Workflow guides

Automation decision guide

How to choose your first AI automation

Start with one task your team repeats. Agree on its inputs, the person responsible, and what a good result looks like. Automate the parts you can test. Keep decisions that need judgment with a person, and record what happened and what still needs attention.

Good fit
Use automation when the work repeats often enough to observe, the source systems are accessible, and a responsible operator can define success, exceptions, and the stop condition.
Pause when
Delay automation when every case is novel, the source records are unreliable, the team cannot name a decision owner, or a mistaken action would be hard to detect and reverse.

When this approach helps

  • People copy the same context between two or more systems on a regular cadence.
  • A queue grows because evidence gathering takes longer than the actual decision.
  • The team can describe a passing case, an exception, and the person who resolves that exception.

Follow these steps

  1. 01

    Choose one repeated task

    Follow a task from its starting event to its expected result. Measure volume, delays, rework, and exceptions before changing it.

  2. 02

    Separate drafts from decisions

    Identify steps that gather facts, organize records, or draft a response. Separate them from spending money, publishing, changing access, or interpreting policy. Start with preparation when decision quality is unproven.

  3. 03

    Test everyday and difficult cases

    Include ordinary cases, unusual cases, denied actions, missing inputs, and duplicates. Compare results with source records. Check that the agent asks for help when information is missing.

  4. 04

    Agree on the launch checks

    Name who can enable live actions. Define monitoring, reasons to pause, recovery, and record retention. Make launch an explicit decision after the checks pass.

Documents and records to keep

Workflow map

The map names the initiating signal, source records, decisions, owners, tool boundaries, exception paths, and proof expected at completion.

Case fixture pack

A reusable set of representative cases makes quality discussions concrete and prevents a polished demo from substituting for operational coverage.

Operating runbook

The runbook explains release state, approvals, monitoring, pause and rollback actions, escalation contacts, and the evidence needed for the next review.

Common questions

Which AI workflow should a company automate first?

Choose a frequent, bounded handoff with accessible records, a known owner, measurable delay or rework, and reversible actions. Avoid beginning with the company’s highest-stakes judgment simply because it sounds strategically important.

Does AI workflow automation require an autonomous agent?

No. Many valuable systems watch for a change, assemble evidence, draft a recommendation, and wait for a human decision. Autonomy is an authority choice, not a requirement for using models or tool connectivity.

When is an AI automation pilot ready for production?

It is ready only after representative cases pass, identity and tool permissions are bounded, exception ownership is staffed, monitoring and rollback exist, and the accountable operator explicitly accepts the remaining risk.

AI workflow system

Find the task where AI could help.

Start with one repeated task. Agree on what AI can do, who checks its work, and how to measure the result.

Verified field result

Automation prepared the evidence. Human judgment still decided.

The system gathered evidence for reviewers. A person still decided whether to approve or reject each submission.

49/ 50
49 of 50 selected cases produced usable evidence packets for human decision.
Workflow
Marketplace template review
Receipt
#FR-2026-01
Verified
May–June 2026
Changes to external systems
0
  1. Map the frictionList the tools, steps, and person responsible.
  2. Test one pathTest the routine work and check that approvals still work.
  3. Keep the proofReview what ran, what stopped, and what needs recovery.