The project management office (PMO): when you need one and how to build it in a mid-sized company

Take a hypothetical manufacturing company with 300 employees. At the same time it is implementing a new ERP system, commissioning a production line, moving its warehouse and running several smaller projects. The same IT and logistics staff are assigned to three projects at once, and the management board only learns about delays when the production start date slips.

In a situation like this, the problem is not any single project but the lack of an overall picture. Nobody decides which project takes priority when people are short, and nobody sees that the quarterly plan exceeds what the teams can deliver. This is a job for a project management office.

There is no single number of projects above which a PMO becomes essential. The following signs will help you judge whether this applies to your company:

A PMO is not a single, predefined institution. The PMBOK Guide, the project management standard published by the Project Management Institute, describes three types of office that differ in the degree of control they have over projects. We compare them in the table below.

In a mid-sized company, starting with the supportive model, with a few firm rules for the largest projects, usually works best. For example, every project above a set budget threshold must have a project charter and report its status once a month. The directive model mainly makes sense in companies that live off projects, such as implementation firms or construction companies.

The choice is not permanent. Many offices start by offering support and only take on more control once they have earned trust. The reverse route, from tight control to support, is much harder.

Many expectations build up around a PMO: that it will improve communication, organisational culture and strategy. In reality, a project office is good at a few specific tasks. It will not change the company’s culture or set its strategy on the management board’s behalf.

The PMO’s most important output is the monthly portfolio review for the management board. It shows all projects on a single page, with their status, risks and the decisions they need. The list below sets out the scope of the office’s day-to-day work.

An office that is to have a real influence on projects usually needs about a year to get from the management board’s decision to a steady working rhythm. A structure on paper can be created faster, but earning project managers’ trust and getting people into the habit of using the reports takes time.

The plan below assumes the supportive model and a start with one person. It is worth showing the first portfolio review to the management board as early as possible, even if it is incomplete. It usually reveals overlapping projects and overloaded people straight away.

In a mid-sized company the PMO team usually starts with one or two people. The most important is the head of the office; the rest of the team can come later, as the portfolio and expectations grow.

The head of the PMO should have experience of running large projects and be able to talk to both the management board and the teams. Certifications such as PMP or PRINCE2 help, but tact matters more: the office works at the interface between departments with different interests. If there is a respected project manager inside the company, it is often better to promote them than to look for someone from outside.

We list the typical roles in a project office below. Further roles usually appear only when the number of projects or the need for data analysis justifies them.

For the first few months, a shared list in SharePoint or an Excel spreadsheet is enough. It lets you find out what information the management board really needs before the company invests in a tool. As the number of projects grows, however, manual updating becomes too time-consuming.

The choice of tool depends mainly on where the teams already work. We compare the most common options in the table below. Whatever you choose, the board’s portfolio review should ideally be generated automatically – for example in Power BI, Microsoft’s reporting tool – rather than put together by hand in a presentation every month.

A project office has costs: salaries, tools and the time project managers spend on reporting. That is why it is worth agreeing, when you set it up, how the management board will judge it after a year. Take a baseline measurement in the first quarter, otherwise there will be nothing to compare with.

Project managers’ opinion can be measured with the Net Promoter Score (NPS), but it has to be calculated correctly. You ask, on a scale of 0–10, whether they would recommend the PMO’s support to a colleague, and subtract the percentage of scores of 0–6 from the percentage of scores of 9–10. The result falls between −100 and +100. A simpler alternative is an average score on a scale of 1–10.

The indicators below are enough for the first year. Set the target values based on your baseline measurement, not on other companies’ data, because projects vary greatly from company to company.

The most common mistake is an office that demands reports and templates from day one but gives nothing in return. Project managers then see it as extra bureaucracy and fill in the documents with as little effort as possible. Trust is built by helping with a specific, difficult project or by securing a quick decision from the management board in a dispute over people.

The second mistake is a PMO without a mandate. When the decision to set it up is taken by the director of a single department, other departments treat the office as someone else’s initiative. Other typical mistakes: