Cadence Standard · Agent Readiness
Agent-Readiness
Before you switch an agent workforce on, know where you stand. Readiness measures one thing: how much of your business an agent could actually run today — scored across six dimensions, function by function.
An agent can only run what's legible to it. These are the six things that decide whether it can — each scored 0–4 with evidence.
1Process docs
Is the work written down, or does it live in people's heads?
0Tribal knowledge, nothing written.
4Every core process documented step-by-step.
2Data legibility
Is your data clean, labelled and in one findable place?
0Scattered across inboxes, spreadsheets, memory.
4Structured, consistent, single source of truth.
3Decision rules
Are the judgement calls written as rules, or "it depends"?
0Every exception is a person's gut call.
4Thresholds, tolerances and ownership are explicit.
4Access & tools
Can a system reach your tools through APIs or logins?
0Everything is manual, no connections.
4Core tools connect cleanly via API / MCP.
5Exception handling
When something's off-script, is the path defined?
0"Ask around until someone knows."
4Clear escalation: who, when, how — every time.
6Verification
Can you tell, fast, whether the work was done right?
0No checks; errors surface late or never.
4Every output has a check and a measurable target.
0–6
Tribal
The business lives in people's heads. An agent can't get a grip yet.
7–12
Fragile
Docs exist, but agents stall on the unwritten decisions. Most SMEs land here.
13–18
Legible
Readable enough that agents can run real work with a human on approvals.
19–24
Agent-Ready
You could switch agents on across several functions now, safely.
▸ The Agent Handover Test
The score tells you where you stand; the proof is the handover. Take one real process, hand it to a live agent cold, and count how many times a human has to step in. Zero interventions = agent-ready. It's the honest test behind the number.
Where a typical business tends to sit — and what usually has to change first. Your own scores will differ; this is the shape of it.
Rate each dimension 0–4. Your score, band and first move appear instantly — free, no sign-up, and the result is yours to download.
JM
John Marston · Cadence Standard
Operations & Lean · warehouse and SME process
I build the agent workforce you've just been talking to — and I put it through the same audit first. I ran it on my own automated business: it scored 11/24. Fragile. That's the point — you fix the gaps before you switch agents on, not after.