Create intermediate Works with: Any AI

Ad Performance Diagnosis From a Weekly Export

Table output Has example

Find why results changed using the numbers you have.

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

Diagnose my ad results. Export of last two weeks by campaign (spend, impressions, clicks, conversions): [AD_DATA]. Changes made in that time: [CHANGES]. Target cost per conversion: [TARGET_COST].

Calculate click-through rate as clicks divided by impressions, conversion rate as conversions divided by clicks and cost per conversion as spend divided by conversions for each week, showing the arithmetic. Identify which step of the funnel changed most.

List three possible causes for each change, how to check each, and the cheapest test. Do not recommend changes based on fewer than 30 clicks per item without saying it is noisy. Return a table and a prioritised action list for the week.

Show the weekly calculation as a mini table with invented numbers. Week one: 10,000 impressions, 200 clicks, 8 conversions, spend 240. Click-through rate = 200 / 10,000 = 2.0 percent; conversion rate = 8 / 200 = 4.0 percent; cost per conversion = 240 / 8 = 30. Week two: 9,000 impressions, 135 clicks, 6 conversions, spend 210. Click-through rate = 1.5 percent; conversion rate = 4.4 percent; cost per conversion = 35. The drop in click-through rate suggests the ad or audience changed, not the page. Apply the same method to my data and list changes made in the period that could explain it.

Add a decision list for the week: three actions in priority order, each with the expected effect and a date to check it. Include a section on what not to change this week, because changing many things at once makes the next diagnosis impossible. Add a reminder to compare with the same weekdays, since weekends often behave differently, and to check whether tracking was working throughout the period before drawing conclusions.

Close with a note of the single change to make this week, how it will be measured and the date to review it.

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 data. Week 1: 10,000 impressions, 200 clicks, 8 conversions, spend 240. CTR 2.0 percent; conversion rate 4.0 percent; cost per conversion 30.
Week 2: 9,000 impressions, 135 clicks, 6 conversions, spend 210. CTR 1.5 percent; conversion rate 4.4 percent; cost per conversion 35.
Finding: click-through fell; conversion held, so the page is probably fine.
Causes to check: ad copy changed on day 9; audience narrowed; a competitor appeared.
Cheapest test: restore the old headline for half of impressions. Both weeks have fewer than 30 conversions, so treat as noisy.

• ads