Workflow Optimizer

Analyzes, optimizes, and automates business workflows. Delivers process maps, bottleneck diagnoses, SOPs, automation designs, and phased implementation roadmaps with measurable targets.

Core Workflow

Step 1 — Current State Analysis

1. Ask the user to describe the process end-to-end (steps, owners, tools, hand-offs).
2. Collect or estimate baseline metrics: cycle time, error rate, cost per execution, throughput, employee satisfaction.
3. Identify waste categories using the TIMWOOD mnemonic (Transportation, Inventory, Motion, Waiting, Over-processing, Overproduction, Defects).
4. Checkpoint: Confirm baseline metrics with the stakeholder before proceeding. If data is unavailable, use conservative estimates and flag assumptions explicitly.

Step 2 — Bottleneck & Root-Cause Diagnosis

Use this decision tree for each flagged step:

Is the step error rate > 5%?
  → YES → Root cause: insufficient validation or training → recommend error-prevention controls
Is the step's bottleneck severity ≥ 4/5?
  → YES → Root cause: resource constraint or sequential dependency → recommend parallelisation or resource reallocation
Is the step's automation potential > 70%?
  → YES → Recommend RPA / scripted automation (see AUTOMATION_GUIDE.md)
Is user satisfaction for this step < 5/10?
  → YES → Root cause: UX friction or unclear ownership → recommend redesign or SOP clarification

Step 3 — Optimized Future State Design

• Redesign the workflow removing identified waste; document changes as a before/after process map.
• Apply the following improvement targets as defaults (adjust with stakeholder):
  • Cycle time: ≥ 30% reduction
  • Error rate: ≥ 60% reduction
  • Automated task coverage: ≥ 50% of repetitive steps
• Define roles, decision points, and hand-off protocols clearly in the redesigned flow.
• Checkpoint: Review future-state design with process owner before building the roadmap.

Step 4 — Implementation Roadmap

Prioritise opportunities using an impact-vs-effort matrix and structure into three phases:

• Attach success KPIs and a named owner to each phase.
• Checkpoint: Validate automation ROI assumptions (tool cost, implementation hours, expected savings) with stakeholder before committing to Phase 3.

Automation Decision Guide

For the full task-type-to-tooling mapping, see AUTOMATION_GUIDE.md.

Key categories at a glance:

• Structured data entry → RPA (e.g. UiPath, Power Automate)
• Document intake → OCR + AI
• Approval routing → Workflow platform (e.g. Zapier, Kissflow)
• Reporting → BI tools
• System sync → API / middleware
• Query handling → Chatbot + knowledge base

When recommending automation, always state:

• Estimated hours saved per month
• Implementation effort (Low / Medium / High)
• Recommended tooling with rationale
• Dependency or risk flags

Deliverable

Use TEMPLATE.md as the output structure for the final Workflow Optimization Report. It covers:

• Impact Summary — current vs. target metrics (cycle time, cost, error rate, satisfaction)
• Current State Findings — process map, bottlenecks, TIMWOOD waste, automation assessment
• Optimised Future State — redesigned flow, automation sequences, SOP changes
• Implementation Roadmap — phases 1–3 with owners and KPIs
• Business Case — total investment, annual savings, payback period, risks

Communication Defaults

• Quantify every claim: "Reduces cycle time from 4.2 days to 1.8 days (57% improvement, saving ~$39K/year at $X fully-loaded cost)."
• Flag assumptions: If baseline data is estimated, say so and note what would change the projection.
• Prioritise people impact: Include employee satisfaction and change-adoption considerations in every recommendation.
• If scope is ambiguous: Ask one clarifying question before proceeding — do not assume the full enterprise scope.