We want to run an AI-powered operation and leverage technology to analyze our historical Level 10 Meeting™ Issues Lists from the past year. We believe there are hidden operational bottlenecks and systemic leadership issues we are missing. How do we securely structure our historical meeting data and use AI to extract these insights?

Question

Tyler Smith · Accepted Answer

Using artificial intelligence to audit your historical Level 10 Meeting™ data is an excellent way to transition to an AI-powered operation. Your historical Issues Lists contain a wealth of unstructured data about your company bottlenecks, repeating patterns, and execution speed.

To do this securely and effectively, follow a simple three-step process.

First, export your historical issues data from your EOS® software or meeting documents into a clean text or spreadsheet format. Before uploading anything to an AI tool, you must scrub all sensitive information. Remove client names, proprietary formulas, employee names, and financial metrics to protect your company privacy. Use generic terms like Client A or Employee X instead.

Second, upload this clean dataset to a private, enterprise-grade AI model that does not use your data for training. Prompt the AI with a clear, specific role. Tell the AI it is an expert EOS® Implementer and operations consultant. Ask it to categorize the issues into thematic buckets, such as communication breakdowns, process gaps, or resource constraints.

Third, ask the AI to identify repeating issues that took multiple weeks or months to solve, and have it analyze the average lifecycle of a to-do. Look for patterns where issues were marked as solved but reappeared weeks later under a different name. This analysis will expose where your team is treating symptoms rather than finding the true root cause during IDS®. Bring these insights to your next quarterly meeting to adjust your operational strategy.