Why So Many Well-Run Projects Still Fail to Deliver Value

A project can look healthy, right up until the moment everyone realizes it has failed.

The schedule is green. The budget is under control. The team is closing action items and producing exactly what the plan called for.

Then the product launches... and customers don't use it.

The new process goes live... and employees quietly work around it.

The facility opens... but operating costs erase the expected savings.

The technology performs exactly as designed... and solves the wrong problem.

None of these are failures of execution. They are failures of clarity. The team delivered something. It just wasn't the outcome the organization actually needed. That distinction matters more than ever.

The uncomfortable evidence

A July 1, 2026 article from the Project Management Institute, "What 22,000 Projects Reveal About Success and Failure," summarizes a conversation between PMI President and CEO Pierre Le Manh and Alexander Budzier, a Fellow in Management Practice at Oxford's Saïd Business School.

Budzier's research dataset spans more than 22,000 capital investment projects across 126 countries. Roughly half delivered on budget or better. Only 8 percent delivered on both budget and schedule. Once the promised benefits were factored in, the success rate dropped to about 0.5 percentone, project in every 200 to 250.

That figure shouldn't be applied indiscriminately to every internal initiative or product launch; the dataset skews heavily toward capital investment and major-program environments. But the underlying lesson travels well beyond it.

Completing a project is not the same as creating a successful outcome.

A project can be managed efficiently and still produce little value. It can also blow through its original schedule and budget and end up creating extraordinary value. Scope, schedule, and cost still matter, but they're just not necessarily sufficient on their own.

PMI's broader research backs this up. Its 2024 Maximizing Project Success study, based on a survey of 10,000 project professionals and 150 in-depth interviews defines success as a combination of execution and value. Seventy-four percent of respondents tied success to finishing on time, on budget; and with a valuable outcome. PMI's conclusion is that organizations improve outcomes by strengthening three disciplines: planning more intelligently, building enough clarity to adapt deliberately, and defining success before execution begins.

Better planning doesn't mean predicting everything

Planning is sometimes framed as the opposite of agility. It isn't. Planning is how a team makes its current understanding visible, what the organization is trying to accomplish, what has to be true for the approach to work, where the real uncertainty sits, which decisions are still open, and how the team will know when circumstances have shifted.

A weak plan produces a date. A strong plan produces understanding.

The PMI article reports that most projects spend roughly 3 to 5 percent of total cost on planning, while the strongest performers in Budzier's research invested 20 to 30 percent. His point isn't that every project should mechanically hit the same percentage, it's that organizations routinely create false urgency by rushing through the exact window when decisions are cheapest to get right.

Infrastructure data points the same direction. The UK's former Infrastructure and Projects Authority found that projects with strong front-end planning ran about 20 percent less costly and 10 to 15 percent faster than average. Early planning lets teams catch design changes while those changes are still cheap to make.

None of this is an argument for months of analysis, excessive documentation, or waiting until every uncertainty disappears. That moment never arrives. It's an argument for doing enough thinking upfront to tell the difference between a calculated decision and a hopeful assumption.

Before major execution begins, a team should be able to explain:

• the problem that needs to be solved

• the outcome the organization expects (and needs)

• who must use, support, or benefit from the result (and seeing this through the lens of their experience)

• the assumptions the plan depends on

• the most consequential risks and dependencies

• the evidence behind the estimated cost and duration

• the conditions that would require the plan to change

Planning should make uncertainty easier to discuss not easier to hide.

A good plan makes adaptation possible

Projects don't succeed because the original plan was perfect. They succeed because the team can recognize when the original plan is no longer adequate. That takes clarity.

Without it, a change in direction looks like failure. People protect old commitments, defend outdated assumptions, and keep reporting against targets that no longer reflect reality.

With it, adaptation becomes a deliberate management decision. The team can ask:

What has changed? What have we learned? Which assumptions no longer hold? Does the expected outcome still justify the remaining investment? What should we preserve, stop, accelerate, or redesign?

This holds across methodologies. A predictive project shouldn't treat its approved baseline as a contract with the future. An agile project shouldn't use iteration as an excuse to start without a clear problem, intended outcome, economic rationale, or decision framework. A hybrid project shouldn't become a poorly integrated compromise; some activities rigid, others improvised, nobody clear on how decisions actually get made.

