LOW / CODEBlueprints
generalai

AI Executive Report Generator

Aggregate data from multiple Google Sheets, then use Claude AI to synthesize insights into a polished executive summary with trends, anomalies, and recommendations. Automatically email to stakeholders and post highlights to Slack.

Advanced~30 minn8nMake.com 101 downloads 304 views

Setup Instructions

1. Identify the Google Sheets containing your data sources (revenue, KPIs, project status, etc.). Note each Sheet ID and the specific ranges to pull. 2. Create a new workflow triggered on a schedule (e.g., every Monday at 8 AM) or via manual webhook. 3. Add multiple Google Sheets "Read Rows" nodes — one per data source. Configure each with the appropriate Sheet ID and range. 4. Add a Code/Function node that aggregates all sheet data into a single structured JSON object with sections (e.g., { revenue: [...], kpis: [...], projects: [...] }). 5. Add an HTTP Request node to call Claude API (POST https://api.anthropic.com/v1/messages). Set headers: x-api-key: {{YOUR_ANTHROPIC_API_KEY}}, anthropic-version: 2023-06-01. In the body, use model "claude-sonnet-4-20250514", max_tokens: 4096. Craft your prompt: "You are an executive report writer. Analyze the following business data and generate a report with: 1) Executive Summary (3 sentences), 2) Key Metrics & Trends, 3) Anomalies & Concerns, 4) Recommendations. Data: [aggregated JSON]" 6. Add a Code node to parse Claude's response and format it as HTML for the email and plain text for Slack. 7. Add a Gmail node to send the HTML report to your stakeholder distribution list. Add a Slack node to post the executive summary and top 3 highlights to your #leadership channel. 8. Test with last week's data first. Iterate on the prompt until the report quality matches your expectations — add examples of good reports to the prompt for better output.
Troubleshooting
**Report is too generic or vague:** Your prompt needs more context. Include company-specific terminology, target ranges for KPIs, and what "good" vs "bad" looks like. Add a few sentences of context about your business so Claude can make relevant comparisons. **Data aggregation fails:** Ensure all Google Sheets nodes return data in a consistent format. Check for empty sheets or changed column names. Add error handling that skips empty sources rather than failing the whole workflow. **Claude response is too long for Slack:** Slack messages have a 4000-character limit per block. Parse Claude's response and only post the Executive Summary and Recommendations to Slack. Send the full report via email only. **Token limit exceeded:** If your aggregated data is very large, summarize each data source before sending to Claude. Use a two-pass approach: first have Claude summarize each section individually, then synthesize the summaries into the final report.

Need a custom version?

We can build a tailored automation workflow for your specific needs.

New blueprints weekly

Get notified when we publish new automation workflows.