These days, it seems many are talking about how AI is going to replace jobs, but here's the truth: AI won't necessarily replace you. A person who's learned AI will. The real divide isn't between humans and machines—in my opinion, it's between people who are learning faster than the rate of change and people who aren't.
In this guide, I want to walk you through how agile leaders are turning learning into a measurable part of their operating system, not just some side activity. Plus, I'll show you a few ways to track progress and experiment with this inside your teams this week.
Hi, I'm Lance Dacy, otherwise known as Big Agile. I spend a fair amount of my time coaching teams, leaders, and even organizations at large in sustainable agility practices—what I like to call next-gen agility. AI is definitely one for the books in recent years, so I'd like to show how we can incorporate it daily instead of making it some fear factor.
The Problem: Learning as a Special Event
Here's the reality: most organizations treat learning as a special event—a workshop, a training day, a certificate to hang on the wall. I think those are valuable, but in today's environment, the half-life of skills is shrinking really fast. In fact, I found some data from McKinsey that estimates most technical skills expire in two to three years. That means half of what we know today will be outdated by the time we finish the next big project.
The good news is if you're already using agile ways of working, you've got a built-in advantage. Agile is built on short feedback loops and continuous improvement—we're already trying to learn. The question is: are we going to use the same loops to continuously upskill our people, or are we still separating the learning from the doing?
The Solution: Learning as an Agile Loop
I see a shift, and I call it "learning as an agile loop." Here's how forward-thinking teams are fixing the learning problem: they treat learning like a sprint deliverable.
Every sprint has two outcomes:
- A product increment that we're trying to build
- A learning increment
In addition to asking "What did we ship this sprint?" we're also asking "What did we learn?" It's a small shift, but it changes everything.
The Agile Learning Loop Process
The agile learning loop looks like this:
- Identify a skill or knowledge gap linked directly to your current work
- Turn that into a micro experiment—something to test this sprint
- Reflect on that in the retrospective: What did we learn? How will we apply it next time?
- Start tracking that learning
Think of it like how some teams handle architectural spikes or technical spikes, but define a learning goal and experiment rather than just a deliverable or a prototype. You don't need a dashboard—just track it in your sprint review or retro to start out.
Three Real-World Learning Outcome Examples
Learning Outcome #1: Prompt Tuning for Accuracy
Goal: Prompt tuning accuracy for our AI search feature
Experiment: Compare contextual versus generic prompts on 50 internal queries
Insight: Contextual prompts might improve accuracy by 30%, but double GPU costs
Action: Adopt contextual prompts only for high-value flows and document prompt templates in a shared repository
Learning Outcome #2: Testing with AI-Assisted Tools
Goal: Upskill test engineers to use AI-powered test case generation like GitHub Copilot or TestGPT
Experiment: Generate automated test cases for one critical module, compare coverage versus manual approach
Insight: AI generated 40% more edge case scenarios but required 15% manual cleanup
Action: Integrate AI test suggestion tool into regression testing pipeline and create an AI test review checklist for next sprint
Metrics: Code coverage improved from 78% to 91%, and manual test authoring time dropped 25%
Learning Outcome #3: AI for Sprint Analytics and Flow Insights
Goal: Upskill Scrum Masters and Product Owners to use AI dashboards for sprint retrospectives to show flow analysis
Experiment: Use an AI analytics plugin to summarize sprint metrics and surface cycle time or lead time anomalies automatically
Insight: AI flagged three high-impact blockers earlier than manual tracking would have, saving roughly 10 development hours during the sprint
Action: Add AI flow report to sprint retrospectives and train the team on interpreting anomaly trends
Metrics: Cycle time variance reduced by 18%, and retrospective prep time down 50%
The Leader's Role: Building a Learning Culture
This shift only sticks if leaders model it, because culture is built around what our leaders constantly reward. If you want a learning culture, here are three things you can do starting this week:
1. Leaders Learn in Public
Share what you're experimenting with, even if you don't have it all figured out. This helps build psychological safety in teams, especially when leaders model vulnerability and say, "Hey, I'm trying this new thing and here's how it's going."
2. Make Learning Visible
Add a line item to retros asking "What did we learn as a team?" and start making that visible to people who aren't on your team. We're taught that retrospectives are closed events just for the Scrum team, but what if we started making some of these learning events visible to build organizational safety?
3. Reward Curiosity, Not Perfection
Highlight teams that took smart risks, learned something new, or improved a process. Showcase and celebrate that at the organization level.
Tracking the Leadership Shift
To measure this leadership shift, track the frequency of learning discussions in your sprint or team meetings. Even simpler, start a shared document or repository where people list what they've learned this month and how it helped customers, teams, colleagues, or product stakeholders. That's real data on growth, not guesswork.
The Competitive Edge: Adaptability Over Automation
Here's the takeaway: The future doesn't necessarily belong to the most automated teams. AI can write 20 billion lines of code more than we can, but it's not always good. The future belongs to the most adaptive teams—the ones who are curious, who are willing to experiment.
AI will give everyone access to these tools, including our competitors and teammates. But how fast we learn to use them? That's the competitive edge.
Your Action Plan: Start This Week
My advice is to treat learning like a system, track it like a process, and celebrate it like a win. This week, try one of these experiments:
- Add learning goals to your sprint planning
- Track learning velocity for one month
- Create a 15-minute skill share at the end of every sprint review to show stakeholders
If you run one of these experiments, I'd love to hear what you discover about how your organization might or might not be open to that approach.
Ready to Build Your Next-Gen Agile Team?
The shift to learning-driven agile practices requires more than just good intentions—it requires structured guidance and proven frameworks. If this sparks ideas for building more adaptive, learning-focused teams in your organization, explore the comprehensive agile transformation classes that Big Agile offers to accelerate your journey toward sustainable agility.
Stay curious, and remember: the most successful teams aren't just building great products—they're building great learning systems that evolve with every sprint.