OpenAI + Gmail proposals, polished PDFs sent fast
You get a new lead, you promise a proposal “today,” and then the copy-paste marathon begins. Intake notes in one place, old proposal templates in another, and your inbox filling up while you wrestle formatting.
If you’re a consultant trying to respond faster, a marketing lead handling inbound requests, or an agency owner who wants AI proposal automation without duct-taping five tools together, this fixes the messy middle. You will send consistent, client-ready PDFs without rewriting the same sections again and again.
This guide shows what the workflow does, what you need, and how the moving parts fit so you can run proposals on demand.
How This Automation Works
Here’s the complete workflow you’ll be setting up:
n8n Workflow Template: OpenAI + Gmail proposals, polished PDFs sent fast
flowchart LR
subgraph sg0["0. Form Trigger (Client Data Input) Flow"]
direction LR
n0["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/form.svg' width='40' height='40' /></div><br/>0. Form Trigger (Client Data.."]
n1@{ icon: "mdi:code-braces", form: "rounded", label: "1. Prepare AI Prompt & Clien..", pos: "b", h: 48 }
n2@{ icon: "mdi:robot", form: "rounded", label: "2. Generate Proposal Content..", pos: "b", h: 48 }
n3["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>3. Generate Proposal PDF (PD.."]
n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>5. Download Generated PDF1"]
n5@{ icon: "mdi:code-braces", form: "rounded", label: "6. Prepare Email Data1", pos: "b", h: 48 }
n6@{ icon: "mdi:cog", form: "rounded", label: "4. Wait (for PDFmonkey Webho..", pos: "b", h: 48 }
n7@{ icon: "mdi:message-outline", form: "rounded", label: "7. Send Proposal Email to Cl..", pos: "b", h: 48 }
n5 --> n7
n4 --> n5
n6 --> n4
n2 --> n3
n0 --> n1
n1 --> n2
n3 --> n6
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n0 trigger
class n2 ai
class n3,n4 api
class n1,n5 code
classDef customIcon fill:none,stroke:none
class n0,n3,n4 customIcon
Why This Matters: Proposal Turnaround Gets Stuck in Draft Hell
Proposals don’t usually fail because you can’t write. They fail because the process is full of tiny interruptions: hunting down scope notes, reusing an old doc that still has the last client’s name in it, and spending too long making the PDF look “real.” Meanwhile the lead is cooling off. And if you have multiple people touching the same draft, the tone drifts, the pricing section gets inconsistent, and you’re stuck doing a final rewrite right when you should be following up.
It adds up fast. Here’s where it breaks down in the real world.
- Every proposal turns into 45 minutes of rewriting “standard” sections you already know by heart.
- Formatting and PDF exporting feels trivial until you do it five times in a week, and then it’s suddenly your whole afternoon.
- Small errors slip in (wrong company name, old scope text), which is the kind of mistake that quietly kills trust.
- Your response time becomes inconsistent, so leads get different experiences depending on how busy you are.
What You’ll Build: Intake-to-PDF Proposals That Send Themselves
This workflow starts with a simple n8n form that captures client details (name, email, company, and the inputs you want for scope). The form submission triggers a formatting step that cleans the data and turns it into a clear prompt for OpenAI. OpenAI drafts the proposal content as structured JSON (not a messy blob of text), which is important because it keeps your proposal consistent from lead to lead. That JSON is sent to PDFMonkey, where it fills your pre-built PDF template and generates a polished document. Once PDFMonkey confirms the PDF is ready, the workflow downloads it and sends a personalized Gmail email to the client with the PDF attached.
The workflow begins at client intake, then moves through AI drafting and templated PDF creation. It ends by delivering the proposal directly via Gmail, so you’re not chasing files or exporting anything manually.
What You’re Building
| What Gets Automated | What You’ll Achieve |
|---|---|
|
|
Expected Results
Say you send 5 proposals a week. Manually, assume about 45 minutes to draft and edit, plus maybe 15 minutes to format/export and attach, so roughly 1 hour each (about 5 hours weekly). With this workflow, you spend about 5 minutes submitting clean intake details (or asking the lead to do it), then wait while the PDF renders, and do a quick 5-minute sanity check before it goes out. That’s about 10 minutes of your time per proposal, not an hour.
Before You Start
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- OpenAI for drafting proposal sections from intake data
- Gmail to send the PDF proposal automatically
- PDFMonkey API key (get it from your PDFMonkey account settings)
Skill level: Intermediate. You’ll mostly connect accounts and paste keys, plus match your PDF template fields to the JSON output.
Want someone to build this for you? Talk to an automation expert (free 15-minute consultation).
Step by Step
Client intake kicks everything off. A new submission comes in through the built-in n8n form, which means you can share one link and keep inputs consistent.
The workflow prepares a clean AI request. A Function step takes the raw fields (name, email, company, needs) and formats them into a prompt that OpenAI can follow without guessing what goes where.
OpenAI drafts structured proposal content. Instead of “write me a proposal,” it generates a JSON outline with sections like problem summary, solution, scope, and milestones, so the output stays predictable.
A PDF is generated and emailed. PDFMonkey merges that JSON into your branded template, the workflow waits for the “document succeeded” callback, downloads the finished PDF, then Gmail sends it to the client as an attachment.
