Reinventing Execution Systems: The Future of Project Management
I’ve seen a “simple” roll-out in Lagos drag out to 6 months because one approval sat on one desk while the market moved on. I’ve seen a smaller team deliver something bigger… faster… because their execution system could see reality and adjust. Here’s my position and I won’t change it: most strategy failures are execution system failures, not strategy failures. In 2026, technology is not a tool you “use”. It’s an ambient co-worker… embedded, intelligent, always on. If your execution model still relies on weekly status meetings and heroic Project Managers “chasing,” you’re already late. This article isn’t hype. It’s not theory without scars. And it’s definitely not a silver bullet. It’s a practical, Nigeria-relevant guide to transforming execution… so strategy stops silently bleeding in your delivery pipe. Recognition: You’re not imagining it, execution has become a different sport If you run projects in Nigeria (public or private sector), you’re managing volatility as a feature, not a bug. FX moves. Fuel moves. Policy changes. Vendor uncertainty. Power and connectivity realities. Distributed teams across time zones. And customers? They don’t wait for your Gantt chart to catch up. So yes, your old playbook feels heavier. Because it is. PMI’s work shows organisations are already moving towards hybrid delivery as a “fit-for-purpose” response, and the adoption curve has been moving fast. (Project Management Institute) That’s the signal. The game is shifting from “managing projects” to running an execution system. Frustration: The part people says out loud Here’s the fact you’ve probably whispered to a colleague: We’re not losing because we can’t plan. We’re losing because we can’t adapt fast enough. Planning isn’t the problem. Your execution system is. I learned this the hard way. A few years ago, I led a cross-functional initiative that looked perfect on paper: scope clear, timeline clean, stakeholders “aligned.” We even had a beautiful RAID log. Then reality arrived. Procurement took longer than promised. A key dependency moved without notice. Two decision-makers went quiet. My team kept building… while the business changed direction. We delivered… technically. But it landed like a missed pass. That failure taught me something I now treat as law: If execution data doesn’t reshape decisions in real time, you’re not executing, you’re performing. And performance is expensive. McKinsey has been repeating a painful refrain for years: around 70% of transformations fail. Not because leaders don’t care. Because execution breaks under complexity. (McKinsey & Company) Different decade, same disease.
Time to get quantitative and make you feel uneasy. PMI’s 2025 report reveals that project professionals with high business acumen are more likely to meet business objectives, adhere to schedules and budgets, and have fewer failures. These are not mere words. These are metrics. Similarly, Gartner has been predicting for a while that by 2030, 80% of the work of project management (status updates, reporting, and other administrative tasks) will be replaced by AI as it automates much of the PM role. That means if you are worth your salt because you can get “status updates,” you are living on borrowed time. If you are worth your salt because you can “create a learning-based delivery ecosystem,” you will be priceless. However, there is a caveat. According to a recent report by Reuters, Gartner cautioned that more than 40% of agentic AI projects are likely to be abandoned by 2027, as they often lack clear business outcomes and are implemented for fears of being left behind. That means the future is not about “AI.” It is about “AI, with a framework and a governance model and an ecosystem that is designed to leverage AI.”
That is what execution must evolve into. A nervous system. Reinvention doesn’t mean chaos—it means orchestration Once leaders get to this point, they usually exhale. “Reinventing execution” is a scary phrase, but it’s really about replacing a rigid, hierarchical delivery mechanism with an execution system that can: recognize that something is happening, immediately decide with context, move quickly, always be learning. But this isn’t just a Silicon Valley phenomenon. In our work at JK Michaels coaching leaders across industries, we continue to see the same dynamic play out: the companies that are winning aren’t the ones with the coolest tech, they are the ones that are innovating in how decisions get made. This is the money quote: It isn’t about being fast. It is about making great decisions fast.
Possibility: A playbook for reinventing the execution system for Nigeria Ok, let’s get real and actionable. Here are 5 strategies I’ve seen work in the real world of messy organisations, high uncertainty and limited resources
1) Stop funding projects. Fund products and outcomes. Projects are finite. Value delivery shouldn’t be. When you fund a “project”, you reward teams for delivery, regardless of changing market conditions. When you fund a product/outcome, you reward teams for continuous delivery until value is received. Example in the wild: A fintech company in Lagos shifted from “Project Alpha” to “SME Lending Product.” The budget remained fixed, but the work to be done changed every month based on repayment data and fraud patterns. The work never “ended.” It evolved. Lesson learned: outcomes promote accountability without the drama. Takeaway: If your funding model rewards project completion more than value, your execution system will optimise for the wrong thing. Build a single “execution control plane” across the ecosystem Most organisations don’t have a delivery problem. They have a fragmentation problem. Work lives in emails. Updates live in WhatsApp. Risks live in someone’s head. Decisions live in “let’s circle back.” Modern execution needs one control layer where work, dependencies, and decision rights are visible. PMI’s 2024 work highlights that performance can be strong across remote, hybrid, and in-person models—if the organisation provides the enablers. The enabler isn’t “more meetings.” It’s shared visibility and fast routing. Example: A construction programme shifted from weekly reporting to daily signal capture: site constraints, approvals, materials ETA. When one supplier slipped, the system rerouted crews and resequenced tasks without waiting for the Monday meeting. Lesson: visibility is not micromanagement. It’s oxygen. Takeaway: If you can’t see dependencies live, you can’t execute at machine speed—even with great people.
