30 Prompts for Workflow Automation: Make, n8n, and Zapier

## Introduction Workflow automation has moved from a nice-to-have to a must-have for modern businesses. Whether you're a solo founder, a marketing manager, or a DevOps engineer, tools like Make, n8n, and Zapier let you connect apps, move data, and trigger actions without writing code from scratch. But the real magic isn't in the tools themselves—it's in the prompts you use to design and debug those automations. A well-crafted prompt can turn hours of manual setup into minutes of configuration. It can help you build error-handling logic, parse complex data, or even generate entire workflows from a simple description. In this guide, I'll share 30 ready-to-use prompts for Make, n8n, and Zapier, organized by use case. Each prompt includes an explanation of what it does, when to use it, and a realistic example. No fluff, just actionable templates you can copy and paste today. ## Why Prompts Matter for Automation Automation platforms have become powerful, but their complexity has grown too. A typical Make scenario might involve dozens of modules, filters, aggregators, and error handlers. Writing a prompt that explains your intent clearly can save hours of debugging. Prompts act as a bridge between your business logic and the tool's configuration—they help you think through the steps, anticipate edge cases, and document your process. For example, instead of manually setting up a Zap that sends a Slack message when a new lead comes in, you can use a prompt to generate the exact steps, including conditional logic for lead score thresholds. This not only speeds up development but also makes your automations more reliable. ## Section 1: Prompts for Scenario Design (Make) ### 1. Build a Lead Scoring Automation **Task:** Create a scenario that assigns a lead score based on form submissions and sends high-scoring leads to a CRM. **Prompt:** "I need a Make scenario that watches a Google Sheets row for new leads. For each new row, check if the 'Company Size' field is > 50 AND the 'Budget' field is > $10,000. If both conditions are true, update the lead score to 'High' and create a deal in HubSpot. If not, update the score to 'Low' and send a notification to a Slack channel. Include error handling for missing fields." **Example:** A B2B SaaS company uses this to prioritize enterprise leads. The scenario runs every 5 minutes, scores leads, and pushes high-value ones straight to the sales pipeline. ### 2. Parse JSON from a Webhook **Task:** Extract specific fields from a complex JSON payload received via webhook. **Prompt:** "I'm receiving a JSON payload from a custom webhook at endpoint /incoming. The payload has nested objects: { 'user': { 'name': 'John', 'email': 'john@example.com' }, 'order': { 'id': 'ORD-123', 'total': 250 } }. I need to extract the user's email and the order ID, then log them to a separate Google Sheet. Handle cases where the JSON might be malformed." **Example:** An e-commerce store uses this to capture order confirmations from a custom checkout system and log them to a sheet. ### 3. Create a Daily Digest Email **Task:** Aggregate data from multiple sources into a daily summary email. **Prompt:** "Every day at 8 AM, I want to collect new records from a MySQL database (table: orders), count the number of orders, sum the total revenue, and also pull the top 3 most recent customer support tickets from Zendesk. Then compose an HTML email with these metrics and send it via Gmail to admin@company.com. Use Make's aggregator to combine the two data streams." **Example:** A support team manager gets a daily digest without logging into multiple dashboards. ### 4. Sync Contacts Between Two CRMs **Task:** Keep contacts updated between Salesforce and HubSpot. **Prompt:** "I need a scenario that runs hourly. It should search for contacts in Salesforce that have been modified in the last hour. For each modified contact, check if they exist in HubSpot by email. If they do, update the HubSpot record. If not, create a new one. Log any errors to a dedicated 'sync errors' sheet." **Example:** A sales team using both CRMs avoids duplicate data entry. ### 5. Automate Invoice Generation **Task:** Generate an invoice PDF when a payment is received in Stripe. **Prompt:** "When a Stripe payment succeeds (event: charge.succeeded), capture the customer email, amount, and description. Use a Google Docs template to generate an invoice, convert it to PDF, and email it to the customer. Include the invoice number as a sequential counter stored in a data store." **Example:** A freelancer automatically sends professional invoices after each client payment. ## Section 2: Prompts for Workflow Logic (n8n) ### 6. Conditional Routing Based on Data **Task:** Route