Toolsday

August 4, 2026 · Last updated on July 27, 2026

AI Toolsday: Real AI, Real Agile, Real Workflows

# Continuous Improvement

Practical AI workflows for Agile teams

Prepare for PI Planning Before Everyone Enters the Room

One of the biggest challenges in PI Planning isn't identifying risks—it's discovering them too late. Teams often spend the first part of planning uncovering issues that have existed for weeks or even months.

This workflow helps you use AI to analyze insights from your previous PI so you can walk into PI Planning with a clearer understanding of recurring risks, dependencies, and discussion topics before the event even begins.

📌 Before You Start

The AI tools featured in this workflow are examples, not requirements. Every organization has different approved tools, security policies, and ways of working.

The goal of AI Toolsday isn't to recommend one specific product—it's to share practical AI workflows that you can adapt using the tools available to you. If your organization uses Microsoft Copilot, Google Gemini, Claude, ChatGPT Enterprise, or another approved AI solution, feel free to substitute those where it makes sense.

Focus on the workflow, not the tool.

🛠️ Tools Used

NotebookLM

Analyze historical planning artifacts and uncover patterns grounded in your organization's own documentation.

ChatGPT

Transform those insights into actionable planning outputs, including risks, mitigation ideas, facilitation questions, and discussion topics.

⚙️ How It Works

Step 1: Build your PI knowledge base

Before jumping into AI, gather the artifacts from your previous PI. The richer the information, the more valuable the insights.

Consider uploading documents such as:

PI Objectives

Business Value scores

ART Metrics

Inspect & Adapt findings

ROAMed risks

Team and ART retrospectives

Confidence Vote feedback

Feature completion reports

Dependency boards

Improvement backlog items

Upload these documents into NotebookLM.

Because NotebookLM grounds its responses in the documents you provide, it becomes an AI assistant that understands your ART rather than relying on generic Agile advice.

Step 2: Identify recurring patterns

Now that your knowledge base is ready, start asking NotebookLM questions that help uncover trends instead of individual facts.

For example:

Which risks appeared multiple times throughout the PI?

What dependencies caused the biggest delivery delays?

Which objectives consistently struggled to achieve Business Value?

What themes emerged during the Inspect & Adapt workshop?

Were there recurring issues between specific teams?

Which improvement items remained unresolved?

NotebookLM is very good at connecting information across multiple documents—something that's difficult to do manually when reviewing dozens of planning artifacts.

The goal is to understand why your last PI unfolded the way it did.

Step 3: Turn insights into a planning guide

Once you've identified the major themes, copy NotebookLM's findings into ChatGPT.

Ask ChatGPT to organize the information into something your ART can use during PI Planning.

For example, have it:

Consolidate similar risks into common themes

Prioritize the most likely challenges for the upcoming PI

Recommend mitigation strategies for each risk

Identify assumptions teams should validate during planning

Generate facilitation questions for breakout sessions

Create a "watch list" of dependencies that deserve early discussion

Rather than simply summarizing information, ChatGPT helps transform historical data into an actionable planning guide.

Step 4: Share the insights before PI Planning

Don't wait until the first day of PI Planning to surface these findings.

Share the planning guide with Product Management, Business Owners, Scrum Masters, and team leaders beforehand so they can begin thinking about potential solutions before planning starts.

This allows conversations during PI Planning to focus less on discovering problems and more on solving them.

💡 Why Agile Pros Care

PI Planning is one of the largest investments of time in SAFe. The more prepared your ART is before planning begins, the more productive those two days become.

Instead of spending valuable planning time rediscovering the same challenges, teams can proactively address known risks, validate assumptions, and make better decisions.

AI doesn't replace the conversations that happen during PI Planning—it helps ensure those conversations start from a stronger foundation.

💬 Prompt to Copy

You are an experienced Release Train Engineer preparing an Agile Release Train for PI Planning.
Review the historical planning insights below and identify recurring risks, dependency patterns, and delivery challenges.
Based on this information:
Group similar risks into common themes.
Rank the five most significant risks likely to impact the upcoming PI.
Recommend mitigation strategies for each risk.
Identify assumptions teams should validate during PI Planning.
Highlight dependencies that should be discussed early.
Generate five facilitation questions that would help teams have more productive planning conversations.

Present the results as a PI Planning Preparation Guide that can be shared with ART leadership before planning begins.
Historical PI Insights: Paste your NotebookLM summary here.

✅ Try It This Week

Even if your next PI Planning event is months away, upload your most recent PI Objectives, Inspect & Adapt findings, and retrospective notes into your preferred AI tool. Ask it to identify the three biggest risks that could carry into your next PI.

You may discover patterns your teams have been experiencing for multiple PIs without realizing it.

💬 Continue the Conversation

When you think back to your last PI Planning event, what was the biggest risk or dependency that caught your ART by surprise? Looking back, do you think the warning signs were already there?

🤝 AI Toolsday Tip

AI is most valuable as a collaborator—not a replacement for your expertise. Use it to accelerate your work, uncover patterns, and prepare for better conversations, but always review AI-generated insights with your teams. The people closest to the work provide the context and judgment that AI cannot.

Comments (0)

Popular