The Smartest Email I Never Wrote: How I Automated My Contact Form (No Code Required)
- Catherine Del Vecchio Fitz
- May 15, 2025
- 7 min read
Series: How I Built My Business – Part 2
This is Post #2 in my “How I Built My Business” series, where I’m sharing the behind-the-scenes systems, tools, and strategies powering Savvyn Insights. You can read Post #1 (about building my website with ChatGPT) here.
TL;DR: I built an automated system using Zapier, GPT-4, Airtable, and Gmail that handles every inbound inquiry to Savvyn Insights—from categorizing the message to sending a tailored follow-up email, summarizing form responses, and flagging replies. It saves hours, improves quality, and creates a consistent, high-touch client experience—without burying me in inbox admin.
The Backstory
🧩 Why Automate the Contact Form?
I launched Savvyn Insights to help life sciences teams streamline messy workflows, but I knew I wanted my own workflows to feel clean, fast, and intentional.
One of the first bottlenecks? New inquiries.
Manually replying to contact form submissions was slow, repetitive, and (ironically) not very scalable. I wanted:
Every message to be acknowledged promptly
A way to qualify and categorize the inquiry
An easy way to collect more info before jumping on a call
So I built a system that does all of that—automatically.
⚙️ What I Built: Three Zaps, One Cohesive System
There are three Zapier workflows behind the scenes:
⚡ Zap 1: Contact Form → Categorize + Auto-Respond
This is the first automation that kicks in the moment someone fills out the website contact form.
It starts with a Wix form submission, which triggers a new record in Airtable, capturing key details like name, email, and the content of the message. From there, the magic happens.
GPT-4 reviews the inquiry and classifies it into one of three categories:
– Services Inquiry (they're interested in a specific offering)
– Partnership Inquiry (they’re proposing collaboration)
– Other Inquiry (general, unclear, or off-topic)
Based on this classification, a second GPT-4 call drafts a follow-up email tailored to the inquiry type. The tone is always professional and warm, but the message changes: for service inquiries, it includes a link to a more detailed google intake form; for partnership or general messages, it asks thoughtful follow-up questions to clarify intent.
That email is sent automatically via Gmail and also logged back in Airtable for future tracking. This process creates a high-quality, customized experience that requires zero manual input from me.
This zap is the cornerstone of my contact automation flow—and ensures that every inquiry gets acknowledged quickly, intelligently, and with context.
⚡ Zap 2: Google Intake Form → Summarize + Recommend Next Steps
Once someone fills out the intake form linked in the initial follow-up email, this zap handles everything that happens next—and it’s probably my favorite part of the whole system.
When a Google Form submission comes in, the zap automatically matches it to the right contact record in Airtable. Then GPT-4 reads through the form responses and generates a clean, high-level summary of what the person is asking for. Not just a regurgitation of answers—it actually distills the core need, interprets what matters, and puts it in plain English I can review quickly.
Next, GPT-4 generates suggested next steps based on what they’ve shared. For example, it might flag that the person sounds like a good fit for a Savvyn Docs sprint, or that their problem is vague and needs more scoping. That summary and recommendation are saved to Airtable.
Finally, the zap sends me a notification in Google Chat so I know I’ve got a new serious inquiry waiting—with everything already summarized for me. This means I can respond with confidence, clarity, and context—without having to dig through form answers myself.
Zap 2 turns raw interest into something actionable. It’s what moves the lead from “possible” to “let’s go.”
⚡ Zap 3: Email Reply → Summarize + Flag for Follow-Up
Not everyone uses the intake form—some people just hit reply and type back a few lines. This zap is built to handle those informal, unstructured responses so they don’t fall through the cracks.
It starts by monitoring my Gmail inbox. When someone replies to an email that came from the system, the zap checks if they already have a record in Airtable. If they do, GPT-4 jumps in to summarize what they said—especially helpful when it’s a short or vague message.
The zap then updates their record in Airtable with that summary, so I have a full picture of the conversation all in one place. It doesn’t send a new email or make any assumptions—it just makes sure I’m not losing valuable context when people choose to engage outside the structured workflow.
Zap 3 is my quiet safety net. It catches responses, organizes them, and helps me keep a clean thread—so I’m not stuck wondering, “Did I already follow up on that?”
🧠 The Prompts That Make It Work
This isn’t just a case of “if form submitted, then send email.” GPT-4 is doing real decision-making behind the scenes. What makes this system powerful is that it doesn’t just respond—it reads, classifies, and adapts.
