When a modernization project fails, technology is often the first suspected cause: an inadequate architecture, insufficient migration tools, or an integration that is more complex than anticipated. While these factors can contribute to the challenges encountered, they are rarely the primary cause.
This finding closely reflects what we see in the field across major transformation programs: the most significant challenges rarely stem from the architecture, tools, or technology selected. Instead, they emerge when decisions are delayed, responsibilities lack clarity, dependencies are not managed at the right level, and governance can no longer support the project’s actual pace. Strong governance enables teams to absorb uncertainty, make trade-offs quickly, and maintain a consistent course. When governance weakens, however, even capable teams and sound technical solutions operate within a framework that is too fragile to support execution at scale.
Large migration and modernization projects do not go off track primarily because of technology. They go off track when governance can no longer support the program’s true complexity.
After more than 25 years spent leading, structuring, or supporting complex technology initiatives, one constant stands out: major challenges arise less from the tools than from how decisions are made, how information flows, and how responsibilities are carried out.
Published findings point to the same conclusion. A recent article published on arXiv by a senior architect with 24 years of experience delivering modernization projects argues that the primary failure mode is not technical, but rather the result of a governance gap. Gartner confirms that 83% of data migration projects exceed their budgets or fail. A 2026 report on the causes of project failure identifies the same root causes: vague scope, sponsor drift, delayed approvals, artificially positive status reports, and teams forced to execute before the operating model has been stabilized.
What is governance?
The word “governance” first refers to the idea of leading, providing direction, and staying the course. It did not originally describe a bureaucratic structure, but rather the act of guiding a group in a consistent direction. This is precisely the meaning we need to restore in major migration and modernization programs. Governance is not a monitoring mechanism. It is an organization’s ability to manage complexity, make information reliable, make trade-offs under pressure, and maintain a strong connection between decisions and execution.
In a migration and modernization project, this ability takes shape through four conditions for maintaining control. They determine whether the organization absorbs uncertainty or is overwhelmed by it.
4 essential governance requirements that maintain control
- Clear decision rights: Who decides, who makes trade-offs, and who removes roadblocks when a technical, functional, or organizational obstacle exceeds the delivery team’s authority? In projects that go off track, these answers remain unclear. Decisions pile up, escalations go in circles, and teams wait while everyone assumes someone else has the authority to decide.
- A reliable view of the project’s reality: Beyond the percentages displayed on a dashboard, we need to see what is actually happening on the ground. Which applications are truly moving forward. Which dependencies are causing delays. Which risks have materialized. Which decisions have been deferred. This visibility depends directly on the quality of communication. When teams are afraid to say that a milestone is slipping, a risk has materialized, or an assumption is no longer valid; information moves upward in a filtered, reassuring, or incomplete form. This is how “watermelon” projects emerge: green on the outside and red on the inside. The danger lies not only in the gap between appearances and reality. It also lies in the fact that decision-makers continue to make trade-offs based on degraded information and ultimately make decisions that are consistent with a false picture of the project.
- Straight talk built on trust: Transparency is not optional. It is a condition of effective governance. If teams do not feel comfortable saying that a milestone is slipping, an assumption was wrong, or a risk is becoming critical, the decision-making structure immediately begins to weaken. Problems do not disappear because no one talks about them. They simply take a different form and surface too late, when they cost more and leave fewer options. The strongest projects are those in which bad news travels early enough to be addressed, because a climate of trust allows people to speak clearly without unnecessary fear.
- The ability to make decisions at the program’s pace: A migration project involves hundreds of small decisions. Every delay leaves a team waiting, pushes back a schedule, and increases risk. Effective governance does not slow the project down. It creates conditions for making timely decisions at the right level, based on reliable information and a clear understanding of the consequences.
Tiered governance: operational, tactical, and executive
Large-scale migration and modernization projects involve dozens, sometimes hundreds, of people across the client, the primary vendor, and potentially other technology partners. Governing this entire group through a single weekly status committee meeting is not enough.
Experience shows that effective governance operates across three complementary levels.
- The operational level manages day-to-day execution: scrums, sprint reviews, and technical blocker tracking.
- The tactical level coordinates teams, monitors dependencies, and resolves competing priorities.
- The executive level maintains strategic alignment, oversees investment decisions, and handles escalations that lower levels cannot resolve.
In a large-scale migration project that Alithya completed in the insurance sector, this multilevel governance model, supported by the active involvement of client and vendor executives, helped remove roadblocks quickly, align teams, and maintain a realistic, shared, and sustainable course. Structured application-level tracking strengthened the management of dependencies, access, validation, and production deployments while providing continuous transparency into progress, risks, and issues.
The human factor: what no methodology can resolve on its own
Methodologies such as Information Technology Infrastructure Library (ITIL), Control Objectives for Information and Related Technologies (COBIT), The Open Group Architecture Framework (TOGAF), and Scaled Agile Framework (SAFe) provide a useful structure. None can guarantee effective governance on its own. The real challenge remains human: the quality of management practices, transparency in communication, the ability to escalate at the right time, and the courage to address problems before they become structural.
A steering committee that exists on paper, but where no one dares to challenge the sponsor. A risk register that is kept up to date, but that no one reads. An escalation process that is clearly defined, but that some people are reluctant or afraid to use because they do not want to bring a problem to management. These situations are common. Another methodology will not correct them. They require a culture of transparency, accountability, and managerial courage.
Successful projects are those where bad news travels as quickly as good news. Where a project manager can say, “We are behind schedule. Here is why, and here is the plan we need to discuss,” without fear of retaliation. Where decision-makers prefer an uncomfortable truth to a reassuring dashboard.
Visible problems are solvable problems
Well-established governance does not make problems disappear. It does something better: it brings them forward earlier, at the right level, and with enough clarity to support real decisions. This is often the price of success. The stronger the governance, the more likely it is that difficult issues, complex decisions, and real tensions will reach the people responsible for resolving them. Executives must therefore understand that learning early that a project is veering off course, a risk has materialized, or an assumption has collapsed is not a governance failure. On the contrary, it is one of the clearest signs that governance is working. A program is not endangered by the existence of a problem. It is endangered when that problem remains invisible at the point when a decision should already have been made.
Sources and references
- Harveen Punihani (2026). EMRGF: A Practitioner Framework for Governance-Driven Enterprise Technology Modernization, arXiv. This article argues that the primary failure mode in modernization programs is a governance gap, not a lack of technical capability.
- Gartner (source cited in recent industry analyses). The statistic that approximately 83% of data migration projects fail or exceed their budgets and schedules is widely cited in industry publications on data transformations and migration programs. It is used here to illustrate the scale of the risk associated with poorly prepared migrations.
- APMIC (2026). 2026-27 Report on Project Failure Rates & Root Causes: Original Data & Analysis. The report highlights recurring causes of failure, including vague scope, sponsor drift, delayed approvals, artificially positive status reports, and governance that cannot support actual delivery.
- Cultivated Management (2024). Watermelon Reporting: When Project Status Hides the Truth. This publication describes “watermelon reporting,” where dashboards show projects as healthy even though their operational reality has already deteriorated. It supports the discussion of transparency, information quality, and the risk of making decisions based on an inaccurate view of what is happening on the ground.
- Alithya. Internal references and lessons learned from large-scale migration programs, particularly in the insurance sector, covering multilevel governance, application-level management, client-partner coordination, dependency management, and operational transparency.