Process hierarchy levels: how to structure a process inventory

You structure a process inventory by sorting work into three levels. L1 names broad process areas. L2 breaks each area into processes with a start, an end and an output. L3 lists the tasks inside them. Each task records its role, applications, business rules and handoffs, so your team can compare work and plan change.

• Process hierarchy levels break work down from broad process areas at L1 to individual tasks at L3.
• Build the L1 frame top down, then fill L2 and L3 with the people who do the work.
• Each L3 task should record its performing role, applications, business rules and handoffs.
• Consistent depth and named owners matter more than the amount of detail.

What are process hierarchy levels?

Process hierarchy levels are tiers that break your organization's work down from broad process areas to individual tasks. Most process inventories use three levels, labeled L1 to L3, so every team describes work at a comparable depth.

A process inventory is a structured catalog of the processes your organization runs. Without process hierarchy levels, a process inventory mixes large value chains with small clerical steps, and nobody can compare them.

Process hierarchy levels give your team a shared vocabulary. When a finance lead and an IT lead both say level 2, they mean the same depth of detail. That shared depth makes ownership, cost and change planning easier to discuss.

Process hierarchy levels also make a process inventory easier to navigate and summarize. Your team can roll volumes and costs up from L3 tasks to L2 processes and L1 areas.

What belongs at L1, L2 and L3?

Each process level answers a different question about the work. L1 names what the business does, L2 names how the business does it, and L3 names who does which task with which tools.

L1: process areas

L1 holds the broad process areas of the enterprise, such as order to cash, procure to pay or hire to retire. An L1 process area usually crosses several departments and has an executive owner.

L2: processes

L2 splits each L1 area into processes with a clear start, a clear end and a defined output. Invoice processing, supplier onboarding and payment runs are typical L2 processes under procure to pay.

L3: tasks and activities

L3 describes the tasks inside each L2 process. An L3 task is done by a specific role, in specific applications, under specific business rules. L3 is where work becomes concrete enough to judge for AI agents.

A simple test helps you place an entry in the process hierarchy. If the entry has its own owner and output, it is likely L2. If a role completes it in a sitting, it is likely L3.

How do you build a process inventory in levels?

You build a process inventory top down for structure and bottom up for accuracy. Start with a fixed L1 frame, then fill L2 and L3 with the people who do the work.

1. Agree on the L1 frame with leadership. Keep the list of process areas short, stable and owned by named executives.
2. Break each L1 area into L2 processes. Give every L2 process a start trigger, an end state, an output and a process owner.
3. Interview and observe the teams that run each L2 process. Capture the L3 tasks in the order people actually perform them.
4. Standardize naming. Use a verb plus an object for every entry, such as approve invoice or validate supplier.
5. Review the inventory with process owners. Fix gaps, duplicates and entries placed at the wrong level.

Short workshops with process owners work well for L1 and L2. L3 usually needs time with the people who perform the tasks, because they know the exceptions and workarounds.

The goal at this stage is as-is process documentation: a record of how work runs today, not how a policy says work should run.

Resist designing the future state while you document the process inventory. An inventory that blends current and target work hides the gaps that your program needs to close.

What should each L3 task record?

Each L3 task in a process inventory should record enough detail to judge the task without a follow-up meeting. Teams often capture a standard set of attributes for every L3 task.

• Performing role: the job role that does the task, not a named person.
• Applications: the business systems and desktop tools the task touches.
• Business rules: the conditions, thresholds and exceptions that govern the task.
• Handoffs: where the task receives work from, and where the task sends work next.
• Inputs and outputs: the documents or data the task starts from and produces.
• Owner: the person accountable for keeping the entry current.

Handoffs deserve special care in a process inventory. Delays, rework and errors often cluster where work passes between roles or teams, so skipping handoffs hides where processes break.

Consistent L3 attributes let your team compare tasks across different L1 areas. Similar tasks with similar rules often turn up in several process areas.

What mistakes weaken a process hierarchy?

Most weak process hierarchies fail on consistency rather than effort. The same few mistakes appear across operations, finance and IT teams.

• Mixed depth: some L2 entries are really L3 tasks, and some L3 entries are whole processes.
• Organization charts as L1: process areas should follow the flow of work, not reporting lines that change with each reorganization.
• Too much detail: listing screen-level steps at L3 makes the inventory hard to read and hard to maintain.
• No owners: entries without an accountable owner go stale quickly.
• Missing exceptions: documenting only the standard path hides the business rules that make a task hard.

A process hierarchy also needs a maintenance rhythm. Process owners should review their L2 and L3 entries whenever systems, policies or teams change.

Teams planning an agentic workforce transformation depend on that discipline, because AI agents need tasks with clear roles, inputs, rules and handoffs.

Documenting process levels in Via

In Via, your team documents processes at up to three levels (L1 to L3), and each task's performing role, applications and business rules. Via also records the handoffs between tasks. Process modeling in Via is done by your team.