What This Guide Covers
This is a practical walkthrough for freelancers, consultants, and small service teams. By the end, you'll have a clear system to automate the repetitive parts of client onboarding using AI tools — without writing code or hiring a developer.
If you've ever copy-pasted a welcome email at midnight, chased a client for a signed contract, or manually created a project folder for every new engagement, this is for you.
The Problem with Manual Onboarding
Here's what manual onboarding actually costs you:
When I tracked my own workflow in early 2024, I spent an average of 47 minutes onboarding a single new client. That was across: sending a welcome email, creating a project folder, setting up intake form data in my CRM, scheduling the kickoff call, and sending the first status update. Every. Single. Time.
For 3–4 new clients a month, that's 2–3 hours of pure admin. Multiply by a year. Then ask whether that time could have been client work, lead generation, or sleep.
The AI-Powered Onboarding Stack I Use
Tools required:
- Tally or Typeform — client intake form
- Make (formerly Integromat) — automation backbone
- Claude or ChatGPT API — personalized email drafts and project briefs
- Notion or Google Drive — project workspace creation
- Cal.com or Calendly — kickoff call scheduling
- HubSpot (free tier) or Airtable — CRM record creation
Total cost: roughly $15–30/mo depending on Make plan and AI API usage.
Step 1: Build Your Intake Form
The intake form is the starting gun for automated onboarding. Every subsequent step depends on what the client submits here.
I use Tally (free, no per-submission fees) with these fields:
- Full name
- Company / project name
- Email address
- Brief project description (long text)
- Budget range
- Timeline / deadline
- How they found you (dropdown)
- Any specific goals or constraints (long text)
AI enhancement at this step: Tally's native Notion integration can push intake data directly to a Notion database. If you're using Make instead, Tally fires a webhook on submission.
Why this matters: You want structured data coming in, not an unformatted email. Structured data is what makes the rest of the automation reliable.
Step 2: Trigger the Automation in Make
Make (the automation platform) watches for new form submissions and kicks off the entire chain.
In my Make scenario, the webhook from Tally triggers:
- Parse the form data
- Look up whether this email already exists in HubSpot
- Create a new contact (if new) or update the existing one
- Pass project description to Claude API for brief generation
- Create a Notion project workspace from a template
- Send a welcome email (drafted by AI, reviewed by me — more on this below)
- Send the client a calendar link for the kickoff call
The entire chain runs in under 90 seconds.
Tip: Use Make's error handling modules. If one step fails (say, the Claude API times out), you want the automation to alert you and pause — not silently drop the client.
Step 3: Generate a Personalized Project Brief with AI
This is where AI adds genuine value. Instead of reading intake form responses and manually writing a project brief, I pass the raw intake data to Claude via Make's HTTP module.
The prompt I use (adapted for your context):
You are a professional project manager. Based on the following client intake:
Client name: {{name}}
Company: {{company}}
Project description: {{description}}
Goals: {{goals}}
Timeline: {{timeline}}
Write a concise project brief (200-300 words) that:
1. Summarizes the project scope
2. Identifies 2-3 key success metrics
3. Flags any potential risks or clarification needed
Tone: professional but direct. Avoid filler phrases.
The output goes into the Notion project page automatically. When I open the new client workspace, the brief is already there — I just edit, not create from scratch.
This alone saves me 15–20 minutes per client.
Step 4: Create the Project Workspace Automatically
In Notion, I have a master project template with pages for: brief, meeting notes, deliverables tracker, and a shared client portal.
Make's Notion integration can duplicate a template database entry with all the intake data pre-filled. The client name, project description, and key dates populate automatically.
For Google Drive users, Make can similarly copy a folder template and rename it with the client name — no manual folder creation.
Naming convention I use: YYYY-MM ClientName — ProjectType
Consistent naming makes search reliable and avoids the "what did I call that folder?" problem.
Step 5: Draft and Send the Welcome Email
The welcome email is where most service providers spend disproportionate time trying to sound warm, professional, and personal simultaneously.
I use AI to draft it and then review before sending — or for repeat client types, I've reviewed enough drafts to trust the automation to send directly.
The Make scenario calls Claude with the intake data and a prompt like:
Write a brief welcome email (150 words max) to a new client named {{name}} from {{company}}. They're working on: {{description}}.
The email should:
- Confirm receipt of their intake form
- Express genuine enthusiasm for their project
- Confirm I'll be in touch within 24 hours to confirm kickoff details
- Include a line to book their kickoff call: {{cal_link}}
Tone: warm, direct, no filler. Sign off as [Your Name].
The resulting email goes out via Gmail (connected to Make). I keep a log of all sent emails in Notion.
Step 6: Schedule the Kickoff Call
The intake form submission triggers a Cal.com (or Calendly) link that goes directly to a dedicated "kickoff call" event type — pre-configured with the right duration (usually 30 minutes), relevant questions pre-filled, and a confirmation reminder.
Combined with the welcome email, most clients book within 24 hours of submitting the form. No back-and-forth email scheduling.
Step 7: Human Review Checkpoint (Don't Skip This)
I keep one human review step: after the automation runs, I get a Slack message (via Make) with a summary of what was created. I spend 2–3 minutes reviewing the AI-generated brief and confirming the welcome email was sent.
This isn't about not trusting AI — it's about maintaining quality control at scale. In six months of running this, I've edited the brief about 30% of the time and rarely needed to touch the welcome email.
What This Automation Doesn't Replace
Be honest with yourself about the limits:
- AI can't replace a real conversation when the project is ambiguous
- Complex or high-stakes engagements should still start with a discovery call before the intake form
- Contract signing and payment should use a dedicated tool (HelloSign, PandaDoc) — don't automate contract generation without review
Summary: The Full Flow
- Client submits Tally intake form
- Make webhook fires
- CRM record created (HubSpot)
- AI generates project brief (Claude via API)
- Notion project workspace created from template
- Welcome email drafted by AI, sent via Gmail
- Kickoff calendar link sent to client
- You get a Slack summary — review in 2 minutes
Total time saved per new client: 35–45 minutes. Recurring, every single time.
FAQ
Q: Do I need coding skills to build this? No. Make's visual scenario builder handles all the logic without code. The HTTP module (for Claude API calls) requires copying a prompt and API key — that's the most technical step.
Q: How much does this cost per month? Approximately: Make Core plan ~$9/mo (verify), Claude API roughly $1–3/mo at low volumes, Tally free, Cal.com free. Under $15/mo total for most freelancers.
Q: Can I use ChatGPT instead of Claude? Yes. OpenAI's API works identically in Make's HTTP module. Claude tends to produce cleaner, more concise professional writing in my testing — but both work.
Q: What if the client submits incomplete intake information? Build a Make filter that checks for required fields. If fields are empty, trigger a follow-up email asking for the missing info instead of running the full chain.