I have worked on projects at a TV channel where every show has a fixed air date, at a game developer opening new offices, and at a bank rolling out new systems. The projects that hurt most rarely had a bad idea. They skipped a stage: started without a plan, or never really ended.
Below: the stages of a project in the order they happen, what each one must deliver, how to decide that a stage is done, an example you can copy, and the mistakes I see most often. To check your own project right now, go to the phase gate checklist.
What are the stages of project management?
A project life cycle is the series of phases a project passes through from initiation to completion. A phase is a set of logically related activities that ends with one or more outputs. In my course I use four stages:
- Initiation — is this project worth doing at all?
- Planning (organization and preparation) — how exactly will we do it?
- Execution — build the team and do the work.
- Closing — accept the result, release the people, learn.
Project management itself is described by five process groups: initiating, planning, executing, monitoring and controlling, and closing. The fifth one is not a separate period in the calendar. It starts with the first task and lasts until closing: compare the results with the plan, deal with problems, keep everyone informed. So when one guide says “four phases” and another says “five phases”, both are right. One counts stages in time; the other counts kinds of work. For a small project all four stages can take just a few days, but the sequence stays.
Stage 1. Initiation: should we do it, and can we do it?
Every project begins with an idea or a need. Sometimes initiation is informal: a short discussion and a verbal agreement. But two questions have to be answered either way:
- Should we do it? Are the benefits worth the costs? Is there a better way to solve the problem?
- Can we do it? Is it technically feasible? Do we have the resources?
The tools for those answers are a cost-benefit analysis (all expected benefits against all costs of doing the project and supporting the change it brings) and a feasibility study. The result of the stage is a decision to move on, plus the project charter: purpose, measurable objectives and success criteria, high-level requirements and boundaries, overall risk, a short milestone schedule, pre-approved money, key stakeholders, exit criteria, the project manager with their authority, and the sponsor who approves it all.
Stage 2. Planning: what goes into a project plan
Planning turns the charter into a document that says how the project will be done. My checklist for a project plan:
- why the project exists and a detailed description of the expected outcomes;
- all constraints and all assumptions;
- all tasks, broken down, and who plays which role;
- a detailed schedule;
- people, money and other resources: equipment, facilities, information;
- how you will manage the major risks and uncertainties;
- a communication plan and a quality assurance plan.
Very large projects are hard to plan in full detail up front. For them I use rolling-wave planning: break the first three months into pieces of two weeks or less, plan the rest in packages of one to two months, and every three months detail the next block.
Stage 3. Execution: preparation first, then the work
Execution does not start with tasks. It starts with a short preparation step that teams like to skip:
- assign people to every project role;
- introduce the team members to each other and to the project;
- give and explain the tasks to everyone;
- agree how the team will perform its core functions;
- set up the tracking systems;
- announce the project inside the organization.
Monitoring and controlling: the phase that runs alongside execution
Monitoring and controlling is three habits, repeated every week: compare performance with the plan, deal with the issues that come up, keep everyone informed. The simplest format I know is a regular team meeting in three parts: progress, plan, problems. What we did, what we do next, what stops us. For a specific issue, call a separate meeting with everyone affected and record each action with an owner.
Monitoring also protects the date. I prefer to release on the date we set at the start, because in my experience everyone takes a slipped deadline very badly.
Stage 4. Closing: the stage most projects skip
Ending is what makes a project a project. Yet when you are pulled into the next assignment, the finished project lingers and fades away. Then you never learn whether it was conceived, planned and run well, and the team never feels the work was finished.
A closing checklist I use:
- finish all open activities and all required deliverables;
- get formal approval and sign-off on the results, including from the customer — a signature or written confirmation, not a thumbs-up in a chat;
- resolve disputes with vendors, pay all bills, close contracts;
- close the project accounts so that no time or money can be charged to them later;
- acknowledge each person’s contribution, thank them, help them move to their next assignment;
- announce to the organization that the project is complete;
- hold a lessons learned review and archive the documents.
At the lessons learned meeting ask: Did we achieve all the goals? Did we meet the schedule and the budget? Could we have foreseen the problems? Did we handle them well when they came? Leave with two lists: practices to repeat and practices to avoid, each with next steps. Collect lessons during the whole project, one line each time something surprises you; the meeting only puts them together.
Two more rules from my certification-prep course. Plan closing at about 5% of the project time — on a four-month project that is a week. And a cancelled project is closed the same way: accept what exists, report, record the lessons, release the people, archive. The next project manager needs to know what was built and why it stopped.
Phase gates: how to decide a project can move to the next stage
A phase gate is the last step of a stage. The project is checked against agreed criteria, and only projects that meet them move on. If one or more criteria are not met, there are two options: keep the project in the current stage and bring the weak points to an acceptable level, or cancel it. There is no third option called “let’s just start and see”.
The checklist above follows that rule: a missing ★ item means “not ready”; other open items can pass the gate with an owner and a date.
Do the stages change for agile and hybrid projects?
The stages stay; their rhythm changes. In a predictive life cycle scope, schedule and cost are fixed early and any change goes through change control: plan, build, test, deliver, one release at the end. In an adaptive life cycle the team works in iterations of one to four weeks: build a small working piece, show it to the customer, learn, repeat. A hybrid keeps a predictive skeleton — fixed milestones and budget — with iterations inside. In large companies most projects are hybrid, even if nobody calls them that.
Five yes-or-no questions pick the life cycle: Are the requirements clear today? Is change expensive? Is the work regulated? Is it a fixed-price contract? Can the customer give feedback every few weeks? Yes to the first four — predictive. No to the first two and yes to the fifth — adaptive. Mixed answers — hybrid. More on the mindset behind it in my article on building an agile culture.
Project stages example: an office move
I know cases where HR led the renovation of a new office and the whole move. Take the example from my course: moving 200 people. Clear requirements, a signed lease, fire safety rules, a fixed-price moving company, and you cannot move half the people and ask how they like it — a predictive project. Its stages:
- Initiation: why we move, costs against benefits, the charter with the sponsor, the budget and the target date.
- Planning: who moves where and when, the task list and schedule, the vendor contract, risks such as a delay in the building works, the communication plan for employees.
- Execution and control: the team and roles, an announcement to the company, weekly “progress, plan, problems” meetings against the schedule.
- Closing: sign-off on the new office, contracts closed and invoices paid, thanks to the team, a lessons learned review for the next move.
Common mistakes at each stage of a project
- Skipping “should we do it?” Nobody compares benefits with costs.
- Starting work before the plan exists. Monitoring then has nothing to compare with.
- Planning a huge project in full detail up front instead of in rolling waves.
- Polishing instead of releasing on the agreed date.
- No visible results for months. People stop believing first.
- Letting the project fade instead of closing it.
- Passing a gate with must-haves missing.
If you want to build all of this into your own work — the charter, stakeholders, scope, schedule, resources, team, control and closing — it is the structure of my Project Management Course: PMI PMBOK 7 Training, ten lessons that follow a project from the first idea to closing, with the templates I use.
