How I Think AI will change Agile Project Management
AI's impact on Agile Project Management and Scrum Mastery will go from “interesting” to “total game-changer” faster than you think.
My team and I have spent years at the intersection between AI and software creation, and we have some fascinating conversations with product managers, product owners and project managers, Scrum masters and the like. Probably people like you.
So I wanted to write about the direction in which AI is taking agile, scrum and project management.
Really good AI is still very green. Not all this tech is ready, but I will stick my neck out and say it will be in the next six months.
TL;DR: Don’t leave it until it’s too late to explore how to safely integrate AI.
Your development team is in the middle of a crucial sprint, and suddenly, an unforeseen issue arises, disrupting the entire project timeline.
In tech, such hiccups can cost you dearly in terms of time and resources. Plus, you have to figure out how to explain this to management and potentially your customer.
But what if AI could help you anticipate and mitigate potential challenges before they even occur?
Enter AI-powered predictive analytics.
By tapping into historical data and employing advanced machine learning algorithms, predictive AI solutions can analyse patterns, identify trends, and forecast potential obstacles in your project's path.
Let me give some examples.
- Estimations. Human estimates are flawed by nature. We’re just not wired to do it. AI will enable realistic sprint planning, release planning and better resource allocation.
- Risks. AI will be able to spot risks and bottlenecks far more consistently and – on average – faster than humans can. That means you can mitigate them before they cause problems.
- Prioritisation. AI-powered analytics will be able to prioritise and adaptively reprioritise your product backlog efficiently. There will be far fewer overheads to this process when driven by AI, and it’ll spot dependencies and keep everybody strategically aligned on what matters automatically.
The backbone of any successful Agile team lies in collaboration and effective communication.
But keeping everyone on the same page is a huge time drain.
Miscommunication (and its consequences) is among the most-mentioned frustrations of the PMs I speak to. That grows exponentially as the complexity (of projects and teams) increases.
And that is not to mention the hours out of every day that engineers and PMs spend catching up on Slack or Teams, fishing through old messages to find resources or working out what work has been done on other areas of the project.
That time spent on information-seeking is, for most teams, necessary. But I think AI will turn that “time spent” into “time wasted”.
Let me illustrate:
- No more trawling. AI will be able to understand everything happening on every project you’re working on and surface the important information from the tools you use like Jira, Slack, Teams and GitHub.
- All-knowing AI. LLMs are now more than good enough to allow you to ask any question you like about project progress, risks or the like and give you a concise, actionable answer.
- Fewer, better meetings. For one thing, there should be no need in an AI world to spend time in meetings on progress updates, or summarising data. Meetings will be more strategic and creative. I don’t know many people in software who wouldn’t leap at this one.
I’m working on a tool called CollabGPT that will do just this. It’s not just a dumb layer between your data and an LLM. We’re using a complex network of AI agents to give your AI long-term memory.
It’s like having a person in your team that’s been there since Day 1 and knows everything that’s happening.
It’s geared for teams that use agile project management to create software. Try the beta, if you’re interested.
Continuous improvement is inherent to the agile methodology, the agile manifesto. It’s all about enhancing your team's efficiency, productivity, and effectiveness with each sprint.
I think AI represents an opportunity for a significant shift – or “step up” if you like – in how continuous improvement happens.
Let’s look at what this could look like for your team.
- Quality. It’s already possible to support processes like code review and deployment with AI, and the development process itself has a wealth of tools available. I wrote in depth about this here.
- Performance insights. AI is already available to help you understand your team's performance, identify patterns, and make data-driven decisions to improve your processes.
It will be far more adept than humans at everything from high-level insights to highly granular and specific insights. Use them to pinpoint areas for improvement. It’s real-time and has almost no time overheads, which speeds the whole thing up and means the agile planning process can be far more dynamic.
- Resource allocation. Make sure everyone is working on the tasks that align with their skills and strengths – or even their opportunity for growth. It’s a win-win. You boost productivity, and your foster a more supportive culture.
Let’s turn down the hype for a moment. Right now, embracing AI to overhaul traditional project management and Scrum practices isn’t an absolute must-have. After all, much of the tech is very green, with many AI tools in Beta or still using old underlying models (like GPT-3, which is fine but isn’t going to change the world.) You’re probably not losing significant ground over your competitors.
This clock is ticking faster than any figurative time bomb I can remember.
It will be a matter of months, not years, before at least partial adoption of AI software development tools is no longer a luxury but a necessity.
Adopting and integrating the right tools safely will be the greatest challenge for team members who make decisions about tools for the agile cycle.
It’s pretty much a full-time job to keep up with advancements in AI.
So I created a concise, practical newsletter specifically for people who create software, to help you stay on top of what’s changing.
My team of AI enthusiasts and I pick out the news that matters – what tools are out there, what stage of development they’re at, and how they can be used – to help you cut out the noise and focus on what matters.