Reduce process to “minimum viable process” All processes carry “technical debt” — old decisions, old processes, old reflexes that served a purpose long ago. In Nigeria, this is more important, because legal compliance, good governance, and reputational risk are all critical considerations. I’m not advocating “removing all controls.” I’m advocating “removing all anti-control”. Example: One team managed to shrink the change management chain from 9 sign-offs to 3 by creating clear decision boundaries: low risk changes, no approval needed. Medium risk changes, duty manager approves. High risk changes, escalate. Same control. Less of a chokepoint. Lesson: process exists to enable value, not slow value. Takeaway: test your process, and make one ruthless decision: does this step make us safer, or just less trusting?
If a process doesn’t protect value, it’s a tax. Shift leadership from “control” to “orchestration” (the mindset change) Old leadership: approve every proposal, review every plan, catch every error New leadership: structure the system to make good choices even when you’re not there .
The PMI 2025 report does a great job of capturing this change: business acumen enables professionals to move from a tactical to a strategic role In the language of execution systems, leaders concentrate on: In the context of decision making, the most relevant roles are decision rights, which involve the right to make or approve a specific decision. No simple definition or explanation can be offered here. Instead, I’ll say that escalation paths are like “in case of emergency, break glass” type situations, when things get hairy, politically charged, or you simply need someone with more authority to step in. Escalation paths can vary, but examples include a direct manager, a director, a VP, or other team members who can offer support or guidance in specific scenarios. railings, the allocation of capacity, indicators of success. Micro-story: I once saw a director require daily meetings with each workstream. It was “just in time.” In reality, it was a single point of failure. After changing the escalation policy, the director got less reporting but had better results because the system was no longer dependent on their contribution.
Lesson: Control seems secure. Orchestration is more secure. Takeaway: Your role is not to review. Your role is to ensure reviews are only needed in exceptional cases.
Leadership is a latency issue.
Treat AI as a collaborator, and then test for “task collapsibility” AI on the job should not be thought of as an author of fancier status reports. It should be thought of as a tool to reduce cycle time: to predict threats to the schedule identifying scope creep indicators The automated summarization of decisions workarounding via constraints coming up with “what-if” situations
It is here that I go back to what Reuters/Gartner cautioned: a lot of agentic AI initiatives fail because the business case isn’t strong. So you will need a metric: collapsibility of tasks. Where can AI compress effort by 75 percent without increasing risk? For instance, a PMO automated weekly reporting, dependency analysis from meeting minutes and risk trending using AI. Reporting cycles were compressed but what benefited the organization more was that the project managers were able to use those hours to align with stakeholders and negotiate priorities which is what AI cannot do. Moral of the story: AI should reduce effort, not increase complexity. A corollary: begin with repetitive, rule-based and high volume tasks. Then climb up.
First fix the mess. Then automate it.
Objections (you’re not naive though, are you?) “We don’t have the budget for this.” Cool. Re-invention is not always a new platform. Sometimes it’s: Decision rights—the power to make and enforce key choices about an organization’s people, strategy, and operations—should be the starting point of any organizational restructuring. streamlining the approval process The shift to outcome-based funding having a single source of truth for execution Those are political more than financial. If you do go down the tool path, make sure it’s after you’ve done the operating model or you’re just going to automate a lot of badness. Tooling alone doesn’t fix dysfunction. It multiplies it. “We tried that once before and it didn’t work out”. Same thing. Most “execution” transformations “fail” because they pursue a method rather than a new system.
Most strategy failures are execution system failures, not strategy failures.
If they didn’t dig into decision flow, capacity allocation, and incentives, that attempt was a rebranding, not a transformation. What if you don’t transform? You will still get things done. Just more slowly. More expensively. More painfully. With each passing quarter. You will exhaust your best talent. You will erode your stakeholders’ confidence. Your strategy will keep shifting — but the way you execute won’t. And worst of all? You will begin to dial back your aspirations to your execution capacity. That’s how organisations quietly die. This is the kind of capability boost organisations like those we partner with at JK Michaels are quietly building — not to keep up with the Joneses, but to be able to execute their strategies in an AI-driven, increasingly turbulent world. So here’s a parting fact, just like the one at the beginning: Most strategy failures are execution-system failures, not strategy failures. Fix the system, and suddenly your people are “geniuses.” Leave the system unfixed, and even geniuses seem “incompetent.”
