thinkhard.SI

Prompt repository Marketing

Report campaign results to your manager

You need a short readout of what a campaign achieved, why, and what you will change next time.

Registry no.
MKT-019
Collection
Marketing
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
6
Added

MarketingAny chat

The prompt

Score 10 of 10, Excellent

Use it when You need a short readout of what a campaign achieved, why, and what you will change next time.

  • {{campaign}}
  • {{targets}}
  • {{manager}}
  • {{next_decision}}
  • {{results}}
  • {{context}}
Full prompt and why it works
Prompt
Background:
Campaign (name, goal, dates, channels, budget): {{campaign}}
What success was defined as before launch: {{targets}}
Who reads this and what they care about: {{manager}}
The decision or next campaign this should inform: {{next_decision}}

The results by channel (spend, reach, clicks, leads, sales, whatever we track), with where each number comes from:
"""
{{results}}
"""

What happened during the campaign (changes we made, problems, outside events):
"""
{{context}}
"""

Task: Write a short readout of the campaign's results for my manager, under 300 words plus one table.

Use these headings:
1. Bottom line: one or two sentences on whether we hit the target, with the main number.
2. Results against target: a table with the columns Measure, Target, Actual, Difference. Add the previous comparable campaign only if the results include it.
3. What worked and what did not: two or three points each, tied to numbers.
4. Why: our best explanation, labeled "likely" or "possible", kept apart from what the numbers show.
5. Next time: three specific changes, each tied to a finding.
6. Decision needed: what I am asking for, if anything.

Rules:
- Judge the campaign against the targets set before launch. Do not swap in a different measure because it looks better.
- Show the arithmetic for any rate or cost per result you calculate, and check that the numbers add up.
- Put business results (leads, sales, revenue) ahead of reach and likes.
- Where the numbers are small, say the result points in a direction rather than proves anything.
- Report misses plainly. A manager trusts a readout that admits what did not work.

Example of a "Why" line, for form only: "Likely: the email outperformed because it went to past buyers, while the ads reached people new to us."

If a number is missing, or two sources disagree (for example the ad platform and our sales records), say so in the table rather than picking one or filling the gap.

Variables

{{campaign}}
The campaign's name, goal, dates, channels and budget.
{{targets}}
What you said success would be before launch, for example "150 leads at under $40 each".
{{manager}}
Who reads it and what they care about most, for example "my manager, who watches cost per lead".
{{next_decision}}
What this should inform, for example "whether to rerun it in spring with the same budget".
{{results}}
The numbers by channel, with the source of each (ad platform, email tool, CRM, sales records).
{{context}}
What happened during the campaign: changes, problems, outside events. Write "nothing notable" if so.

Why it works

  • What good looks like. Results are judged against the targets set before launch, with business results ahead of likes, so the readout cannot quietly move the goalposts.
  • Format. Bottom line first, then a target-against-actual table, then reasons and changes: the order a manager reads in when they have two minutes.
  • If unsure. Missing numbers and sources that disagree are shown as such, and causes are labeled likely or possible rather than stated as fact.

How to use it

  1. Press Fill in variables and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/campaign-results-readout

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "campaign-results-readout"}, or just ask for it by name.