Choosing a Role in Project Data Analytics

Following on from our previous posts on How to get a job in Project Data Analytics and Who applies for project data analyst internships I’ve taken the opportunity to mull over what you should consider when choosing a role.

There is rapidly growing interest in project data analytics; its an area that will experience huge growth and opportunity within the next 5 years. But how do you find a role in it?

The job market is split into the following segments:

Product based organisations.  There are a number of companies, particularly startups, who are beginning to develop project data analytics products. These could be organisations that are developing platforms or niche AI solutions. They will initially be looking for high end data scientists or software developers. People who can help them to develop concepts into commercially viable solutions. As they expand, they are likely to extend their service offerings into consultancy.

Consultancy based organisations.  All of the major consultancies are expanding their capabilities in project data analytics. They are likely to recruit across al levels, from junior roles through to seasoned consultants. If they want to charge the big fees then they will need to demonstrate that their team has a high level of professional training, competence and experience. The smaller consultancies are likely to be hesitant about investing in advanced data analytics capabilities because the market isn’t quite ready. But this will change over the coming months.

Project based organisations.  Larger organisations are just beginning to understand the potential. Most are operating in the realm of excel, with some extending into PowerBI or Tableau. From my experience, most organisations haven’t yet understood the intrinsic value in the exhaust plume of data that is emitted from a project. But this will change.

I would envisage the roles in the large organisations to fit into the following categories:

  • Extracting insights from swathes of data using excel. They will probably require an understanding of VBA, DAX and how to apply a pivot table.
  • Organisations used to be hooked on word or powerpoint based reports, but the more enlightened ones are now starting to move on to business intelligence tools such as PowerBI or Tableau. There is plenty of online material to help you to upskill in the use of these tools. Datasets are also available to practice on.


But this is an area that is likely to progress rapidly. Organisations initially moved static powerpoint based charts into PowerBI and made them interactive. But the real progress will come from automatically extracting, cleaning and integrating data. You’ll probably need to have a firm grasp of python and understand the utility of the libraries of algorithms.

You’ll also be required to have a high level appreciation of data engineering, i.e. understanding the data architecture, how data is permissioned and how it can be glued together. You’ll also need to understand data security and be familiar with GDPR and data protection.

The majority of the roles in large organisations will probably sit within the Portfolio Management Office, but these roles will extend into projects as capabilities grow. Project managers will have an increasing appetite to call on the advice of data analysts to enable evidence driven decision making, displacing some of the conventional project management roles.

This will see an evolution in roles from reporting towards:

  • Extracting insights.Understanding the predisposition of a project to variance. Predicting risks that may arise on a project based upon the dataset of projects that have gone before. Understanding how a schedule is likely to evolve.
  • Functional maturity.Understanding the maturity of project management across the organisation. Running scripts to understand how often project management systems are updated, whether the update is significant and to what extent does the content compare with best practice.


This will require an increasing level of competence in data science, extending into machine learning.

Which type of organisation do you want to work for?

This is a matter of personal choice. I spent 20 years working for government and they provided me with a supportive environment in which to learn. I was mentored throughout and there were various controls that limited my degrees of freedom. Its very structured and most large organisations have strong support mechanisms in place. There is plenty of variety and opportunity. You are a small cog in a big machine, but you could be working on a huge, multi million project.

At the other end of the business is working for a startup. Training is likely to be less structured and more hands on. You’ll be expected to try and find your own answers, pulling on forums, communities and your own network; but support is there when you need it. You’ll probably be covering a far broader breadth, ranging from recruitment through to deploying machine learning algorithms. Its likely to be less structured, with less support mechanisms. Its also more difficult to hide; if you don’t perform then a small organisation probably can’t afford to carry you for long. You are a large cog in a small machine.

There is no right or wrong answer to this. If I was to try and help someone, I would ask:

  • Are you a self starter or do you benefit more from supervision?
  • Do you prefer structured training or are you happy to dip into Youtube and MOOCs?
  • Are you looking for a structured career path or a more organic one?
  • How driven are you and how fast do you want to rise to stardom?
  • Are you happy to invest time in your own development or would you prefer your organisation to structure it for you?

The product of this discussion would help you to decide which road to follow. But I would also say that whatever you choose, its not a life decision, you can always change afterwards; but wherever possible give it a year before jumping. If you jump too often then employers will begin to think that you may struggle to hold down a regular job.

If you get to interview try and understand the organisation’s passion for project data analytics. Are they doing it because they have been told to, or are they setting the pace and setting the vision? If it is viewed as something of strategic importance then you’ll be joining a rollercoaster.

Your portfolio.

Keep track of everything that you have done. Use Github for any code and keep a library of your portfolio of reports and insights. You may need to anonymise some of them, but proving examples helps to demonstrate your grasp of the subject. It doesn’t hurt to turn up to an interview with a few pages of examples. But be careful about commercial confidentiality; if your future employer has any inkling that you have breached trust then your credibility will be damaged.


All organisations are different. There are some huge generalisations in the statements above, but the intent is to help you to navigate a route through.