What is project management, and what does a project manager actually do?

What project management means

A project is temporary and unique: it has a start, an end and a specific output. Building a warehouse, launching a website, moving an office and implementing a new finance system are projects. Running payroll every month is not. That is operations, because it repeats with no end date.

Project management is the discipline of getting a project from idea to handover in a controlled way. The PMBOK Guide describes it as applying knowledge, skills, tools and techniques to project activities to meet the project's requirements. In practice it comes down to answering five questions and keeping the answers current:

• What are we delivering, and what is out of scope?
• When will each part be done, and what depends on what?
• How much will it cost, and how much have we spent so far?
• Who does the work, who decides, and who needs to know?
• What could go wrong (or better than expected), and what will we do about it?

What a project manager actually does

The job title sounds administrative, but the core of the role is decision support. A project manager rarely does the technical work. They make sure the right work happens in the right order, and that problems surface early enough to fix. A typical week includes:

• Planning: breaking the scope into a work breakdown structure, estimating durations and costs, and sequencing tasks into a schedule.
• Coordinating people: agreeing who is responsible for each deliverable (a RACI matrix helps), running short check-ins and removing blockers.
• Tracking: comparing actual progress and cost with the plan, and forecasting where the project will land.
• Managing risk and change: keeping a risk register, agreeing responses with named owners, and putting scope changes through a simple approval step.
• Communicating: a regular status report for the sponsor, and a clear answer when someone asks "are we on track?"
• Closing: confirming acceptance, handing over, releasing the team and recording lessons learned.

The project manager is accountable for the plan and its delivery. The sponsor owns the business case and makes the big calls: funding, priority and major scope changes. Keeping those two roles distinct avoids a lot of confusion.

The project life cycle in five stages

Most methods describe the same broad flow, even if the names differ:

1. Initiate: agree why the project exists, what success looks like and who sponsors it. The output is usually a project charter.
2. Plan: define scope, schedule, budget, risks, roles and communication. Save the agreed plan as a baseline so you can measure against it later.
3. Execute: the team does the work, and the project manager coordinates and removes obstacles.
4. Monitor and control: runs alongside execution. Compare actuals with the baseline, forecast the outcome and act on variances.
5. Close: formal acceptance, handover, final cost and lessons learned.

These stages are not strictly sequential. Planning continues as you learn more, and on longer projects you repeat the plan, execute and control loop every week or month.

Scope, time, cost and quality

Every project balances four things: scope (what you deliver), time (when), cost (how much) and quality (how well). Change one and at least one other moves. Add features with a fixed date and you need more people or money, or you accept lower quality. A large part of project management is making these trade-offs visible so the sponsor can choose, rather than discovering them late.

Two techniques make the trade-offs measurable on larger projects. The critical path method shows which chain of tasks sets the finish date, so you know where a delay really hurts. Earned value management compares the value of work done with what it cost and with what was planned, so cost and schedule performance become numbers rather than opinions.

A worked example: a small office move

Suppose a team of 40 people must move to a new floor in 10 weeks with a budget of $60,000. A project manager would:

1. Write a one-page charter: goal (everyone working from the new floor by week 10), scope (fit-out, IT, furniture and the move itself, but not new phones), sponsor (the operations director) and budget ($60,000).
2. Break the work into five packages: fit-out ($30,000, 6 weeks), network and IT ($12,000, 3 weeks after fit-out), furniture ($10,000, 2 weeks), move day ($5,000, 1 week) and contingency ($3,000).
3. Sequence them. Fit-out must finish before IT starts, and the move follows IT. That chain is 6 + 3 + 1 = 10 weeks, so it is the critical path with no spare time. Furniture can be delivered during the IT work, so it has slack.
4. Track weekly. At week 4, fit-out should be 4 / 6 = 67% complete. If the contractor reports 50%, the project is about one week behind on the critical path, and the project manager acts now: an extra crew, weekend work, or a sponsor decision to accept week 11.

Nothing here is complicated. The value is in doing it explicitly and early, so a one-week slip becomes a decision rather than a surprise on move day.

Predictive, agile and hybrid approaches

Predictive project management (often called waterfall) plans the whole scope up front and controls against that plan. It suits work where requirements are stable and change is expensive, such as construction or equipment installation. Agile delivers in short iterations and re-plans often. It suits software and other work where requirements emerge as users see results. Hybrid mixes the two: for example, a fixed overall schedule with milestones, and agile sprints inside the software stream. Choose the approach per project, not once for the whole organisation.

Common mistakes

• Starting without an agreed goal. If the sponsor and the team describe success differently, no schedule will fix it.
• A plan nobody updates. A schedule made in week one and ignored afterwards is decoration. Update progress at least weekly.
• Tracking activity instead of outcomes. "We were busy" is not progress. Measure finished deliverables.
• Hiding bad news. Late warnings remove options. Report variances as soon as you see them.
• Too much process for a small project. A three-week campaign needs a task board and a short list of risks, not a 40-page plan.

How to do this in Critova

Critova lets you start at the level the project needs. A small project can be a simple task board with assignees and due dates. A bigger one becomes a timeline with dependencies and milestones. A complex one gets full project controls: critical path scheduling, baselines, cost and earned value, and a risk register. Team members update their progress on a My Work page that works on a phone, and each project has a printable status report. See how the scheduling features work. Every plan is free during the launch, until 31 March 2027.