PMI's 2024 research found agile, hybrid, and predictive approaches can all produce similar performance results. Success depends less on loyalty to a methodology and more on choosing an approach that fits the work and giving the team room to use it well. Methodology matters. Context matters more.

A construction project, a software implementation, an organizational transformation, and a community program won't need identical planning practices. But every one of them needs a shared view of the outcome, transparent assumptions, real feedback loops, clear decision authority, and a disciplined way to respond to new information.

The goal isn't to prevent change. It's to prevent drift change that happens without a conscious decision about what it does to value, risk, cost, schedule, or stakeholder expectations.

Define success before you measure progress

Many projects launch with an approved solution and an incomplete definition of success.

"Install the system." "Open the facility." "Redesign the process." "Launch the program." "Complete the transformation."

These are outputs. They don't explain why the work matters.

A stronger definition of success answers at least four questions:

• What measurable condition should improve? Cycle time, customer experience, revenue, cost, safety, quality, access, compliance, employee capability, mission performance, pick the one that actually matters here.

• Who has to experience the benefit? A technically sound solution that users reject rarely produces lasting value.

• What constraints matter? Schedule, affordability, quality, risk exposure, regulatory obligations, and operational disruption still require disciplined management.

• When will success actually be evaluated? Some benefits show up at delivery. Others don't become visible until months or years after the project team has moved on.

This doesn't abandon the traditional project-management (iron) triangle, it puts it in context. Cost and schedule tell you whether the organization delivered efficiently. Benefits and outcomes tell you whether it delivered wisely.

PMI's research encourages project professionals to stay accountable for value, reassess project parameters as needs and technologies shift, and continually realign stakeholders around measurable outcomes. That accountability shouldn't end at handover. Someone has to stay responsible for adoption, operational integration, benefits measurement, and corrective action after the project's outputs land in the organization. Otherwise, the project gets declared complete exactly when the hard work of realizing value begins.

Look outside the project before looking deeper inside it

One of the more interesting findings in the PMI article is what Budzier calls the uniqueness trap.

When people convince themselves their project is unlike anything attempted before, they stop looking for useful comparisons. They discount historical evidence, lean too heavily on internal assumptions, and talk themselves into believing that other teams' difficulties don't apply to them.

Research by Bent Flyvbjerg, Alexander Budzier, M.D. Christodoulou, and M. Zottoli found a statistically significant link between perceived project uniqueness and underperformance. Their recommendation: reference-class forecasting, premortems, similarity-based forecasting, and other forms of "decision hygiene" to counter the bias.

Every project has distinctive features. Very few are genuinely unprecedented. This is why recognizing common patterns across seemingly unrelated fields can help teams uncover solutions that might otherwise remain hidden. A hospital system and a financial platform look nothing alike on the surface, but both involve data conversion, user adoption, workflow redesign, training, security, and operational transition. A public infrastructure project and a corporate transformation look nothing alike either, but both involve uncertain requirements, multiple stakeholders, scarce expertise, governance delays, and optimistic forecasts.

Good planning asks two questions, not one. "What's special about this project?" helps you tailor the approach. "What is this project similar to, and what happened when others attempted comparable work?" keeps you honest.

The discipline beneath every methodology

Project success is sometimes discussed as though the answer is finding the right framework. Adopt agile. Strengthen governance. Improve scheduling. Use artificial intelligence. Create a PMO. Each of these can help. None of them substitutes for clarity.

Regardless of methodology, successful projects tend to share the same underlying disciplines. They invest enough thought to understand the work before accelerating it. They define success in terms of outcomes, not just outputs. They make assumptions, risks, and dependencies visible instead of implicit. They use evidence from comparable efforts rather than treating the project as exempt from history. They build feedback loops that let the plan evolve without letting the project drift. And they keep someone accountable for value long after the deliverable is done.

The purpose of planning was never to create the illusion that the future is controllable. It's to help people see clearly enough to make better decisions as the future unfolds.

That may be the most transferable lesson in the research PMI highlighted. Successful projects aren't defined by the absence of surprises. They're defined by an organization's ability to recognize what matters, learn faster than conditions change, and keep directing its resources toward an outcome worth achieving.

