B2B lead automation with n8n is not a consultant’s promise: it is a system that today can replace between 3 and 5 hours of manual work per week in any sales team. Capture the lead from a form, enrich it with company data, score it according to BANT criteria and assign it to the right salesperson, all in less than 90 seconds and without anyone touching anything. That is what the flows you will see in this post do.
Not theory about what n8n is, but the exact workflows, with the real tools, that B2B teams are using right now to multiply their volume of qualified leads.
The problem no B2B CMO wants to admit to
Your marketing and sales team is copying data. From Typeform to HubSpot. From HubSpot to Excel. From Excel to an email to the salesperson. No one says it at the results meeting, but according to McKinsey, 72% of repetitive marketing tasks are automatable with the tools available today. Not in the future. Today.
The real cost is not the hours lost: it is the speed of response. A B2B lead contacted in the first 5 minutes is 9 times more likely to qualify than one contacted an hour later. Every time a salesperson enters the CRM on Monday morning to review what came in on Friday, you are paying that cost.
The solution is not to hire more people. It is to set up a machine that does not sleep.
Why n8n and not Make or Zapier for B2B lead generation?
This is the question every operations manager asks before approving the project. The answer has three concrete legs:
Self-hosted means that your data does not leave your infrastructure. For B2B companies with sensitive customer data, this greatly simplifies GDPR compliance. You are not sending lead information to third-party servers in the US; the flow runs on your own European server. n8n offers DPA with Standard Contractual Clauses and its security practices are aligned with SOC 2.
2. No limit to the number of self-hosted executions. When you have 1,000-5,000 leads a month, Zapier’s per-task billing model starts to hurt. A 5-step workflow executed 3,000 times is 15,000 tasks in Zapier.
3. Native AI Agents since version 2.0. n8n includes AI agent nodes ready to use with OpenAI, Anthropic or local models via Ollama. You don’t need middleware or additional APIs to put intelligence into your qualification flow.
| Tool | Pricing model | Estimated cost (3,000 leads/month, 5-step flow) | Self-hosted | Native AI Agents |
|---|---|---|---|---|
| Zapier | By task | ~€150-€250/month | No | Yes (limited) |
| Make | Per operation | ~€50-€100/month | No | Yes |
| n8n Cloud | By execution | ~€60/month (Pro plan, 10,000 executions) | Optional | Yes (native) |
| n8n Self-hosted | Infrastructure only | ~€20-€40/month (VPS server) | Yes | Yes (native) |
Flow 1: automatic lead capture and enrichment
This is the base flow. The one that eliminates 80% of the manual work in the first week of implementation. The sequence is as follows:
- Trigger – New submission in Typeform or HubSpot Forms. The workflow is triggered as soon as someone fills out the contact or content download form.
- CRM Verification Node. Before creating anything, n8n checks if the email already exists in HubSpot or Pipedrive. If it does, update the record instead of duplicating it. This step alone avoids hundreds of duplicate records per month.
- Node Enrichment via Hunter.io or Clearbit. With the email domain, the flow calls the Hunter.io API to verify if the email is valid and get the company name. Clearbit adds staff size, industry, estimated turnover and technologies used by the company.
- ConditionalScoring Node. An IF node evaluates the data obtained: company with more than 50 employees + target sector = hot lead (score 80-100). Small company or non-priority sector = cold lead (score < 40). This logic is defined by you; the flow executes it without human intervention.
- Assignment and notification node. The hot lead goes to the senior sales person with a notification in Slack that includes name, company, title, phone and score. The cold lead goes into a nurturing sequence in Brevo or Lemlist.
Result: from form sent to commercial notified in less than 2 minutes. Without anyone touching anything.
Flow 2: intelligent nurturing that adapts to behavior
Nurturing that doesn’t convert is nurturing that sends the same email to everyone on the same day. This flow is solved by using the actual behavior of the lead as an activation signal.
- Trigger – Lead opens the email but does not click. Brevo or Lemlist activate the webhook in n8n when opening without conversion is detected.
- IA Node – Personalized email generation. An OpenAI node analyzes the lead’s profile (sector, position, company, previous emails) and generates a new adapted message. The prompt inside the node has this structure:
“You are a senior SDR specializing in [sector]. Lead [name] works as a [position] in [company, size]. He opened our email about [topic] but did not respond. Write a follow-up email of 120 words max that connects to their main pain point in [sector] and proposes a 20-minute call. Tone: direct, no pressure, no generic opening sentences.”
