Most tutorials on this topic are written for sales teams trying to scrape LinkedIn and enrich cold contact lists. That is a real use case, but it is not the one that actually costs small businesses money every day. The more common, more urgent problem is simpler: someone fills out your contact form right now, and nobody replies for six hours. This tutorial builds the opposite of a cold outreach machine. It builds a system that answers a real inquiry the moment it arrives.
Quick answer: You can build a basic automated lead reply in n8n using four connected pieces: a webhook trigger that catches your form submission, a node that drafts a reply using your own information, an email node that sends it immediately, and a logging step that records the lead. This tutorial walks through each one. Expect to spend thirty to sixty minutes building it the first time, and know upfront what it will not do, handle a complex question, negotiate, or replace judgment on anything outside a simple acknowledgment and next step.
Would rather have this built, tested, and connected to your real CRM than build it yourself? We do this exact kind of build for a fixed price, starting with a free audit.
Book a Free Automation AuditKey takeaways
- Four nodes are enough for a working first version, a trigger, a drafting step, a send step, and a log step, you do not need a complex setup to get real value.
- This workflow handles acknowledgment and basic information, not negotiation, complex questions, or anything requiring judgment, and building it to try anyway usually backfires.
- n8n gives you a test webhook URL for building safely and a separate production URL once you are ready to go live, and mixing the two up is the most common early mistake.
- Expect to spend real time on testing edge cases, a blank field, a duplicate submission, a strange email format, since this is where most first builds actually break.
- If this workflow works well for you, the natural next step is usually connecting it to your CRM and adding a follow up sequence, which is a larger build most people eventually bring in help for.
What you need before you start
You will need an n8n account, either self hosted or on n8n's cloud plan, a way to trigger the workflow, most commonly your website's contact form pointed at a webhook URL, and an email account you can connect, Gmail or another provider n8n supports. You do not need to know how to code. Every step below uses n8n's visual node editor.
Step 1: Create the webhook trigger
Open a new workflow in n8n and add a Webhook node as your first step. This node listens for incoming data and starts your workflow the moment it arrives. n8n generates two separate URLs for every webhook, a test URL and a production URL. According to n8n's own documentation, you should build and debug using the test URL, since it lets you see the incoming data directly in the editor, and only switch to the production URL once your workflow is fully built and published.
Set the HTTP method to POST, since that is what most contact forms use to submit data. Point your website's form, or a tool like Typeform, at the test webhook URL for now. Submit a test entry and confirm you can see the name, email, and message fields arriving correctly in the node's output panel before moving on.
Step 2: Structure the incoming data
Add a Set node right after the webhook. Incoming form data often arrives with inconsistent field names or extra formatting, and this step cleans it up into a predictable structure, name, email, message, and source, so every node after this one can rely on the same field names no matter how the original form was built. This small step saves real time later, since a broken field name is one of the most common reasons a workflow fails silently.
Step 3: Draft a reply using your own information
This is the step that actually answers the inquiry. Add an AI node connected to a language model, and give it clear, specific instructions, not just answer the customer, but answer using only the information you provide it, your actual services, pricing ranges, and next steps, and to say plainly that a team member will follow up if the question falls outside that information.
Being specific here matters more than almost anything else in this tutorial. A vague prompt produces a vague, generic reply that sounds like it came from nowhere. A prompt that includes your actual service list, your actual response time promise, and clear instructions to stay within that information produces something that actually sounds like your business.
Step 4: Send the reply automatically
Add an email sending node, connected to the inbox you want replies to come from, and configure it to send the drafted message to the lead's email address captured back in step two. Test this by submitting another form entry and confirming the reply lands in an actual inbox, checking spam folders too, since a new sending setup sometimes lands there during the first few tests.
Step 5: Log the lead somewhere your team can see it
Add a final node that writes the lead's details into a spreadsheet, a CRM, or wherever your team currently tracks new inquiries. This step matters more than it seems. Without it, an automated reply that goes out with no record anywhere creates a different problem, a lead your team does not know exists until they happen to check the inbox that sent it.
Step 6: Switch to the production webhook and test again
Once every step works reliably against your test URL, switch the Webhook node's setting to Production and update your website's form to point at the new production URL. Submit one final real test end to end. This is also the point where it is worth deliberately testing a few edge cases: a submission with a blank message field, a name with an unusual character, and two submissions sent close together, since these are exactly the situations that expose a gap in an otherwise working workflow.
Where this simple version stops being enough
This four step workflow genuinely works for its actual job, acknowledging a lead instantly and capturing it reliably. It is not built to negotiate, answer a complex or unusual question accurately, follow up automatically if the lead goes quiet, or update a real CRM with proper lead scoring and routing. Trying to stretch a simple version like this into handling all of that usually produces a system that occasionally gives a confident, wrong answer, which is a worse outcome than the slow reply you started with.
If you have built this and it works well, the honest next question is whether the manual patching required to extend it further is actually worth your time compared to having it built properly the first time, with real error handling, a CRM connection, and a follow up sequence included. That is exactly the kind of project we scope for free before anything gets built.
Want the version of this that includes CRM integration, error handling, and follow up sequencing, built and tested for you? A free audit tells you honestly what that would take.
Get My Free Build ScopeWhat a fully built version typically includes
Our own email automation and lead automation builds start from the same basic shape as this tutorial, then add the pieces most self built versions skip: proper error handling when a node fails, a real CRM connection instead of a spreadsheet, multi step follow up sequences for leads that go quiet, and a human handoff for anything outside routine questions. Pricing for a complete build like that typically runs between one thousand and five thousand dollars, in the same range as our other full systems, detailed further in our guide on what AI automation actually costs. For a wider comparison of the platforms this kind of workflow can be built on, our guide on automation tools for service businesses covers n8n alongside Make and Zapier in more depth.
Built this yourself and want a second pair of eyes on it, or ready to have the fuller version built? A free audit gives you a straight answer either way.
Book Your Free Automation AuditFrequently asked questions
Do I need to know how to code to build this?
No. Every step in this tutorial uses n8n's visual node editor. Some steps, like writing clear instructions for the AI node, benefit from being specific and detailed, but none of it requires writing code.
How long does this actually take to build the first time?
Expect thirty to sixty minutes for the basic four step version, plus additional time testing edge cases properly, which is where most of the real learning happens.
Can this handle a complex or unusual question from a lead?
No, and it should not attempt to. This version is built for acknowledgment and basic information only. Anything more complex should be flagged for a person to review, not answered automatically by a simple workflow like this one.
What is the difference between the test and production webhook URL?
The test URL lets you see incoming data directly in the n8n editor while you are building, which is essential for debugging. The production URL is what you switch to once the workflow is fully built and ready to run automatically without you watching it.
Why would I pay someone to build this if I can follow this tutorial myself?
You genuinely might not need to. This tutorial gets you a real, working basic version. Where people usually bring in help is once they want proper error handling, a real CRM connection, multi step follow up, and the testing needed to trust the system running unattended.
Does this work with any website contact form?
Most contact form tools, including Typeform, Webflow forms, and custom HTML forms, can be pointed at a webhook URL. The exact setup depends on your specific form tool, and some page builders require a small configuration change to send data to an external webhook.