Is your site agent-ready? →

Home › Tools › The Continuity Runbook Check

Free check · 7 items · printable register

How many people could pick this up on Monday?

Seven items and four numbers return a one-page register you can take into a planning meeting — key-person exposure, coverage ratio, and how much of a publication batch your evaluation window would misclassify.

7 items Printable register Corpus of 29 postings No email

No email required. Nothing is sent anywhere — your answers stay in this browser.

0 of 7 answered

The Continuity Runbook Check

Answer for the programme as staffed today. The output is a register: print it and take it to the planning meeting. “Not sure” scores as no.

Item 1

Can you produce, today, a written decomposition of your lead search role into its individual specialisms?

Technical, content, analytics, AI visibility, schema, engineering liaison — named separately.

Item 2

Does every one of those specialisms have a second named person who could pick it up on a Monday?

A name, not a team. “The agency” is not a name.

Item 3

Does your supplier contract require the decision log, query and prompt history, instrument inventory and citation-history export as deliverables at renewal?

The apparatus, not the deck. If it leaves with the supplier, you were renting the measurement.

Item 4

Do you have a retrieval baseline — which pages AI engines cite, and which passages they quote — taken in the last 90 days?

Without a baseline, no future change can be shown to have worked.

Item 5

Does your monthly report carry impressions, clicks and citations as three separate lines with separate owners?

Three behaviours that now move independently.

Item 6

Is your evaluation window for newly published pages 45 days or longer, with a written pending-not-failing exception?

The exception matters: it is what stops a page being killed at day 30.

Item 7

Does a named person run a weekly non-zero check on every measurement instrument, with a dated log?

This is the item that caught us. An instrument that returns zeros looks exactly like a result.

Your register (optional)

These four numbers produce the printable register: exposure, coverage and the days your window would misclassify.

Your result
Read the method
What you are being scored against

Our own numbers, from a corpus of 29 public postings and our own estate

What employers ask one person to carry, read from public job requirements. It shows what employers ask for, not what the market pays. No employer is ever named.

MeasuredWhat it is
median 8, range 4–12Specialisms demanded of one lead search hire17 of the 29 postings demand 8 or more. 10 of those 17 name no team at all. · Indexable job-req corpus, n = 29 public postings, June–August 2026
$176K–$221KMedian advertised band for the equivalent internal hire16 of the 29 postings disclosed a band. Public postings only, mixed geographies and seniorities: it shows what employers ask for, not what the market pays. No employer is ever named. · Indexable job-req corpus, n = 29 public postings, June–August 2026
median 6.81 days, P90 37.10 daysHow long citations lag publicationThis is the reason a 45-day evaluation window beats a 30-day one: at 30 days you are reading the lag, not the result. Third-party measurement, not ours — and we hold it second-hand, so treat it as directional. · Profound, 2026, as reported in LinkedIn's 2026 AI Search Visibility guide

The corpus is self-selected: public postings only, mixed geographies and seniorities. It shows what employers ask for, not what the market pays. Our own first run scored 2 of 7 — item 7 was the one that caught us.

The register is the artefact. The contract clause is the fix.

Two of these items need somebody else's signature. Both are one sentence: the renewal deliverables in item 3, and the pending-not-failing exception in item 6.

Get an AI search audit

Copied ✓