What is project management maturity?
Project management has major benefits (Demir and Kocabas, 2010). However, PM maturity is an evolution of the process - a way of keeping track of how effective your decision-making, strategies, and overall processes are. A company’s maturity level will differ based on the industry and it’s short and long term goals. Additional factors include the resources available, the strategies in place, and the scope of projects undertaken (Crawford, 2007).
One way to improve project maturity is to utilise artificial intelligence (AI). As Elrajoubi describes, AI can augment your pre-existing capabilities, helping your teams work more efficiently.
Despite these benefits, AI hasn’t taken a hold in project management as you would expect. Here, we’ll look at how AI can be used to improve PM maturity levels, as well as looking at the reasons that might be preventing widespread adoption.
How is AI used in PM?
Artificial intelligence includes an incredibly broad range of tools and solutions. One that can benefit PM, in particular, is machine learning. Najjar and Sarraj (2019) discuss AI’s potential to provide actionable insights, and this is primarily through machine learning.
Machine learning algorithms take data from previous projects and use this data to build estimates of project requirements. This could include:
- Project length
- Required resources
- Staff scheduling
- Risk management.
By building strategies based on the data from previous projects, AI can help reduce overall costs and increase efficiency.
Of course, machine learning isn’t the only aspect of AI that can help. There are many tasks that are incredibly time-consuming, yet relatively easy to automate, such as booking rooms, scheduling meetings, filing reports and requesting team updates.
These tasks can be assigned to an AI virtual assistant (Mendieta (2018). This reduces the amount of time project managers have to spend on administrative tasks, letting them focus on more important things.
Using AI to improve PM maturity
Evidence shows that companies who use AI in their PM achieve higher levels of project maturity. The ‘AI Innovators’ global survey by Villanova University notes that:
- 81% of companies were being impacted by AI technologies
- 37% consider AI adoption to be a high priority
- 23% of projects currently use AI, and this is projected to rise to 37% in the next three years.
In a study based on the ‘Cognizant’ and ‘ESI ThoughtLab’ survey (which spoke to 2,491 executives between April and July 2019), Greenstein (2019) considered project management maturity - particularly, digital maturity. After segmenting respondents into four stages (beginner, implementing, advancing, and leading), he found a clear correlation between those that used AI and those with a high level of digital maturity.
In general, respondents that fell into the ‘beginner’ category (what we would call level 1 maturity) didn’t see themselves as being advanced in AI. Respondents categorised as ‘leaders’ (akin to level 5 PM maturity), however, did.
While this doesn’t prove a direct link, it does show that ‘beginner’ companies are likely to use basic AI, whilst leading companies use it to its fullest. In particularly, they’re able analyse data, gather actionable insights, and understand how to apply them.
Barriers to AI adoption in PM
As mentioned, adoption of AI is slower than you may expect. Here are some of the main things holding it back…
Many people have preconceptions about what AI is, and what it can do. This can lead to companies not knowing what the benefits are, or being unable to visualise how it might fit in with their strategies. Another issue, highlighted by Marr (2019), is a reluctance to ‘hand over control’ - either to the AI, or to the staff who manage it.
Of course, even companies that do want to adopt AI face certain hurdles. As a relatively new industry, talent is in high demand. As Hupter (2020) notes, ‘companies need the right mix of talent to translate business needs into AI systems to enable efficiencies’. It’s not just the need to learn new technologies, however. The way companies work is also changing. Hupter states that:
- 71% of those who adopted AI report a change to skills and job roles
- 82% believe it will change them in the next three years.
In order to use AI to its full potential, management needs to find the right split between hiring external talent, and training existing employees. Whilst hiring experts can bring a boost of skills to any team, investing in training narrows the skill gap, strengthening expertise throughout the company.
One aspect of PM maturity is the role of standardised processes. Level 1 maturity companies have no standard processes at all. Level 2 have some, but they’re ‘per project’ rather than structurally mandated. At level 3, these processes are standardised throughout the company.
However, when we get to level 4 and 5, companies begin to reflect on these processes. Companies at level 4 regularly measure and review them, while at level 5, there is a concentrated effort to measure, review and constantly improve them.
As you can see, the more efficient your processes are, the more likely your projects are to succeed. We can apply a similar scale to the adoption of AI. At low levels of adoption and understanding, data can be gathered on how projects are currently doing - but insights are not shared or analysed.
At higher levels of adoption, companies can use this data to provide predictions, handle risk management, and plan ahead.
Why is this a barrier, you might ask? Low maturity companies who adopt AI at a basic level may quickly be put off as it won’t give the results they’re hoping for. However, this is an issue with their processes, rather than with the technology.
Data and systems
AI is only as good as the data it works from. As Lahmann, Keiser and Stierli (2018) state, for deep insight, AI needs a huge data set. To achieve the highest level of maturity, companies need to invest in large amounts of data gathering and preparation.
That’s not to say AI is useless without such a large data set - companies with limited data sets can still gain insights into current and past projects. This is an important step to building integrated AI processes into projects. However, companies need to be aware that they need to invest in building strong data sets alongside implementing AI solutions in order to reach the highest levels of maturity.
Building reliable, useful data sets can be a challenge in and of itself. Not only do large amounts of data need to be gathered, but also processed. The steps of preparing data - identification, cleansing, and transforming - can be time consuming, particularly for low maturity companies. Companies at higher maturity levels are likely to find this easier, as they already have processes in place that can speed up this step.
As mentioned, finding people with the right skill set can be a problem. Marr (2019) notes that ‘a bottleneck exists’ when implementing these solutions, and one such bottleneck can occur at the data gathering/preparation stages without the right talent. However, with time and further development, AI should be able to take over this step - accelerating the overall process.
In order to achieve high project maturity, standardised processes need to be backed up by appropriate support systems. Teams need to be able to answer the following:
- What types of data are being stored?
- Where is the data being stored?
- Who has access - and what level of access do they have?
- What needs is this data set meeting?
As you can see, it quickly becomes complicated. For companies with standardised support systems, AI can easily be implemented. However, companies at lower maturity levels will face another bottleneck as they set up these systems in advance.
Given the positive data from Villanova University (2019) and the findings from Greenstein's (2019) study, it’s clear that AI has great potential. Companies that implement AI in their project management should see their maturity growth increase - as long as they build the right team and commit enough resources. It’s no longer enough to factor in people and processes into maturity levels. Business must now factor in AI technologies too.
As we’ve seen by looking at the barriers companies face, there are challenges for companies at lower maturity levels. Whilst they may benefit the most from AI, they’re also likely to find implementing it challenging. These challenges can be overcome, but unless tackled, it increases the chances of their projects remaining inconsistent. For these companies, it’s vital to invest the time and effort needed to get past the initial hurdles. Once they’ve done so, they’re likely to see a rise through the PM maturity levels as well.
About the author
Lloyd Skinner, Chief Executive Officer, Greyfly Ltd, is a project professional with 25 plus years of experience working in multiple sectors and projects in both support and delivery roles. For the last two plus years, he has been investigating the use of AI in project management and developing the Greyfly proposition.