Skip to content
KatafactsBeta

Visual management

Go-see protocol

Go-see protocol is licensed CC BY 4.0. Attribution: Katafacts (katafacts.com).

Customise with AISkill file ↓

1 · What it is

What it is

A go-see protocol is a gemba walk with real technique behind it — the work instruction shape (major step, key point, reason) applied to observation instead of a task. Describe the walk roughly, and real steps get drafted: what to check, how to observe without distorting what you see, and why each technique matters.

2 · When to use it

When to use it — and when not to

Use it when

  • You want to verify that a visual board or training matrix reflects what's actually happening, not just what's documented.
  • A new leader or team member needs a real, repeatable way to observe the work, not just wander over and watch.
  • You want to catch drift between the plan and reality before it becomes a real problem.

Not when

  • The observation is genuinely a one-off, not a repeatable practice — a work instruction adds structure a single visit doesn't need.
  • You'd be observing to catch someone doing something wrong rather than to understand the real work — that's not what a go-see walk is for.

3 · How to fill it in

How to fill it in

Purpose
What does this walk verify, and what would go unnoticed without it?
Steps, key points & reasons
Drafted from your description of the walk — major steps, each with zero or more key points and, for every key point, the reason it matters.
Narrative
Drafted from the steps actually built.

4 · What good looks like

What good looks like

The example below is a go-see walk that checks the renewal-risk queue's visual board and ILUO matrix against what's actually happening on the floor — three of four steps have a real key point, and the fourth is honestly left flagged rather than padded.

Same example, as a downloadable xlsx workbook.

Download .xlsx

Work instruction

Go-see walk — renewal-risk queue

Support — renewal-risk queue

Priya Nair · 2026-10-09

Purpose

Verifies that what the visual board and training matrix claim is actually true on the floor — catches drift between what's documented and what's really happening before it becomes a real problem.

Steps, key points & reasons

4 steps · 3 key points

1. Stand at the point of work without interrupting

  • Watch a full ticket-handling cycle before asking any questionsInterrupting early means you see the explanation, not the actual work.

2. Compare what the board shows against what's actually happening

  • Check the KPI tree summary's numbers against the live dashboard, not from memoryA board that's a day stale looks fine at a glance but hides real drift.

3. Ask the person doing the work what's slowing them down

  • Ask open questions, not leading onesA leading question gets confirmation, not a real answer.

4. Note any gap between the ILUO matrix and who's actually handling each ticketNo key point yet

Narrative

Three of four steps have a real key point grounding the technique in a specific reason. The fourth, noting gaps between the ILUO matrix and actual handling, is still just an action with no key point yet — worth adding once there's a clearer sense of what to specifically look for.

5 · Common mistakes

Common mistakes

  • Asking questions before watching a full work cycle.

    An early question changes what you see — you get the explanation, not the real work.

  • Comparing the board against memory instead of the live data.

    A board that's a day stale still looks fine at a glance — checking against live numbers is what actually catches drift.

6 · What it connects to

What it connects to

upstream

  • Visual board design guide

    A go-see walk exists to verify the board's own claims against reality.

  • ILUO training matrix

    Checking who's actually handling the work against who the matrix says is trained is a natural go-see step.

downstream

    7 · Where AI helps

    Where AI helps

    Judgement — stays yours

    • Deciding whether a step genuinely needs a key point or is simple enough not to
    • Deciding what a drifted observation actually means for the team

    Analysis — AI helps

    • Drafting major steps, key points, and reasons from a description of the walk
    • Drafting the narrative from the steps actually built

    Drudgery — automated

    • Counting steps and key points
    • Identifying which steps still have no key point
    • Exporting to xlsx in the house format

    9 · Rate this kata

    Rate this kata