incoming data to different branches depending on a field value. **Prompt:** "I have an n8n workflow that receives orders from a Shopify webhook. If the order total is over $500, route it to a 'VIP processing' branch that sends a Slack alert and adds a tag in Shopify. If under $500, route to 'standard processing' that just updates a Google Sheet. Include a fallback for missing total." **Example:** An online store prioritizes high-value orders for faster fulfillment. ### 7. Error Recovery with Retries **Task:** Retry a failed API call with exponential backoff. **Prompt:** "I'm using an HTTP Request node to call a third-party API that occasionally returns 429 (Too Many Requests). Configure a retry mechanism: wait 5 seconds on first failure, 15 on second, 30 on third. If it still fails after 3 retries, send an email to the admin and log the error to a separate database." **Example:** A data integration pipeline that pulls weather data from an external API stays resilient. ### 8. Code Node for Data Transformation **Task:** Transform a CSV string into a structured JSON array. **Prompt:** "I have a CSV string as input: 'name,email,score\nAlice,alice@example.com,85\nBob,bob@example.com,92'. Use a Code node (JavaScript) to parse this into an array of objects. Then filter out rows where score < 80 and pass only the remaining rows to the next node. Handle empty lines." **Example:** A marketing team imports survey results from a CSV file attached to an email. ### 9. Merge Two Data Streams **Task:** Combine data from two different APIs based on a common key. **Prompt:** "I have two webhooks. Webhook A sends user profile data (userId, name, email). Webhook B sends order data (userId, product, amount). I need to merge these two streams on userId using a Merge node (join type: inner). Output a single object per user with profile and order info. Handle cases where userId is missing." **Example:** A customer support agent sees both profile and recent order history in one view. ### 10. Schedule and Send Reports **Task:** Generate a weekly report from a database and email it. **Prompt:** "Every Monday at 9 AM, run a SQL query on PostgreSQL to get last week's sales, grouped by product category. Format the result as an HTML table. Then use an SMTP node to send it as an email to the management team. Add a line chart image generated from the data using a Code node." **Example:** A sales manager receives a formatted weekly report without manual work. ## Section 3: Prompts for Multi-Step Zaps (Zapier) ### 11. Capture Form Submissions to CRM **Task:** When a Typeform submission comes in, create a contact in Salesforce and send a confirmation email. **Prompt:** "When a new entry is submitted in Typeform, create a new contact in Salesforce with the fields: First Name, Last Name, Email, Phone, and Company. Then send a confirmation email via Gmail to the submitter with a custom message based on the form's 'interest' field. If the interest is 'Demo Request', include a link to schedule a demo." **Example:** A lead capture form on a landing page automatically enters the CRM and triggers a personalized response. ### 12. Social Media Cross-Posting **Task:** Automatically share new blog posts from RSS to LinkedIn and Twitter. **Prompt:** "Monitor an RSS feed from my blog. When a new post appears, extract the title, URL, and a short excerpt. Post the title and link to LinkedIn (with a custom image if available). Also post a shorter version to Twitter (now X) with the link. Add a filter to skip posts that contain the word 'internal'." **Example:** A content marketer maintains a consistent social media presence without manual posting. ### 13. Invoice Payment Follow-Up **Task:** Send a reminder email when an invoice in Xero becomes overdue. **Prompt:** "When an invoice in Xero has its status changed to 'Overdue', check the invoice date. If it's been overdue for more than 7 days, send a reminder email to the customer with the invoice details and a link to pay. If it's been overdue for more than 30 days, also create a task in Trello for the collections team." **Example:** A small business reduces late payments with automated reminders. ### 14. Slack Notifications for Critical Alerts **Task:** Forward specific PagerDuty alerts to a Slack channel. **Prompt:** "When a PagerDuty incident is triggered with severity 'critical', post a message to a dedicated Slack channel (#alerts) with the incident title, description, and a link to the PagerDuty dashboard. Include the assigned responder's name. If the incident is acknowledged, post a follow-up message." **Example:** An engineering team gets real-time alerts without checking multiple dashboards. ### 15. E-commerce Order Fulfillment **Task:** When a Shopify order is paid, send order details to