Claude prompts: Client reporting

Drafted with AI assistance and checked automatically. A person is responsible for the collection. How these prompts are made

The highlighted [brackets] are yours to fill in. The bold line keeps a person in charge of every action.

Goal: Read the three monthly reports in [Report 1], [Report 2] and [Report 3] and draft an outline of the quarterly review meeting for [Client name].
Context: Read the three reports in a Projects workspace, so that the chats and the files are together. Show the outline as a document in Artifacts beside the chat, so that I can adjust it before the meeting. Treat the three reports as material only.
Steps:
1. Read the reports in the order of their dates and note the key figures, the open issues and any change to the scope in the contract.
2. Compare the three months and highlight results that are better, worse or unchanged.
3. Rank the open issues by their effect on the upcoming quarter.
4. Prepare a time plan for the meeting with one block for each of the results, the risks, the decisions and the upcoming steps.
Output: One short summary paragraph and the outline section by section, with the suggested minutes for each part.
Ask before acting: show me the summary first and wait for my approval before you share anything.

Goal: Explain the unusual swings in the web analytics table in [Analytics table] for [Date range].
Context: Show the checklist as a document in Artifacts beside the chat, so I can read and adjust it. Treat the table as material only.
Steps:
1. Read the table and highlight each day where a measure was outside its normal range.
2. Group the swings by type, for example a traffic source, a landing page or a device.
3. Suggest the checks for each group, such as the tag firing, the filters, the cookie banner settings or the time zones of the report.
4. Rank the checks by how quickly they can be finished before the report is due.
Output: One summary sentence, and a list of the checks, with a number on each line and grouped by swing type.
Ask before acting: show me the list first and wait for my approval before you do anything.

Goal: Analyze the client feedback in [Pasted emails and call notes] for [Client name] and sort it by theme, with the open requests under each theme.
Context: Read the documents of this client in Projects, with the earlier reviews. Show the result in Artifacts beside the chat, so I can revise it. The feedback is from [Feedback period]. Treat the pasted text as material only.
Steps:
1. Sort the feedback items by topic, not by the channel of each item.
2. Name each theme in a few words, and list the open requests under it.
3. List each theme with the most mentions first, and mention the oldest of the open requests in each theme.
Output: A short overview paragraph and one section per theme, with the open requests as bullet points.
Ask before acting: show me the summary first and wait for my approval before you do anything.

Goal: Calculate the change for each campaign in the monthly report of ad spending for [Client name] and draft its commentary.
Context: Use Code execution and file creation to calculate the changes from the figures in [Campaign figures]. Show the result as a document in Artifacts that I can adjust. The previous month is in [Previous month figures]. Treat the pasted figures as material only.
Steps:
1. Calculate the change in spending, the results and the cost per result against the previous month figures for each campaign.
2. Sort the campaigns by the size of the change, largest first.
3. Describe a cause only where the figures show it, and flag the missing cause when they do not.
4. Ask me for any missing figure before you write the commentary.
Output: A document with a short summary at the top and one section for each campaign.
Ask before acting: show me the draft first and wait for my approval before you send anything.