# PandaDoc AI automation and integrations > PandaDoc has a REST API authenticated with an API key or OAuth 2.0, webhooks for document status changes, and built-in apps in Zapier and Make; n8n connects through its HTTP Request node. The highest-value AI automations draft proposals and quotes from CRM data, fill pricing tables, and follow up on documents that were viewed but not signed. API access depends on your PandaDoc plan. - Category: Docs & Productivity - Vendor: https://www.pandadoc.com/ - How to connect: PandaDoc REST API; Webhooks; Zapier and Make apps; n8n HTTP Request node; Native CRM integrations - Authentication: API key for internal integrations, or OAuth 2.0 for apps that act on behalf of PandaDoc users - Notes: Sandbox keys are free but limited to 10 requests per minute. Production API access depends on your plan, so confirm it with PandaDoc before designing around the API. - Last updated: 2026-10-14 ## What you can automate with PandaDoc - **Proposal drafting from the CRM.** When a deal reaches the proposal stage, draft scope and approach sections with an LLM and create a PandaDoc document from your template. Guide: https://www.lumeor.studio/guides/automate-proposal-generation/ - **Quote generation.** Build a pricing table from CRM line items and a product catalog, apply discount rules in code and send the quote for approval. Guide: https://www.lumeor.studio/guides/automate-quote-generation/ - **NDA and simple agreements.** Generate NDAs from a request form with PandaDoc variables and send them for signature without manual setup. Guide: https://www.lumeor.studio/guides/automate-nda-generation/ - **Viewed-not-signed follow-up.** Use document status webhooks to detect stalled proposals and draft a personalized follow-up for the account owner. Guide: https://www.lumeor.studio/guides/automate-sales-follow-up-sequences/ - **Signed deal to accounting.** When a document completes, create the customer and first invoice in your accounting system from the pricing table. Guide: https://www.lumeor.studio/guides/automate-quickbooks-data-entry/ - **RFP answer drafting.** Draft responses to RFP questions from approved content library answers and past proposals, then assemble them into a PandaDoc document for review. Guide: https://www.lumeor.studio/guides/ai-rfp-response-automation/ ## What makes PandaDoc easy to automate PandaDoc is designed for documents built from templates: variables (tokens) fill names and fields, pricing tables carry line items, and content library items hold reusable sections. The API mirrors that model. You create a document from a template, pass recipients, token values and pricing rows, then send it. Webhooks report status changes such as viewed, completed and declined. Zapier and Make both have PandaDoc apps. n8n has no dedicated node, but the API is small enough that HTTP Request nodes are easy to maintain. The main surprise for new integrations is asynchronous creation. A document created from a template is not immediately sendable. Workflows must wait for it to reach draft status. The second surprise is plan gating: check that your plan includes production API access. ## The AI layer PandaDoc already handles assembly and signature. AI adds value in the parts salespeople write by hand: - **Narrative sections** such as the problem statement, scope and approach, drafted from call notes. Our [proposal generation guide](/guides/automate-proposal-generation/) covers the review step. - **Line items** suggested from discovery notes, with prices and discounts always applied by code from your catalog. - **Follow-up** on stalled documents, using view events as the signal, as described in [sales follow-up sequences](/guides/automate-sales-follow-up-sequences/). ## Limits to plan around - **Per-endpoint rate limits.** Production limits vary by endpoint, for example 500 requests per minute for creating from a template and 100 per minute for downloads. Sandbox keys allow 10 per minute on everything, which makes load testing in sandbox misleading. - **Async processing.** Build a wait-for-draft step with a timeout and an alert. - **Template drift.** Renamed tokens or pricing table names break integrations silently. Keep a template change log. - **Plan gating.** API access and some workflow features depend on plan. Confirm current terms. ## How Lumeor typically connects it We trigger from the CRM deal stage in n8n, assemble AI-drafted sections and validated pricing rows, create the document from a locked template, wait for draft status and route it to the rep for review. A webhook listener updates the CRM on view and completion and, on signature, hands off to accounting. When the source of pricing is complex, the same flow follows our [quote generation](/guides/automate-quote-generation/) pattern. ## Frequently asked questions ### Why does PandaDoc say my document is not ready to send? Documents created through the API are processed asynchronously. They start in an uploaded state and move to draft once processing finishes. Wait for the draft status, by checking status or listening for the webhook, before calling send. ### Does n8n have a PandaDoc node? Not a built-in one. n8n uses the HTTP Request node with an API key or OAuth credential, and a Webhook node to receive status events. Zapier and Make both have built-in PandaDoc apps for teams that want a packaged connector. ### Should I use PandaDoc's HubSpot or Salesforce integration instead of building one? For creating documents from a deal and syncing status back, the native integration is usually enough and cheaper to own. Build a custom workflow when you need AI-drafted sections, pricing logic the native integration cannot express, or actions in other systems after signing. ### Can AI write the whole proposal? It can write a strong first draft of narrative sections from discovery notes and past wins. Pricing, legal terms and commitments should come from templates and validated data, and a salesperson should read every proposal before it goes out. --- Source: https://www.lumeor.studio/tools/pandadoc/ Published by Lumeor Studio, Newtown, PA. hello@lumeor.studio. Get a written scope: https://www.lumeor.studio/start/