We need a radical overhaul in how we deliver projects, and now is a good time to start, writes Martin Paver
When we emerge from the COVID-19 crisis, the government will be introducing several measures to stimulate the economy. Will the new world order provide the trigger we need to be bold and think differently about how we deliver projects? I certainly hope so, but investment in new approaches will be hard to come by. We need realism and to build momentum organically.
I am convinced that advanced data analytics lies at the heart of this. Now, more than ever, we need to be able to model the impact of delays on our portfolio, conduct scenario analysis, share our experiences of the impact of emergent risk, understand supply-chain dynamics and maintain critical capacity. We need to move beyond the world of spreadsheets.
APM’s recent Salary Survey found that 60 per cent of project professionals are using data analytics to some extent in their work. And momentum is building via the Project Data Analytics Community, a non-profit community offering designed to share best practice on leveraging data within a project, programme and portfolio environment. Nearly 6,000 people have now joined this initiative.
In wider society, we have seen the positive impact of community, with people working selflessly to combat the effect of the coronavirus. If we want to transform how we deliver projects, we need a similar approach. But we all need to recognise that our professional futures will look very different. We need to disrupt or be disrupted.
I envisage this evolving via the following:
We need case studies where innovators are using data science to significantly impact project out-turn. These are emerging, but we need many more. We need to experiment, but in a coordinated way that pools our collective resources to map the way ahead.
In 2011, Alex Budzier’s research into 1,471 IT projects observed that “one in six projects developed into a black swan with 200% or more cost overruns and schedule delays of 70%”. Black swans are characterised by their improbability of happening, but is one in six improbable? Did they all suffer from a unique failing or are there are patterns in the data that demonstrate themes, many of which may be entirely foreseeable and predictable? Data can help us to challenge this bias, identify lead indicators, evidence the need for action and shape future forecasts.
In 2008, the national risk register identified a pandemic as a high impact, high likelihood event, with a likelihood the same as a severe weather incident. But the report concluded that “the containment of the SARS outbreaks globally reconfirmed that traditional public health and infection control measures can be successful in containing a new infectious disease”. Their conclusion differed to those countries with direct experience of the 2003 SARS epidemic, who appear to have been much more prepared for COVID-19. Is there a pressing need to challenge bias as these ‘fat tail risks’ become increasingly probable?
Our professional community is upskilling a new cadre of professionals who will be pathfinders within their organisation – change agents who drive a culture that values project data and helps to improve quality.
Meanwhile, meet-ups are at the core of the Project Data Analytics Community initiative, helping us to understand the art of the possible and the benefits of greater adoption.
We also need to remove the fear and barriers to implementation of data analytics. We hold community hackathons every four months to provide hands-on experience. It’s impressive what can be achieved in a weekend.
- Data availability
One of the biggest barriers to innovation is the availability of data. Start-ups really struggle to get hold of it. When they do, it is often of variable quality. We need to change this. We’ve been working with the construction and oil and gas sectors to develop the concept of a data trust: a facility that enables us to securely pool data for the benefit of the collective. We will also put a governance framework around it to spiral up data scope and quality to ensure that it aligns with our AI-driven aspirations. Without this, we won’t be harnessing the data that we need.
- Democratising tools
Power apps and Python in Power BI look scary, but a lot of the fundamentals are available via off-the-shelf tools. Why do we repeatedly pay people to create simple apps, code and dashboards when we could collaborate to get them out there a lot quicker and drive up data volumes? A win-win for everyone, while protecting IP on top-end algorithms.
It’s time to be bold
No single organisation can deliver this future. But by creating a movement, we can change multiple professions over multiple sectors. We pivot how we deliver projects. We automate the repetitive, develop apps to harvest data and use our collective experience at a functional level to flag up areas of focus.
By way of example, if I lead a major project, do I want a risk manager in the future who helps me to administrate risk and conduct qualitative risk assessment? Or do I want someone who understands the principles but can reach into a vast dataset to understand the probability and impact of risks, pulling on people with similar experiences on similar projects?
We begin to challenge bias and scenario-model the implications of management action, overlaid onto the models of the implications of COVID-19. I would want someone who can provide a crystal ball that is founded on evidence. Just one example, but it will impact every discipline.
We have a pressing need for change. Not a tweak, but a radical overhaul.
To get involved in the Project Data Analytics Community, click here. Join the big conversation on future trends in project management at APM’s Projecting the Future page.