Create intermediate Works with: Any AI

Newsletter Metrics Review and Next Experiments

Table output Has example

Read the numbers honestly and choose small tests.

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Review the performance of my newsletter. Metrics for the last [PERIOD]: [METRICS]. Goals: [GOALS]. Changes made recently: [RECENT_CHANGES].

Explain each metric in plain words: list growth, unsubscribe rate, click rate, reply rate and, with caution, open rate, noting that privacy features distort opens. Calculate the net growth and unsubscribe percentage from my figures, showing the arithmetic.

Identify three things the numbers suggest and, for each, an alternative explanation. Propose three experiments with a hypothesis, one change, a measure and a minimum run time. Say plainly that small samples cannot prove an effect. End with a one-line summary for my own records.

Add a monthly dashboard layout with six numbers and a line of commentary under each. Warn against changing too many things at once, since I cannot tell which change worked. Provide a record template for experiments with columns for date, hypothesis, change, result and decision. Remind me that percentages from lists under a thousand readers move around a lot from issue to issue.

Fill in before use

Optional: type your details and the prompt above updates. Nothing is saved or sent.

Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.

Invented figures: start of month 2,000 subscribers, 150 joined, 40 unsubscribed.
Net growth = 150 - 40 = 110, which is 110 / 2,000 = 5.5 percent.
Unsubscribe rate = 40 / 2,000 = 2.0 percent.
Click rate: 180 clicks from 1,500 delivered = 12 percent.
Reading: growth looks healthy, but unsubscribes may have followed the longer issue on the 14th. Alternative explanation: a small list clean-up.
Experiment: shorten one issue per month for two months; measure clicks and unsubscribes; small samples cannot prove an effect.

• analytics