- Time to First Value (time needed before the first tangible benefits can be observed)
However, these indicators only have their full value if project leaders have the courage to truly pursue them.
“Most transformations fail because they measure what is easy, not what is essential,” Antoine Desjardins (opens in a new tab) reminds us.
Do we have the courage to delay a deliverable to protect the vision?
Even with the right objectives, digital transformation projects are often rooted in “excessive optimism” about the schedule and scope of the project.
Leaders tend to see technology as a magic wand, overlooking the complexity of execution.
Holding on to an unrealistic schedule to save face invariably leads to what experts call project “distress”: adverse conditions that can no longer be resolved through routine and that require immediate, agile intervention.
Mature digital leadership is measured by its ability to say no: no to a premature rollout, no to an unrealistic schedule, no to a solution that betrays the original vision.
This maturity requires breaking with the culture of illusion.
In many fiascos, leaders prefer to keep up a façade of compliance rather than admit a budget blowout or a technical lag.
This lack of transparency creates what experts call “project distress,” where problems pile up without being solvable through routine methods.
Managerial courage here means revealing the real state of progress, even if it disappoints the board’s immediate expectations, to secure the organization’s long-term future.
In short, delaying a deliverable to protect the vision is the most responsible management decision a leader can make.
It takes the humility to revisit one’s assumptions and the strength to pivot when performance indicators, whether financial or human, signal that things are slipping.
From promise to proof: rebuilding trust in public digital services
Public megaprojects don’t derail because digital is “too complex” or technologies move too fast for the state.
They derail mainly because we still apply a management model inherited from an era when we bought systems, without recognizing that digital projects are in fact deep transformations of the way the state decides, learns and serves citizens.
Past fiascos, like Phoenix and SAAQclic, are not isolated accidents, but symptoms of a cultural and organizational model ill-suited to today’s digital ambitions.
One constant emerges: megaprojects don’t fail at the end, they fail at the beginning.
They fail when the conceptual, human and organizational foundations are neglected.
The levers for turning things around are now well identified:
- Clarify the mission before choosing the solution;
- Mobilize internal talent;
- Measure value in use rather than contractual compliance;
- Develop the ability to say “no” to a premature rollout to protect real adoption.
Trust in public digital services won’t be rebuilt by decree or by a new “big platform,” but by adopting a new daily discipline: that of proof, patient, incremental and deeply rooted in citizens’ reality.
One more thing before you go…
This article is part of my Turbulences series (in French).
If you enjoy unfiltered reflections and bold explorations of the digital trends shaking up our world, from marketing to AI to the environment… then Turbulences is for you.
A newsletter designed as a spontaneous conversation, where ideas take flight before they are perfectly polished.
Sources
- Tabrizi, B., Lam, E., Girard, K., & Irvin, V. (2019). Digital transformation is not about technology. Harvard Business Review.
- Joukhadar, G., Jiang, R., Harrington, K., & Thorogood, A. (2023). Promoting digital innovation for sustainability in the public sector. Communications of the Association for Information Systems, 53, 240-277. https://doi.org/10.17705/1CAIS.05310 (opens in a new tab)
- Pombriant, D. (2021). Do you have the right software for your digital transformation? Harvard Business Review.
- Andriole, S. J. (2017). Five myths about digital transformation. MIT Sloan Management Review, 58(3), 20-22.
- Flyvbjerg, B., & Gardner, D. (2023). How big things get done: The surprising factors that determine the fate of every project, from home renovations to space exploration and everything in between.
- Commission d’enquête sur la gestion de la modernisation des systèmes informatiques de la Société de l’assurance automobile du Québec. (2026). Sommaire : Rapport et recommandations. Gouvernement du Québec.
- Goss Gilroy Inc. (2017). Lessons learned from the transformation of pay administration initiative. Gouvernement du Canada. Lien (opens in a new tab)
- Mercure, P. (2025). Projets numériques : Fais-moi un sandwich beurre d’arachide-confiture. La Presse.