A project and a programme are distinguished by what is being led, whatever their size. In common usage, the word project refers to a time-bound effort, carried out to deliver a defined result: an application put into service, a site opened, an organisation put in place. The word programme refers to a set of linked projects, brought together under a single leadership to obtain a benefit beyond the reach of each one taken in isolation. The difference between a programme and a project therefore lies in what is accounted for: one delivery in the first case, the effect of several deliveries in the second.
These working definitions do not reproduce the text of any standard. They are enough to frame the choice that matters to senior management: run several projects separately, or bring them together into a programme. We propose to settle it on four criteria (a common benefit, dependencies, shared resources, duration), then to weigh up the price of a programme, in structure and in decision-making delays. Set up without a common benefit, a programme adds a layer of management to projects that did not need one.
The difference between a programme and a project, and the place of the portfolio
A project manager accounts for a delivery, on the agreed date and at the agreed cost. A programme director accounts for an effect, which the deliveries must produce together. Each project keeps its manager, its plan and its budget. The programme carries what lies between them: the order of go-lives, the commitments of one project to another, the trade-offs, the preparation of the teams who will receive the result.
We propose a test to place a set of projects. If one of them is removed, do the others keep their reason for being? If so, the set is a list. When removing a single one deprives the others of their usefulness, the set is run as a whole.
The portfolio is a third object. It brings together projects and programmes, linked or not, because they draw on the same budget and the same teams, and it is used to choose those the organisation commits to. Our entry on PPM gives its vocabulary. Our page on programme management sets the three levels side by side in a table.
The word “programme” can also name a large project, or a department’s budget line. The former remains a project as long as it has a team, a plan and a delivery; the latter is a share of the portfolio. The Project Management Institute awards a separate credential for project, programme and portfolio, as presented in our entry on PMI certifications.
The criteria that justify setting up a programme
The table presents these criteria in the order in which we propose to examine them; the grid is adjusted to each organisation.
| Criterion | What to look at | Argues for a programme | Argues for separate projects |
|---|---|---|---|
| Common benefit | The result expected by senior management, and the person who promised it | A result that no project produces alone, carried by a single sponsor | An investment case and a benefit specific to each project |
| Dependencies | The deliveries of one project that another needs | Cross dependencies, renewed at each stage | An isolated, one-way dependency, dated once |
| Shared resources | Scarce skills, key users, cutover periods | The same people called on by several projects for the same result | Separate teams, or sharing that the portfolio already arbitrates |
| Duration | The horizon of the result and the number of go-lives | A long horizon, divided into stages that each call for a decision | Short projects, delivered once |
In this grid, the criteria do not carry the same weight. The common benefit is the condition: without it, a programme director has nothing to account for, and the other criteria merely describe projects that get in each other’s way. Dependencies then say whether this benefit is exposed. The more they cross, the more a delay spreads without any project manager having the authority to arbitrate.
Shared resources, on their own, do not make a programme. Two unrelated projects competing for the same architect raise a question of capacity, and this is settled at the level where the organisation orders all its projects: this is the purpose of project portfolio management. Duration, finally, acts as a threshold. A programme requires a set-up (mandate, governance bodies, programme office) that only pays off if the whole lasts long enough to use it.
Suppose a haulier wants to merge two warehouse networks after an acquisition, and entrusts four project managers with the warehouse management system, the item master data, the transport plan and the reorganisation of the sites. None of these projects produces the unified network alone. The new system cannot open a site without the master data. The same operators are expected in acceptance testing by all four teams, and the sites cut over in waves. All four criteria are met.
A programme has a price, in structure and in decision-making delays
Setting up a programme adds a level between the projects and senior management: a programme director, a sponsor who devotes time to it, a body that arbitrates and, as soon as there are many projects, an office that holds the overall plan and the registers. Our article on the role and responsibilities of the PMO describes this office at programme level.
The price is also paid in the projects. Each project manager reports twice, to their sponsor and to the programme. They lose control of certain dates, which the programme sets for everyone. A decision that used to be taken between them and their management goes through one more body, and therefore one more delay if the delegation thresholds are not written down. This allocation is covered in our article on programme governance.
The absence of a programme also has a price when the criteria are met: nobody holds the order of go-lives or the commitments of one project to another. The choice therefore compares two costs, that of the structure and that of its absence.
When separate projects are enough, and when the programme exists without saying so
Several situations do without a programme:
- projects that have nothing in common but a department or a budget envelope belong to the portfolio, which orders and arbitrates them;
- a large project divided into work packages, with a single management team and a single go-live, remains a project, and its work packages are coordinated in its plan;
- two projects linked by a single delivery make do with a written commitment between the two project managers, with a subject and a date of need.
The reverse situation exists: projects run separately form a programme that nobody has set up. Project managers negotiate dates among themselves without anyone able to decide between them. A benefit has been announced to senior management, and no project file carries it. A delay in one shifts the go-live of another, and senior management learns of it through the second.
The choice can be revised. Projects already launched can be brought together along the way, provided a sponsor is appointed and their mutual commitments are brought into an overall plan, described in our article on the master schedule. A programme can also hand its projects back to their departments on the day they no longer depend on one another.
The first criterion provides the starting point: ask the prospective sponsor to write down the common benefit and to name the person who will account for it after the last go-live. If they cannot, the projects remain projects, and their links are handled at portfolio level.