# End-to-End Process Automation Blueprint

Design a complete automation blueprint for a manual business process, including trigger logic, tool selection, exception handling, and rollback plans.

## Prompt

Act as an automation consultant designing a production-grade workflow. Here is the manual process I want to automate:
{{manual_process_description}}

Build a blueprint covering:

1. Trigger definition — specify exactly what event should start the automation (e.g. new form submission, scheduled time, webhook), and any conditions that should prevent it from triggering unnecessarily.
2. Step-by-step flow — break the process into discrete automated steps, noting which tool or system performs each (e.g. Zapier, Make, a script, an API call).
3. Data mapping — identify what data moves between each step and any transformation needed (e.g. reformatting a date, matching a record by ID).
4. Exception handling — for each step, define what should happen if it fails (retry, alert a human, log and skip, halt the entire workflow).
5. Rollback plan — describe how to safely undo or correct the automation's actions if an error is discovered after the fact.
6. Monitoring — recommend 2-3 specific things to log or alert on to catch problems early.

Present this as a structured technical blueprint.

## Best for

Operations leads, RevOps/marketing ops professionals, and technical founders designing a reliable automation rather than a fragile one-off script.

## Compatible tools

- Claude
- ChatGPT

## How to use

- Describe your actual manual process in {{manual_process_description}}, including every current manual step and the tools involved.
- Use the trigger definition section to configure the actual automation tool's trigger settings precisely, avoiding overly broad triggers.
- Implement the exception handling logic before going live — this is the most commonly skipped step in real automations and the most common source of production issues.
- Set up the recommended monitoring/alerts on day one, not after the first failure.

## Customization tips

- Specify your actual tools (e.g. 'we use Make, HubSpot, and Slack') so the blueprint recommends realistic integration steps rather than generic placeholders.
- If the process handles sensitive data (financial, health, PII), mention this so the exception handling and rollback sections account for compliance requirements.
- Ask for a version of this blueprint formatted as a checklist for your engineering or ops team to implement directly.

## Example input

Act as an automation consultant... Manual process: When a new customer signs a contract in our e-signature tool, someone manually creates a record in our CRM, sends a welcome email, creates a folder in our shared drive, and notifies the account manager on Slack.

## Example output

1. Trigger definition: trigger on the 'document completed' webhook event from the e-signature tool, filtered to only contracts tagged as 'New Customer Agreement' to avoid firing on unrelated document types.
2. Step-by-step flow: (a) Webhook receives signed document event, (b) automation tool (e.g. Make) extracts customer name and email from the document metadata, (c) creates a new CRM record via API, (d) triggers a templated welcome email via email tool, (e) creates a folder in the shared drive using a naming convention script, (f) posts a Slack message to the account manager's channel.
3. Data mapping: customer name and email must be extracted from the e-signature tool's metadata fields and mapped to the CRM's 'Contact Name' and 'Email' fields; date signed should be reformatted from the e-signature tool's timestamp format to the CRM's expected date format.
4. Exception handling: if the CRM record creation fails (e.g. duplicate entry), log the error and alert the ops team via Slack rather than silently retrying, since duplicate customer records are a common source of downstream errors; if the welcome email fails to send, retry once after 5 minutes, then alert a human if it fails again.
5. Rollback plan: if a CRM record was created in error, provide a clear reversal step (delete or archive record) and ensure the welcome email and Slack notification are not automatically re-sent after a rollback, to avoid confusing the customer or account manager.
6. Monitoring: log every trigger event with timestamp and customer name, alert if more than 2 failures occur within a day, and track the end-to-end completion time to catch slowdowns.