The Evolution of Project Analytics

Blue Down Arrow

Welcome to the first in a five-part series examining the fast-changing world of Project Analytics and its vital role in the major projects sector.

This first article explores how the major projects sector is responding to a need for more intelligent project analytics. We explore how recent advances are beginning to dramatically improve the decision-making ability of major projects.

Where are we today with major projects reporting?

Today, major projects get delivered with the help of a wide range of project management, reporting and coordination systems.

This collection of software tools have one thing in common – they create large amounts of disparate data that mostly provides reporting on the ‘here and now’.

Although these systems contain data relating to the same or similar projects, there is a lack of data integration between project applications. Large projects create disconnected ‘islands of data’ that make it hard to convert the raw data into actionable intelligence, particularly for trend analytics and forecasting.

Another challenge project teams face is a lack of standardisation for key terms and data definitions. For example, trying to create a report on a common term such as ‘profit’ across a portfolio of projects can open up a can of worms when each project has a slightly different profit calculation.

Many organisations are striving to make data sharing a priority across their projects, and introduce more advanced data analytics tools, such as Power BI, for cross-project analysis, but despite the desire to move forward, there are still major obstacles to overcome.

Due to the nature of these project ‘data silos’, creating project analytics has always required considerable manual effort and a great deal of technical skill to pull data together for meaningful decision-making. As a result, most projects are forced to rely on the standard reports from their existing project tools, or whatever analysis they can pull together with Excel.

Traditionally, these standard reports have satisfied the needs of project stakeholders. But stakeholder demands are changing, driving the need for more intelligent project analytics.

Why do we need more intelligent project analytics?

With the amount of money and risk tied up in major projects, stakeholders clearly need more intelligence from their project analytics to help make smarter decisions.

As a result of the challenges discussed earlier, it’s fair to say that major project analytics has lagged behind the mainstream data analytics innovations we’ve observed in other sectors.

For example, at The Oakland Group, we’ve been deploying advanced data analytics across many industries for over twenty years so it was only a matter of time before the information locked away in major projects initiatives would begin to fall under the analytics spotlight.

But what can we do with better data analytics, and how will this benefit those endeavouring to deliver more successful project outcomes?

Here are some typical use cases:

  • Timely, accurate reporting: It’s fair to say that most major projects would benefit from more accurate and timely project reporting. It’s not uncommon for many projects to be reporting one or two months in arrears. Merely being able to accurately report what’s happening right now would be a considerable improvement.
  • Answering questions: We need to ask more questions of project data and get accurate, insightful answers quickly and easily. Project staff don’t care about data analytics; they want to ask questions that provide the guidance they need to help their projects move faster and safer. We need to provide the right environment for staff to answer questions via simple, easy-to-use tools in a sustainable and repeatable way.
  • Intelligent forecasting: How can we forecast the trajectory of a major project with greater accuracy? How do we get better at estimation, cost control, quality, and all other well-known project paths? Forecasting has historically been a challenge on major projects, but recent technological advances have finally made this a reality.
  • Artificial Intelligence (AI) and Machine Learning (ML): With greatly improved data integration and data quality, we can incorporate advanced practices such as AI and ML to create even more intelligent forecasting and richer analysis of major projects initiatives.

Improving project analytics presents obvious challenges, but as you’ll learn in this series, these challenges are starting to be overcome within a growing number of UK organisations.

Advice for those starting their project analytics journey

Throughout this four-part series, we’re going to introduce practical examples and stories of project analytics success. For those interested in embarking on this journey, here are some pointers we can share right now that will act as your project analytics ‘North Star’.

#1: Begin with the end in mind

Instead of ‘shooting for the moon’, strive to deliver something achievable that satisfies some of the most pressing use cases – those that typically fulfil a need for improved efficiency.

Obvious candidates here will focus on use cases such as:

  • Improving project assurance: Is there a better way of spotting ‘distressed’ projects so you can more effectively target your project assurance and audit activities?
  • Intelligent forecasting with historical data: Major projects typically have large amounts of data but little insight into what caused specific outcomes. By capturing historical data, you can deliver predictive analytics capabilities that ‘learn’ from historical events. A single view of historical project data also satisfies a requirement for robust project auditing, particularly on taxpayer-funded major programmes.
  • Optimised project reporting: Project stakeholders don’t need more attractive dashboards. Instead, the real demand is for more timely information that is accurate and trusted. Therefore, the assurance and provenance of the numbers contained within your reports should be a key milestone deliverable for your project analytics journey.

#2: Develop an end-to-end appreciation of the journey

It’s tempting to aim for a single, technical ‘silver bullet’ that provides a tactical fix, but it’s only by taking a holistic view that you reap the rewards of project analytics.

Improve your raw material: On major projects, staff typically have to update multiple systems and are often not fully briefed or incentivised on the importance of entering good quality data at source.

An end-to-end focus must begin with your sources of data. Project staff need the right training, support and tools, to record and update data quickly and accurately. If the raw material lacks quality, the output from your project analytics efforts won’t be trusted.

Deliver real-world impact: Your project analytics initiative has to create real-world outcomes that stakeholders and project staff value. The goal is not to develop colourful dashboards but to solve real problems.

When looking at the end to end processes required to deliver a project analytics capability, start with the most valuable linkages between systems (e.g. projects and maintenance) so you can provide some immediate, short-term benefits.

Consistently measure and communicate the benefits your initiative is delivering to help grow the momentum for change.

#3: Assemble the skills required

We occasionally see organisations hiring a data scientist in a bid to ‘fix the project reporting problem’. Whilst data scientists add value, they can’t deliver on the goal of project analytics single-handedly.

You need a specialist team, that typically requires:

  • Data engineers capable of bringing the data together in the first place.
  • Business analysts who understand the system landscape and the art of what’s possible with the disparate data sources available.
  • As you integrate ever-increasing volumes of data, an added challenge of maintaining data governance emerges, requiring data governance analysts to define appropriate controls and stewardship of the data, as well as forging relationships with Group IT.
  • Executives form a crucial element of your team. They require regular communication to understand where you are on the journey, particularly the likelihood of early mistakes and wrong turns.

You need to create a coalition of experienced staff and senior management, all working towards a common goal of translating your vision for project analytics into reality.

#4: Deliver value quickly to reinforce the business case

Building the next generation of project analytics is a considerable undertaking. However, most stakeholders will not sanction a three-year analytics project with some nebulous offer of a reward in the future.

You therefore need to balance a need for the appropriate frameworks, technologies and controls, with the demand for delivering some short-term value. Otherwise, your business case will soon run out of steam.

For example, you’ve no doubt witnessed projects that have gradually moved from a status of green to amber then eventually red, but what if you could have predicted that trend much earlier?

What if you could reduce risk across an entire portfolio of projects without requiring weeks of manual labour and ‘Excel-hell’ to spot the project warning signs?

Your organisation may have a desire to improve material or workforce efficiency, but whatever the focal point, try and achieve some small wins that attract early plaudits and approval.

What next in this series?

Coming up in the next article, we’re going to discuss how to ‘Know Your Data’ when it comes to project analytics.

We’re going to take a deeper look at where organisations go wrong when assembling a project analytics team and how they can start building some practical uses cases that help create early traction.

Previous Post: How to build a Data Governance Program by Stealth: Introducing the Lighthouse Projects ConceptNext Post: Project Analytics: Getting to know your data.