You can easily modify the form fields to collect different inputs (budget, timeline, package tier) based on your needs. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the Form Trigger
This workflow starts when a client submits a form and passes their details into the proposal pipeline.
- Add the Client Form Intake node as your trigger.
- Open Client Form Intake and configure the form fields you need for client details (name, email, scope, budget, etc.).
- Ensure Client Form Intake is connected to Assemble AI Prompt Data.
Step 2: Connect the Data Preparation Layer
Client inputs are formatted and structured for AI prompt readiness and downstream usage.
- Open Assemble AI Prompt Data and implement logic to format the submitted form fields into a structured prompt object.
- Confirm Assemble AI Prompt Data outputs to AI Proposal Drafting.
⚠️ Common Pitfall: If the function output is not valid JSON or missing fields, AI Proposal Drafting will generate incomplete content.
Step 3: Set Up AI Proposal Generation
The AI node generates the proposal content that will be sent to PDFMonkey.
- Open AI Proposal Drafting and configure your prompt and response format requirements.
- Credential Required: Connect your OpenAI credentials in AI Proposal Drafting.
- Ensure AI Proposal Drafting connects to Create PDF via PDFMonkey.
Step 4: Configure PDF Generation and Retrieval
This section sends AI output to PDFMonkey, waits for generation, and retrieves the finished file.
- Open Create PDF via PDFMonkey and configure the HTTP request to PDFMonkey’s document creation endpoint.
- If authentication is required by your PDFMonkey setup, add the necessary headers or auth in Create PDF via PDFMonkey.
- Ensure Create PDF via PDFMonkey connects to Await PDFMonkey Callback.
- Open Await PDFMonkey Callback and verify the webhook URL is registered inside PDFMonkey for callback completion.
- Open Retrieve PDF File and configure the HTTP request to download the generated PDF from PDFMonkey’s API.
- Confirm Await PDFMonkey Callback outputs to Retrieve PDF File.
⚠️ Common Pitfall: If the PDFMonkey callback URL is not set correctly, Await PDFMonkey Callback will never resume the workflow.
Step 5: Configure Email Assembly and Delivery
The workflow prepares the email payload and sends the finished proposal PDF to the client.
- Open Compose Email Payload and ensure it maps client details and PDF binary data into an email-ready structure.
- Verify Retrieve PDF File connects to Compose Email Payload, and Compose Email Payload connects to Dispatch Proposal Email.
- Open Dispatch Proposal Email and configure recipient, subject, and attachment fields.
- Credential Required: Connect your Gmail OAuth2 credentials in Dispatch Proposal Email.
Final Step: Test and Activate Your Workflow
Validate each stage of the workflow to ensure proposals are generated, retrieved, and emailed correctly.
- Click Execute Workflow and submit a test entry via Client Form Intake.
- Confirm that AI Proposal Drafting generates content and that Create PDF via PDFMonkey receives it.
- Verify the workflow pauses at Await PDFMonkey Callback and resumes once PDFMonkey completes generation.
- Check that Retrieve PDF File returns a binary PDF and that Dispatch Proposal Email sends it to the client.
- When successful, toggle the workflow to Active for production use.
Troubleshooting Tips
- Gmail credentials can expire or need specific permissions. If things break, check your n8n Credentials page and reconnect the Gmail OAuth credential first.
- If you’re using Wait nodes or external rendering, processing times vary. Bump up the wait duration if downstream nodes fail on empty responses.
- Default prompts in AI nodes are generic. Add your brand voice early or you’ll be editing outputs forever.
Quick Answers
About 30 minutes if your accounts and template are ready.
No. You’ll connect credentials and edit a couple of text fields. The “hard part” is just mapping your PDF template to the JSON sections you want.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenAI API usage and PDFMonkey charges for document generation.
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
Yes, and you should. The easiest place to customize is the “Assemble AI Prompt Data” Function step, because that’s where you decide what inputs matter and how the AI should structure the JSON. You can also swap Gmail for a different email provider by replacing the “Dispatch Proposal Email” step, while keeping the rest intact. Common tweaks include adding pricing tiers, inserting your guarantee/terms section, and creating different PDF templates for different services.
Usually it’s an expired or revoked OAuth connection. Reconnect your Gmail credential in n8n, then confirm the sending address matches the authenticated account. If it still fails, check Google’s security settings and make sure the n8n project is allowed to send mail.
A lot for a small team. On n8n Cloud, your limit depends on your plan’s monthly executions, while self-hosting mainly depends on your server. PDFMonkey and OpenAI can rate-limit you if you try to burst hundreds at once, so if you’re sending a big batch, stagger intake or add a short wait between runs.
Often, yes, because this flow needs a few things that get annoying in simpler tools: waiting for a webhook callback, handling binary PDF downloads, and shaping structured JSON for templating. n8n also gives you Functions for custom logic without paying for “paths” every time you add branching. If you want full control, self-hosting is a big deal since execution volume won’t surprise-bill you. Zapier or Make can still be fine if you only need a basic “form → email” automation and you never plan to customize it. If you’re on the fence, Talk to an automation expert and we’ll help you pick the simplest option that won’t break later.
Once this is live, proposals stop being a “later today” task and become something you can deliver reliably, even on busy weeks. Set it up once, then let the workflow handle the boring parts.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