At your next project review, the most useful question probably isn't "Are we following the plan?"

It's this: Are we still solving the right problem, for the right people, with a plan that reflects what we now know?

This is the gap I spend most of my time closing with clients, not writing better plans on paper, but building the discipline that lets teams tell the difference between progress and drift before it causes them pain, or costs them a program. If your organization is delivering on schedule but not on outcome, that's usually a clarity problem, not an execution problem, and it's solvable.

Sources:

• Project Management Institute. "What 22,000 Projects Reveal About Success and Failure." July 1, 2026.

• Project Management Institute. Maximizing Project Success: What Is Project Success? November 2024.

• Project Management Institute. "Essential 2024 Insights for Project Professionals." December 19, 2024.

• Infrastructure and Projects Authority. "Setting Up for Success: The Importance of Front-End Loading." September 9, 2020.

• Flyvbjerg, Bent; Alexander Budzier; M.D. Christodoulou; and M. Zottoli. "Uniqueness Bias: Why It Matters, How to Curb It." Saïd Business School Working Paper, 2024.

Clarity Is the New Competitive Advantage

From Noise To Clarity

There was a time when the companies with the most information usually won. That made sense; information was expensive. Market research took months. Reports arrived by mail. Data lived in filing cabinets, and finding the right document often depended on knowing who had the key.

Today, information is almost free. Every manager has access to more data before breakfast than most executives saw in an entire year thirty years ago. Dashboards update by the minute. AI summarizes reports in seconds. Every meeting produces another slide deck, another spreadsheet, another list of action items.

Yet something curious has happened: the more information organizations collect, the less certain many leaders seem to become.

The meetings are longer. The decisions take longer. The projects become larger. The priorities multiply. Everyone is busy. Few people are clear.

That should concern us, because history suggests organizations rarely fail from a lack of intelligence. They fail because intelligent people become overwhelmed by complexity.

The garage with 200 bins

A friend of mine once described his garage as "well organized." Every tool had a place. Every shelf was labeled. There were hundreds of plastic bins.

Then one Saturday he needed a Phillips screwdriver. He spent twenty minutes looking for it.

His garage contained every tool he needed. What it lacked was clarity.

Many organizations look remarkably similar. They have strategic plans, performance dashboards, project portfolios, risk registers, steering committees, status meetings, customer surveys, financial reports, and now AI-generated summaries on top of all of it.

None of those things are , each exists for a good reason. The problem is that they accumulate faster than anyone stops to ask a simple question:

What problem are we actually trying to solve?

That question is surprisingly rare. Instead, organizations tend to ask questions that are easier to answer: Should we create another dashboard? Schedule another meeting? Gather more data? Wait until we're certain?

The irony is that certainty rarely arrives. More information frequently creates more questions. Research on this is consistent with what most of us feel day to day: studies on knowledge work have found that a large share of the week goes to reading and responding to messages, leaving less room for the thoughtful, value-creating work leaders are actually there to do. The result isn't simply lost productivity. It's fragmented thinking.

Driving in the fog

Imagine driving across the countryside on a clear morning. You don't know everything; not what lies fifty miles ahead, not what every other driver intends to do. Yet driving feels easy, because you can see the road.

Now imagine driving through dense fog. The road hasn't changed. Your car hasn't changed. You haven't forgotten how to drive. Only one thing has changed: you can no longer see clearly.

Leadership works much the same way. Most executives aren't searching for more information. They're searching for visibility.

What Lean actually teaches

This is one of the reasons I've always admired Lean thinking. People often assume Lean is about reducing cost. It isn't, cost reduction is usually a consequence. Lean is really an exercise in making work visible.

When Toyota developed practices like visual management, standard work, and the Five Whys, they weren't trying to create more reports. They were trying to remove confusion. Every improvement began by asking, "What is actually happening?" Only then did they ask how to improve it.

That sequence matters. Clarity comes before improvement.

Why AI raises the stakes

Artificial intelligence makes this lesson even more important. AI can summarize a thousand pages before you've finished your coffee. It can draft proposals, analyze customer comments, write code, forecast demand, produce beautiful charts.

What it cannot do is decide which problem deserves your attention. That remains a human responsibility.

Organizations that use AI simply to generate more content may find themselves drowning in faster-moving information. Organizations that use AI to remove noise will discover something far more valuable: clarity.

