Get The Facts & Set a Performance Baseline

Purpose

To help new Caddis customers establish clear visibility into their operations by collecting objective data on machine performance, cycle times, and downtime. This playbook walks users step-by-step through building a performance baseline that can be used to drive meaningful improvements.

“Before we had it, we were just taking people’s word for it… we weren’t measuring whether we made an improvement by uptime or runtime.”

‍— President, Aluminum Manufacturer

Timeline

7-10 days from integration to activation

Outcomes

• A 10-day snapshot of real-time machine performance
• Objective baseline metrics (runtime %, stop count, cycle consistency)
• Visual reports to align teams
• Key insights that surface first improvement opportunities

What You'll Use In Caddis

• Real-time machine dashboards
• Cycle time and stop count reports
• Shift comparison tools
• CSV or Excel export tools (for internal review or presentations)

Who's Involved

• Plant Manager (Project Sponsor, Drive Engagement)
• Shift Supervisors (Provides Content into Data Trends)
  
• Maintenance Lead (Helps Identify Mechanical Factors)
• Production Operators (Shares Process Feedback, Validates Assumptions)
• Caddis Champions (Oversees Platform Usage, Pulls Reports)

Step-by-Step Guide

Step 1: Select 3–5 Focus Machines

Choose machines that are:

• Mission critical (impacting daily production goals)
• Historically problematic (frequent issues)
• Visible to multiple shifts (helps with accountability analysis)

✅ Tip: Start small. Too much data early can lead to analysis paralysis.

Step 2: Let the Data Flow (7–10 Days)

Let Caddis do its job. Do not attempt to fix anything yet.

• Let Caddis run uninterrupted
• No operator interventions

💡 Use this period to observe behavior. Where do machines sit idle longer than expected?

Step 3: Schedule a ‘Get the Facts’ Team Meeting

Hold a 30–45-minute meeting after the observation period.

Suggested Agenda

1. Objective: Understand our actual performance vs. assumptions
2. Data Review: Use Caddis to show:
   • Total runtime hours vs. available hours
   • Most frequent stops per machine
   • Longest running cycle
   • Differences between shifts (if any)
3. Operator/Team Feedback: Are these results expected? Surprising?
4. Document Observations: What stands out? What trends do we see?

Step 4: Build the Baseline Report

Create a 1-page summary including

• Uptime %, Downtime %, Idle Time
• Avg. Cycle Time and Variance
• Top 3 Unplanned Stop Reasons
• Shift-to-Shift Performance Comparison)

Optional Template Sections

• Key Insights:
      “Shift 1 shows 14% more runtime than Shift 2.”
• Immediate Opportunity:
      “#2 Die Cast Machine has 4+ stops/day without clear reason codes.”
• Screenshots from Caddis dashboards (add visuals!)

Step 5: Assign Owners to Track Trends

Assign one owner per machine or department to:

• Monitor dashboards daily
• Note deviations from baseline
• Update team weekly on observations

👥 Sample Roles: Supervisor, Maintenance Lead, Cell Lead

Step 6: Hold a Weekly ‘Data Huddle’

Duration: 15–20 mins, weekly
Attendees: Caddis Champion, Shift Leads, Maintenance

Topics:

• Where are we above/below baseline?
• Any new failure patterns?
• Did any improvements correlate to uptime?
• Pick one area for deeper analysis next week.

Metrics to Track:

Metric

• Runtime %
  
  • Shows actual machine availability
• Downtime
  
  • Identify bottlenecks or mechanical issues
• Cycle Time
  
  • Reveals consistency in process
• Shift Comparison
  
  • Help identify training or process variation

Common Challenges (and What to Do)

Challenge

• “This doesn’t match what we were told by the team.”
  
  • Response: That’s expected — the point is to replace guesswork with facts. Use it to drive positive change, not blame
• “The data feels overwhelming.”
  
  • Focus on trends, not perfection. You’re looking for outliers, not explanations for every detail.
• “Operators feel exposed.”
  
  • Reframe Caddis as a tool for improvement, not surveillance. Share wins that come from the data.

What Success Looks Like

• Your team no longer debates what happened- you can show it
• Leadership meetings include data, not guesses
• You've identified your first improvement project (changeover, PM, etc.)