a fulfillment service. **Prompt:** "When a Shopify order is marked as 'paid', extract the order ID, customer name, shipping address, and line items. Send this data to a custom webhook URL for my fulfillment partner. Also update a Google Sheet row with the order status as 'Sent to fulfillment'. If the webhook fails, log the error." **Example:** An online store streamlines order processing from payment to shipping. ## Section 4: Prompts for Error Handling & Logging ### 16. Centralized Error Log **Task:** Log all automation errors to a single database or sheet. **Prompt:** "For every automation I build, I want a generic error handler that catches any module failure. Capture the error message, the module name, the input data (sanitized), and a timestamp. Write this to a central 'error_log' table in Airtable. Send a daily summary of errors to the admin." **Example:** A team maintains a single source of truth for debugging all automations. ### 17. Graceful Degradation **Task:** If an API is down, use cached data instead. **Prompt:** "When calling the Weather API, if a 500 error occurs, instead of failing, use the last successful response stored in a data store. Add a flag 'cached: true' to the output so downstream modules know the data is not fresh. If the cache is empty, send an alert." **Example:** A dashboard that shows current weather always has data, even if the source is temporarily unavailable. ### 18. Human-in-the-Loop Approval **Task:** Pause automation until a human approves a critical action. **Prompt:** "When a Zap detects a new order over $10,000, instead of processing it automatically, send a Slack message to the finance team with the order details. Wait for a Slack reaction (thumbs up) to proceed. If no reaction within 30 minutes, escalate via email." **Example:** A high-value order requires manual review before fulfillment. ## Section 5: Prompts for Data Transformation & Parsing ### 19. Extract Email Attachments **Task:** Save email attachments to a cloud storage folder. **Prompt:** "When a new email arrives in Gmail with the subject line 'Invoice', download all PDF attachments. Rename them using the pattern: Invoice_[date]_[sender].pdf. Save them to a Google Drive folder named 'Invoices'. Then create a row in a Google Sheet with the filename and date." **Example:** An accounting department automatically archives vendor invoices. ### 20. Parse Web Scraped Data **Task:** Extract structured data from an HTML page. **Prompt:** "I have an HTML string from a web scraper. Use a Code node in n8n to parse the HTML and extract all tags and their following tags. Output an array of objects: { heading: '...', paragraph: '...' }. Ignore empty tags." **Example:** A research team collects article summaries from a list of URLs. ### 21. Convert Currency with Live Rates **Task:** Convert amounts from one currency to another using an exchange rate API. **Prompt:** "When a new order comes in with a 'currency' field (e.g., 'EUR') and 'amount', call the ExchangeRate-API (free tier) to get the latest USD rate. Multiply the amount by the rate and add a new field 'amount_usd'. If the API call fails, use the last cached rate from a data store." **Example:** A global e-commerce store tracks revenue in a single currency. ## Section 6: Prompts for Integration Specifics ### 22. Telegram Bot for Order Updates **Task:** Send order status updates to a Telegram group. **Prompt:** "When a Shopify order status changes to 'shipped', send a Telegram message to a group chat with the order ID, customer name, and tracking number. Use the Telegram Bot API (sendMessage endpoint). Include a button with the tracking URL if available." **Example:** A fulfillment team gets instant notifications on mobile. ASI Biont supports connecting to Telegram via API — for more details, check asibiont.com/courses. ### 23. Google Analytics Event from Form Submission **Task:** Log a custom event in Google Analytics 4 when a form is submitted. **Prompt:** "When a new row appears in a Google Sheet that tracks form submissions, send an event to Google Analytics 4 via the Measurement Protocol. Include parameters: event_name = 'form_submission', form_id, user_email (hashed). Use a custom HTTP request module." **Example:** A marketing team tracks form conversions without adding JavaScript to the site. ### 24. Salesforce Lead from LinkedIn Ad **Task:** Capture LinkedIn Ad leads and push them to Salesforce. **Prompt:** "When a LinkedIn Lead Gen Form submission comes in via webhook, map the fields (first name, last name, email, company, job title) to Salesforce Lead fields. Create the lead. Then add a note to the lead with the ad campaign name and timestamp." **Example:** A B2B company automates lead intake from LinkedIn Ads directly into their