Map a process before automating it: a practical guide
A six-step method to understand real work, make decisions explicit and choose a measurable first automation scope.
Automating a poorly understood process does not clarify it; it can accelerate its ambiguities. A useful map shows how work is actually done, which data moves, which decisions are made and where a person must intervene.
Define the start, end and expected value
Set the process boundaries before detailing its steps. A clear scope prevents a focused need from becoming a general transformation project.
- Triggering event
- Observable outcome
- Beneficiary person or service
Observe the real flow, not the ideal process
Interview the people involved and follow several representative cases. Record waits, rework and workarounds that do not appear in official procedures.
- Actions actually performed
- Handoffs between teams
- Waits, returns and duplicate entry
Inventory roles, tools and data
For each step, state who acts, where information lives and which permissions are required. This view reveals dependencies and sensitive access.
- Owner and contributors
- Applications and sources
- Read, write and retention
Make rules and exceptions explicit
Turn implicit habits into understandable decisions. Separate stable rules from cases that still require judgement.
- Decision criteria
- Missing or conflicting data
- Escalation and human approval
Measure the baseline
Choose a small set of indicators before any pilot. The comparison baseline should cover the same period and scope as the future test.
- Volume and frequency
- Lead time and human effort
- Errors, rework and exceptions
Choose a testable first scope
Start with repetitive, sufficiently stable and reversible steps. Document what remains under human control and the criteria to continue, revise or stop.
- Included and excluded steps
- Mandatory approvals
- Success, revision or stop
NOVAMIND
Map to choose, not to freeze
The map is not a final document. It creates a shared language for choosing a pilot, testing assumptions and improving the process from observed evidence.
General methodological framework: rules, measures, responsibilities and controls must be adapted to your organisation, country and sector.