The first prompt kicks in right after someone submits the form. GPT-4 reads their message and decides what kind of inquiry it is:
If someone’s asking for help with AI, data, or workflows, it gets tagged as a Services Inquiry.
If they’re floating a potential collaboration or joint initiative, it’s a Partnership Inquiry.
If the message is more general—like “what do you do?” or “how can I learn more?”—it’s labeled as an Other Inquiry.
That categorization step determines everything that follows. It tells GPT how to write the email, what tone to take, and whether or not to include a link to the intake form.
Once the inquiry is categorized, a second GPT prompt writes the actual follow-up. These responses are short, professional, and to the point. For service inquiries, the email includes a link to a simple form to collect more detail. For partnership or general inquiries, the email includes a few quick questions to help clarify what they’re looking for. In all cases, the goal is the same: make it easy for people to move forward without creating more back-and-forth for me.
📌 One funny (and slightly annoying) quirk: GPT really, really wants to say “Rest assured.”
Every. Single. Time.
I now include the line “Do not say ‘Rest assured’” in every follow-up prompt. Because this is Savvyn Insights—not Delta Airlines circa 2002 😁.
✉️ Sample Emails
Here are two actual examples that came straight from this workflow—written by GPT, personalized by type, and sent automatically:
📨 Example 1: Other Inquiry
(No form link—just a few thoughtful questions)
📨 Example 2: Services Inquiry
(Includes direct link to the intake form)
📨 What Happens After the Form
If someone completes the intake form, Zap 2 kicks in to summarize their responses and generate a recommended next step. That includes a draft follow-up email—like this one:
🧾 Sample Follow-Up: Post-Form Summary + Next Steps Email
Subject: Next Steps for Your Process Automation Project
Hi Catherine,
Thank you for your prompt response and for expressing your interest in moving forward with process automation. We're excited to assist you in kicking off this project within the next month!
Next Steps:
Schedule a Discovery Call – To better understand your specific needs and goals for the automation project, I recommend we schedule a discovery call. This will allow us to discuss your objectives, current processes, and any challenges you may be facing. Please let me know your availability or feel free to book a time directly using this link: Schedule a Call.
Identify Key Processes – Before our call, it would be helpful if you could identify the specific processes you are looking to automate. This will help us tailor our discussion and provide you with the most relevant solutions.
Explore Our Resources – I can share case studies or resources that demonstrate how we have successfully implemented process automation for other organizations. This may give you additional insights into what to expect and how we can support you.
I look forward to your response and the opportunity to collaborate on this exciting project!
Best Regards...
Right now, this email isn’t sent automatically—and that’s intentional. I’m not quite ready to hand off the conversation completely at this stage. I still want my eyes on the responses and the tone before moving forward. This draft gets generated and added to Airtable, giving me a head start. If it looks good, I can copy, tweak, and send it manually. Eventually, I might automate it—but for now, I want to keep a hand on the wheel for this part of the process.
It’s a great example of how automation doesn’t have to be all or nothing. Even partial automation can give you speed, consistency, and structure—while keeping the human touch where it matters.
🚀 Why It Matters
This system isn’t just about saving time—it’s about creating a better experience for everyone involved. Every inquiry gets a thoughtful response in minutes, not days. I don’t have to dig through emails or retype the same follow-up over and over. And everything stays clean, searchable, and actionable in Airtable.
It reflects exactly what I offer my clients: clear, efficient, AI-powered systems that take the chaos out of everyday workflows.
And look—none of this is rocket science. I’m not claiming it is. These are basic tools (Zapier, GPT, Airtable, Gmail), strung together with a little intention. But that’s the point: you don’t need custom software or an engineering team to automate real work. Sometimes it just takes a form, a prompt, and a well-placed Zap.
Hopefully this gives you a window into how simple, repeatable automations can free up your time, reduce friction, and let you focus on the work that actually matters.
Thanks for reading,
—Savvyn (your partner in ruthless efficiency)
🌀 ➝ 📊 ➝ 💡 ➝ 🚀
From Chaos to Clarity. Amplify Your Impact.
P.S. I built this entire blog—from idea to publish—in under an hour. Almost half that time was spent futzing with formatting (which I will automate next). I didn’t spend longer because that’s all the time I had today—sick kid home from school. Done is better than perfect, and sometimes “good enough” is exactly what you need to keep moving.
Comments