CRM. ## Section 7: Prompts for Advanced Patterns ### 25. Dynamic Workflow Chaining **Task:** Trigger a different workflow based on the output of the first. **Prompt:** "After a Make scenario completes, pass the output (a JSON object with a 'status' field). If status = 'success', call a webhook to trigger a second scenario that sends a confirmation. If status = 'failed', call a third scenario that logs the error. Use Make's webhook response module to return a status." **Example:** A multi-step order processing pipeline handles success and failure separately. ### 26. Data Deduplication **Task:** Prevent duplicate entries when syncing data. **Prompt:** "Before creating a contact in HubSpot from a new form submission, search HubSpot for an existing contact with the same email. If found, update the existing contact's fields instead of creating a duplicate. If not found, create a new one. Log the action (created vs updated). " **Example:** A lead generation system maintains clean data even with multiple form submissions from the same person. ### 27. Batch Processing with Throttling **Task:** Process a large batch of records without hitting API rate limits. **Prompt:** "I have a list of 10,000 email addresses in a CSV. I need to check each against a third-party email verification API that allows 100 requests per minute. Use a loop with a 600ms delay between each request. If a request fails with 429, wait 60 seconds and retry. Write the results to a new CSV." **Example:** A marketing team cleans their email list before a campaign. ## Section 8: Prompts for Testing & Debugging ### 28. Dry Run Mode **Task:** Test an automation without making real changes. **Prompt:** "Add a toggle variable 'DRY_RUN' at the start of the scenario. If true, instead of actually creating/updating records in external services, just log the action to a debug sheet and stop. This allows testing without side effects." **Example:** A developer validates a new automation before enabling it in production. ### 29. Debug Logging with Context **Task:** Log detailed debug information for each step. **Prompt:** "After every module in the workflow, capture the input and output (truncated to 500 chars) and write it to a debug log. Include the module name, timestamp, and a unique run ID. Only enable this logging if a variable 'DEBUG' is set to true." **Example:** A support engineer investigates a rare bug by enabling debug mode for specific runs. ### 30. Simulate Webhook Payload **Task:** Generate a test payload to manually trigger a workflow. **Prompt:** "Create a simple script or module that generates a sample webhook payload matching the expected schema. Include random data for fields like userId, email, and amount. The output should be a JSON object that can be copied and sent to the webhook URL for testing." **Example:** A developer quickly tests a new webhook integration without a real event. ## Conclusion Prompts are the secret weapon of efficient automation engineers. They force you to think through every step, anticipate failures, and document your logic in a reusable format. Whether you're using Make's visual builder, n8n's code-friendly nodes, or Zapier's quick Zaps, a well-written prompt can turn a vague idea into a production-ready workflow. Start by copying the prompts from this guide and adapting them to your specific tools and data. As you gain confidence, you'll develop your own library of prompts for common patterns—lead scoring, error handling, data parsing, and more. The time you invest in writing clear prompts will pay back tenfold in reduced debugging and faster implementation. Automation is not just about connecting apps; it's about designing reliable processes. And great prompts are the first step toward that reliability.

30 Prompts for Workflow Automation: Make, n8n, and Zapier

Comments

Recent articles

ASI Biont AI Agent Fleet Management Integration: Automate Vehicle Tracking, Maintenance, and Compliance with No-Code API

Информационная безопасность (право): как защитить коммерческую тайну и не нарушить ФЗ-149 — обучение на asibiont.com

CE Marking and EU Technical Regulations — Product Compliance: How AI Training Helps Enter the European Market

Jersey Mike’s IPO Illustrates How Bad the AI Hype Has Become

Web Security (OWASP Top 10): How to Protect Your Website from Hacking in 2026 and Earn More

How GitHub Maintains Compliance for Open Source Dependencies: A Deep Dive into Modern Dependency Governance

How to Master Multimodal AI: A Practical Guide to GPT-4V, CLIP, and Whisper with asibiont.com's Course

Wordgard: The New In-Browser Rich-Text Editor from the Creator of ProseMirror — A Hands-On Expert Review

Mastering AWS Solutions Architect Professional (SAP-C02) with AI: A Smarter Path to Cloud Mastery in 2026