Complexity isn't going anywhere on its own, the volume of organizational data and communication keeps climbing every year, and most leaders already feel it. That means the competitive advantage of the next decade won't belong to the organization with the most information. It will belong to the organization that can distinguish the important from the merely interesting.

Clarity by subtraction

This is harder than it sounds, because our instincts tell us to add: another report, another metric, another meeting, another approval, another process, another initiative.

But clarity usually comes from subtraction. A sculptor doesn't create a statue by adding marble, he creates it by removing everything that doesn't belong. Leadership is much the same.

Every unnecessary meeting removed creates time for thinking. Every confusing metric eliminated sharpens attention. Every redundant approval accelerates delivery. Every priority abandoned gives another priority room to succeed. None of this is glamorous. It is, however, effective.

Perhaps that explains why the best leaders often appear calm. It isn't because their organizations are simpler, it's because they've learned to simplify what matters. They know which numbers deserve attention, which problems deserve immediate action, and most importantly, which distractions deserve neither.

Peter Drucker put it well: there is nothing quite so useless as doing efficiently what should not be done at all. The observation feels even more relevant today. AI allows us to do almost everything faster. It doesn't tell us whether we should be doing it. That judgment still belongs to leaders.

Where the real advantage lives

The organizations that thrive over the next decade will certainly use better technology. They will almost certainly use more AI. But those won't be the reasons they succeed.

They'll succeed because they cultivate something much harder to copy; they create clarity where others create complexity. Their people understand the mission. Their projects support the strategy. Their measures reflect what matters. Their meetings end with decisions instead of more meetings.

That is not simply good management. It is a competitive advantage.

In a world overflowing with information, clarity has become one of the rarest resources in business and like every scarce resource, it's becoming more valuable every year.

This is exactly the kind of noise-to-signal work I help operating teams and PE portfolio companies do, cutting through complexity to find the few things that actually move performance. If that's a conversation worth having, let's connect.

The Future of Management Consulting: Why AI Will Increase the Demand for Human Judgment

As AI commoditizes analysis and information, the consultants who thrive will help organizations make better decisions, redesign workflows, and deliver measurable outcomes.

Artificial Intelligence is transforming nearly every industry, and management consulting is no exception. As AI tools become more capable, many organizations are beginning to ask an important question:

If clients can perform their own research, generate reports, analyze data, and create business plans using AI, what role will management consultants play in the future?

It is a fair question. After all, many of the activities traditionally associated with consulting, research, analysis, benchmarking, presentations, and documentation can now be completed in minutes using generative AI.

Some have interpreted this as a threat to the consulting profession. I see it differently. In fact, I believe AI will increase demand for the most valuable aspects of consulting: judgment, leadership, change management, and organizational transformation.

We've Seen This Before

Throughout history, major technological advances have changed how organizations operate.

Personal computers transformed how we created and stored information.

• Email transformed communication.

• The internet transformed access to information.

• Social media transformed marketing and customer engagement.

Each innovation initially generated tremendous excitement. Organizations rushed to adopt new technologies, often without a clear understanding of how to use them effectively. Eventually, the technology matured. Organizations learned where it created value, where it did not, and how to integrate it into everyday operations. AI appears to be following a similar pattern.

Today, many organizations are experimenting with AI. Employees are using it to write reports, summarize meetings, generate presentations, create project plans, and conduct research. While these applications are valuable, they often represent isolated productivity improvements rather than true transformation. The organizations creating the greatest value from AI are not simply using AI to do existing work faster. They are redesigning work itself.

AI Is Not Just a Productivity Tool

Many discussions about AI focus on productivity gains. Certainly, AI can help professional's complete tasks faster. Project managers can draft schedules in minutes. Analysts can summarize thousands of pages of documentation. Marketing teams can generate content at scale. These improvements are significant, but they represent only the first stage of AI maturity.

The second stage involves process redesign. Organizations begin asking questions such as:

• If AI can automate 40% of a process, why does the process still exist in its current form?

• Which activities require human judgment?

• Which decisions can be delegated to AI?

• How should roles and responsibilities evolve?

