How Do You Eliminate Project Uncertainties?

The steering committee meets for the third time. The choice between the Montreal integrator and the Quebec City one is postponed again: everyone is waiting for one last client reference. Meanwhile, the team can neither plan the migration nor book the experts. Eliminating project uncertainties starts with admitting that waiting costs more than deciding.

The figures cited come from public studies by McKinsey and PMI; they cover large projects and serve as reference points, not as a forecast for your project.

According to McKinsey, in collaboration with the University of Oxford, large IT projects run 45% over budget and 7% over time on average, while delivering 56% less value than predicted.

Why are project uncertainties inseparable from risks?

Because a risk is, according to the PMBOK definition cited by PMI, an uncertain event or condition that, if it occurs, affects project objectives: without uncertainty there is no risk, only problems. The process of identifying uncertainties therefore goes hand in hand with risk management.

Managing risks and reducing uncertainties demands all of a project manager's skill. They must identify the possible risks, put in place actions that give better control during execution, and keep the objectives in sight. Almost all of those actions boil down to one thing: making decisions.

Uncertainty in project management is unavoidable. The project manager's job is not to make it vanish, but to validate and clarify assumptions one by one, so the project becomes clearer and simpler for the teams involved.

Where do uncertainties hide in an IT project?

In every facet of the project, but above all in the delivery process: the approach, the phasing, the proof of concept, deliverable approval, procurement and the use of external partners. Each of these dimensions is a question nobody has answered yet at kickoff.

The concrete questions: do we proceed in agile mode or in planned iterations? Do we need several phases? What kind of proof of concept do we run, and on what scope? Who approves deliverables, and against which criteria? How do we handle procurement? Do we call on external partners? These questions form the project manager's arsenal. They must be answered quickly and precisely.

The cost of waiting has been measured. Reference points from the McKinsey and Oxford study of more than 5,400 IT projects:

• 45%: average budget overrun of large IT projects (initial budget above US$15 million).
• 7%: average schedule overrun of those same projects.
• 56%: shortfall in value delivered compared with predicted benefits.
• Half of the large IT projects analyzed massively blew their budgets, with software projects the most exposed.

Why does every decision create new uncertainties?

Because choosing a solution removes the uncertainty of having to choose, but immediately activates doubt about whether the choice was right: if the doubt persists, new decisions and actions are required. That is normal. Some uncertainties are never fully resolved, since every decision opens the door to unpredictable situations.

Take an analogy. To travel from Montreal to Quebec City, you first choose the vehicle. Then the route: the South Shore or the North Shore, the old Quebec Bridge or the Pierre-Laporte Bridge. Each choice closes one question and opens another. But without the initial decisions, the trip never happens.

In a car, the means of managing risk are standardized: seatbelt, insurance, weather forecast. In project management, they vary from case to case, and every situation is more complex. Paired with risk management, reducing uncertainties moves the project forward with foresight instead of leaving it in limbo.

Isn't the biggest risk not deciding at all?

Yes: not making a decision is neither a safety net nor an option, it is a risk in itself, because it keeps every uncertainty open and delays every subsequent decision. A good project manager therefore speeds up the decision process while planning ways to reduce uncertainty. Otherwise, they will end up managing surprises with no preparation.

Postponing a decision does not reduce the risk, it only changes who owns it: the team carries it in your place.

The steering committee that has been waiting for one last client reference for three meetings is the perfect example. While it waits, the project cost estimate rests on assumptions nobody can validate, and every week of drift ends up in the budget.

This is precisely what a third-party project health check reveals: the list of pending decisions, their age and what they block. In the projects we support, that list explains more delays than all the technical problems combined.

How do you concretely reduce project uncertainties?

Keep an assumptions log, give every open decision an owner and a deadline, test critical assumptions with a targeted proof of concept, and make the decision review a fixed item of every steering committee. Four simple moves, rarely done together.

The assumptions log is the most neglected. Every assumption in the IT project plan, about team availability, data quality or a vendor's delivery date, is an uncertainty disguised as a certainty. Naming it lets you validate it or replace it with a decision.

PMI's Pulse of the Profession 2025 report, based on more than 3,000 project professionals, stresses business acumen: understanding the context, knowing how to navigate the organization and making things happen. In other words, the ability to get decisions made, not merely to document them.

By moving forward with a chosen solution, the next steps become easier to visualize and define. It then becomes simpler to make the following decisions and shrink the uncertainties. It is a virtuous circle, provided you dare to take the first turn.

Can you really eliminate project uncertainties?

No, but you can replace them one by one with dated, owned and revisable decisions, which is exactly what is expected of a project manager. A clear project is not a project without unknowns. It is a project where the team knows which ones remain open, who must close them and by when. The rest is called risk management, and it always starts with a decision.

Frequently asked questions

What is the difference between a project uncertainty and a project risk?

An uncertainty is a question without an answer: which approach, which vendor, which scope. A risk is an uncertain event that, if it occurs, has a positive or negative effect on project objectives, according to the PMBOK definition cited by PMI. Every unresolved uncertainty feeds one or more risks; every decision closes some and opens others.

Can you eliminate all project uncertainties?

No. Uncertainty is inherent to every project, and each decision creates new ones. The realistic goal is to reduce it progressively: validate assumptions, settle open questions on time and document what remains unknown. A project where the team knows what it does not know is already better managed than one that claims to know everything.

Why do large IT projects overrun their budgets so badly?

According to McKinsey and the University of Oxford, who analyzed more than 5,400 IT projects, large projects run 45% over budget on average and deliver 56% less value than predicted. The causes often trace back to decisions made too late: fuzzy scope, assumptions never validated and technology choices postponed until the moment they cost the most.

How do you speed up project decisions without rushing them?

Give every decision an owner, a deadline and the minimum information needed to decide. A targeted proof of concept lasting a few weeks beats six months of analysis. Set a steering committee rule as well: a decision postponed twice gets escalated. Speed comes from a clear process, not from haste.