Meet Jennifer, Operations Manager at a 75-person logistics company
- Oversees warehouse, shipping, and inventory management
- Tasks: process improvement, team coordination, vendor management, KPI reporting
- Pain points: documenting processes, training new hires, coordinating across departments
The Before State
- Process documentation is outdated or doesn't exist
- Tribal knowledge lives in people's heads (risky when someone leaves)
- Spends 10+ hours per month answering the same questions from different people
- KPI reports are manual, time-consuming, and often out of date
Knowledge Base Setup
- Folder structure:
Operations/[Function]/[Processes|Training|Metrics|Projects]- Functions: Receiving, Warehousing, Shipping, Inventory, Vendors
- File naming:
Function_DocType_ProcessName_Version_Date - Context documents:
- Standard Operating Procedures (SOPs) for each major process
- FAQ document by function (updated monthly based on actual questions received)
- Vendor contact sheet and SLA summary
- Team responsibilities matrix (who owns what)
- KPI definitions and calculation methods
Tool Integration Choices
- Email AI for team and vendor communication
- Project management tool with no AI (team prefers manual task tracking)
- Standalone AI for process documentation and training material creation
- Excel AI for KPI analysis and narrative
Three Core Workflows
Workflow 1: Creating/Updating Standard Operating Procedures
- Process changes due to new software, vendor, or efficiency improvement
- Document current state by shadowing team members, taking notes
- Use AI to draft initial SOP: "Create an SOP for this warehouse receiving process: [paste notes]. Format: purpose, scope, step-by-step instructions, quality checks, troubleshooting."
- Review AI draft with team members who do the work
- Revise based on feedback (AI often misses nuances, edge cases)
- Add visuals (photos, diagrams) manually
- Use AI to generate training quiz questions based on SOP
- Publish to shared drive, announce to team
Workflow 2: Answering Recurring Questions
- Notice same question being asked repeatedly (e.g., "How do we handle damaged inventory?")
- Review existing documentation or create answer if none exists
- Use AI to draft FAQ entry: "Write a clear, concise answer to: [question]. Context: [relevant SOP or policy]. Audience: warehouse staff. Tone: friendly, practical."
- Review AI draft, simplify language if needed
- Add to FAQ document
- Next time question is asked, send FAQ link instead of re-explaining
- Monthly: review FAQ for most common questions, consider creating formal training
Workflow 3: Weekly Operations Report
- Pull KPIs from systems (shipments, inventory accuracy, order fill rate, etc.)
- Calculate week-over-week and target variance
- Use AI to draft report narrative: "Summarize this operations data [paste table]. Context: [link to KPI definitions doc]. Focus on: anything off target, trends, recommended actions."
- Review AI summary, add operational context it couldn't know:
- "Fill rate down because of supplier delay on Product X"
- "Overtime up due to unexpected large order from Key Customer Y"
- Use AI to suggest talking points for leadership meeting
- Format report, add charts manually
- Distribute to management team
Quality Control Checklist
- Does the SOP accurately reflect how the work is actually done (not how I think it's done)?
- Have I validated the process documentation with the people who do it daily?
- Is the FAQ answer clear enough that someone could follow it without asking for help?
- Do the KPI trends in my narrative match the actual numbers?
- Are my recommendations practical and actionable, not just AI-generated generic advice?
- Would a new employee understand this documentation?
The After State
- Saves ~8 hours per month on process documentation and reporting
- New hire onboarding time cut by 30% (better training materials)
- Team asks fewer repeat questions (FAQ is comprehensive and current)
- Leadership reports are clearer and more actionable
- Still manually handles sensitive HR issues and vendor negotiations
Common Mistakes for Operations Managers
- Letting AI write SOPs without validating with frontline workers (misses reality)
- Using AI-generated KPI explanations without operational context
- Not maintaining FAQ document (AI needs reference material)
- Assuming AI understands your specific operational constraints
- Over-documenting processes that change frequently (documentation debt)