For example, consider customer service. Historically, customer inquiries were routed to employees who reviewed, categorized, prioritized, and resolved requests. Today, AI can perform many of those initial activities automatically. As a result, organizations are redesigning workflows rather than simply accelerating existing ones. This is where substantial value begins to emerge.

The third stage is operating model transformation. At this level, AI influences how organizations make decisions, structure teams, manage knowledge, govern operations, and create value. This is not a technology challenge. It is a leadership challenge.

What Clients Will Want From Consultants Five Years From Now

As AI becomes more accessible, clients will need less help gathering information and more help making sense of it.

The future consultant will create value in five critical ways.

1. Turning Information Into Insight

AI can generate enormous amounts of information. What it cannot do is fully understand the unique context, politics, culture, priorities, and constraints of a specific organization. Executives will increasingly need trusted advisors who can help answer questions such as:

• What matters most?

• What should we prioritize?

• Which risks are hidden beneath the surface?

• Which opportunities deserve investment?

The value will no longer be information. The value will be interpretation.

2. Leading Organizational Change

Technology has rarely been the hardest part of transformation. People are.

Organizations have spent decades implementing ERP systems, CRM platforms, Agile frameworks, Lean initiatives, and digital transformation programs. The common challenge has never been the technology itself. The challenge has been helping people adopt new ways of working. AI will be no different. Organizations will need guidance on:

• Leadership alignment

• Workforce adaptation

• Skills development

• Governance

• Change management

• Organizational culture

These remain fundamentally human challenges.

3. Redesigning Workflows and Operating Models

Many organizations currently ask: "How can we add AI to our existing processes?"

The more important question is: "How should our processes change because AI exists?"

This distinction is critical. Adding AI to a broken process often accelerates inefficiency.

Redesigning the process creates sustainable value.

Consultants who can help organizations rethink workflows, decision rights, governance structures, and operating models will be in increasing demand.

4. Providing Judgment in Complex Situations

AI excels at identifying patterns. Leaders still must make difficult decisions. Organizations face trade-offs every day:

• Cost versus quality

• Speed versus risk

• Innovation versus stability

• Short-term gains versus long-term value

These decisions often involve competing stakeholder interests, incomplete information, and political realities. Human judgment remains essential. The consultants who thrive in the AI era will be those who help leaders navigate complexity rather than simply analyze it.

5. Building Trust and Accountability

One of the most overlooked aspects of consulting is trust. Organizations do not hire consultants solely because they lack information. They hire consultants because they need confidence.

Executives often seek independent validation before making major investments, launching strategic initiatives, or implementing organizational change.

AI can generate recommendations. It cannot assume accountability. When a critical transformation succeeds or fails, leadership remains responsible for the outcome. Trusted advisors will continue to play an important role in helping organizations make informed decisions with confidence.

What This Means for Project and Program Management

The implications are particularly significant for project and program leaders. Within a few years, AI will likely generate many traditional project management artifacts automatically:

• Project charters

• Work breakdown structures

• Risk registers

• Communication plans

• Status reports

• Meeting summaries

• Draft schedules

As a result, project managers will spend less time creating documents and more time creating alignment. The future project leader will focus on:

• Stakeholder engagement

• Governance

• Prioritization

• Risk-based decision making

• Organizational change

• Strategic alignment

• Team leadership

In many ways, AI may allow project and program managers to return to the most valuable aspects of their profession: leading people and delivering outcomes.

The Future Consultant

There is a common misconception that AI will replace consultants. A more accurate prediction is that AI will replace some consulting activities while making other consulting capabilities more valuable than ever.

The consultants who succeed over the next decade will not simply be experts in AI. They will be experts in leadership, decision-making, organizational behavior, change management, governance, and business transformation, while also leveraging AI as a force multiplier. The future consultant will be:

• Part strategist

• Part change leader

• Part organizational architect

AI can generate possibilities. Organizations will still need help turning those possibilities into measurable outcomes. That is where great consulting creates its greatest value.

Final Thought

The future of consulting is not about providing more information. Information is becoming abundant. The future of consulting is about helping organizations make better decisions, redesign how work gets done, build confidence in times of uncertainty, and transform potential into performance.

As AI becomes more capable, the value of expertise may decrease. The value of judgment, leadership, and trust will increase. And that is why management consulting is unlikely to disappear. It is more likely to evolve.