Blank editorial worksheet · E94-NT-01

Creator campaign next-test worksheet

One worksheet covers one disappointing result and one next test. Use the blank CSV register if you keep several. The decision map shows the order.

Local blank template. No submit action, no saving, no example creator or result, and no benchmark. Every number you write here is yours, set by your team before launch. This is an editorial method, not statistical or legal advice.

A. Record what happened

Goal as written before launch, with the date you wrote it

Result as observed (number, post, date range)

Where each number comes from, who read it, and when (date, time, time zone)

B. Check the measurement

What could have kept the result from being recorded (link, consent, tagging, report delay, settings changed mid-campaign)?

Measurement state

Evidence for that state

C. Rival explanations

Explanation 1: what it predicts, what you hold, what is missing

Explanation 2: what it predicts, what you hold, what is missing

Explanation 3: what it predicts, what you hold, what is missing

D. One hypothesis, one change

If we change (A) to (B) and hold the rest the same, then (metric) will (direction) by (read date)

The one change, from A to B

Held constant: creator, offer, destination, window, measurement settings

E. Pre-state the read, then freeze it

Metric, source and window

Read date (date, time, time zone)

Owner of the read

Result that supports the hypothesis

Result that contradicts it

Result that is inconclusive, including no data by the read date

What you will do in each case

Frozen at (date, time, time zone)