n8n Webhook Workflow: Draft Product Review Replies (JSON)

Draft on-brand public product review replies with an n8n webhook + JSON

This n8n Webhook Workflow receives product reviews via a POST webhook and uses an OpenAI-compatible Chat Completions API to draft a public reply in valid JSON, including token usage and an estimated cost—with a single corrective retry if the output is invalid.

What this workflow does

• Accepts review data through a webhook request (rating, review text, product, reviewer_first_name, and an optional review_id).
• Loads configuration such as the system prompt, brand voice, model name, API base URL, temperature, and per-token pricing.
• Calls an OpenAI-compatible /chat/completions endpoint to generate a JSON object containing the drafted reply plus additional fields like tone and publish_ready.
• Validates the response by parsing the returned content as JSON and checking that required keys are present.
• Retries once (only if needed): if the model output is invalid or the API errors, the workflow sends one corrective follow-up request and validates the retried JSON.
• Responds to the webhook with the reply draft (including usage, attempt count, and estimated cost). If it still can’t produce valid JSON, it returns needs_human=true.

Use cases

• Automatically generate customer-friendly responses for product review management pipelines.
• Run a lightweight SaaS customer success process: draft replies instantly, then route only failures to a human.
• Improve consistency with a defined brand voice and system prompt across all review replies.

Technical details

• Webhook (POST) to receive review details and Respond to Webhook with the JSON result.
• HTTP Request to the OpenAI-compatible /chat/completions endpoint.
• JSON validation logic using nodes such as if and set (plus a single corrective retry flow).
• Optional sticky note for workflow documentation.

Setup note: Create an HTTP Header Auth credential with Authorization: Bearer YOUR_API_KEY, update the LLM base URL and model name to match your provider, and adjust system prompt/brand voice/temperature and token pricing.