- Node Waiting. The flow waits 3 working days before shipping.
- Bifurcation Node. If there was a response: the lead is upgraded in the CRM and the salesperson is notified. If there was no response after 2 cycles: the lead goes to a low impact monthly cadence or is marked as passive nurturing.
Companies with automated nurturing and AI generate 451% more qualified leads than those using manual processes or static sequences, according to HubSpot data. The difference is not in sending more emails, it’s in sending the right email at the right time.
Flow 3: automatic lead scoring with AI and BANT criteria
The two previous flows use conditional logic to score. This flow goes one step further: it uses an LLM to read the free text that the lead typed in the form and assign an automatic BANT (Budget, Authority, Need, Timeline) score.
The AI node receives the form message and returns a structured JSON:
{
"budget": "alto",
"authority": "decisor",
"need": "urgente",
"timeline": "1-3 meses",
"score_bant": 87,
"razon": "Menciona presupuesto aprobado, cargo de Director Comercial, problema activo con el CRM actual y proyecto para el Q2"
}This score is written directly into the custom field in the CRM. The salesperson arrives at HubSpot or Pipedrive and already has the qualification done. In documented implementations in the official n8n community – which has more than 627 published lead generation workflows – this approach reduces manual qualification time by 55% to 65%.
You can use OpenAI GPT-4o for maximum accuracy or, if the data is especially sensitive, a local model via Ollama that runs on your own server without sending anything to the outside.
Mistakes 80% of people make when setting this up for the first time
Implementing n8n flows without prior experience has three recurring pitfalls that turn a promising project into a problem:
No lead deduplication. The most common mistake. If you don’t check before creating, each duplicate form generates a new contact in the CRM. In six months you have thousands of corrupted records. The solution is in the verification node of Flow 1: always check before creating.
No error handling. If the Hunter.io API fails or Clearbit returns an error, the workflow stops silently. The lead is lost and no one knows it. n8n allows you to configure error nodes that send an alert to Slack when something fails, so the flow always has a plan B.
No traceability logs. When did this lead arrive? What data did it have when it came in? Why was it assigned to this salesperson? Without logs, answering these questions is impossible. The simple solution is to write each execution in a Google Sheets sheet that acts as an audit trail. Add a Google Sheets node at the end of each flow with the key data and date of execution.
How to measure if your automation is working?
A flow that is not measured does not improve. These are the three metrics that determine whether your lead machine is working or just seems to be working:
Time to First Touch (Time to First Touch). The target in B2B is under 5 minutes for hot leads. If your flow works well, this number drops drastically from day one.
Lead → SQL (Sales Qualified Lead) conversion rate. If the scoring is good, the percentage of leads that the salesperson accepts as qualified should go up. A reasonable benchmark for B2B is between 20% and 35%.
Weekly volume of qualified leads. The absolute number of leads that pass the automatic filter each week. This number tells you if the funnel mouth is well calibrated.
To monitor these three metrics without additional BI tools, add a secondary flow in n8n that consolidates data from HubSpot or Pipedrive into Google Sheets every Monday at 8:00 and generates an automatic summary that arrives in the team’s email. The same n8n that captures leads gives you the dashboard.
The next level: AI agents acting on their own
The previous flows react to events: a submitted form, an open email, a signal from the CRM. The native n8n 2.0 AI Agents do something different: they act autonomously with a goal in mind.
A lead research agent configured in n8n can receive a lead’s name and company, search for updated information on LinkedIn, analyze recent news about the company, evaluate whether it fits the ideal customer profile and make the decision to pass the lead to sales or discard it, all without human intervention and with traceable reasoning that is recorded in the CRM.
This is what differentiates tactical automation – doing the same old thing faster – from strategic automation: delegating decision making to a system that doesn’t get tired, doesn’t forget criteria and scales without hiring.
At Inprofit we design and implement these systems for B2B commercial teams: from the diagnosis of the current process to the deployment and training of the team. If you want to see how it would work in your company, we talk without obligation.
Frequently asked questions about n8n and B2B lead automation
n8n is an open source workflow automation platform that connects applications, APIs and AI models into visual flows. In B2B lead generation it is used to capture, enrich, score and distribute leads automatically between forms, CRMs and email marketing tools, eliminating manual work between systems.
Not for base flows. n8n has a visual drag-and-drop interface that allows you to assemble post workflows without writing code. For advanced features like the BANT scoring node with AI, it is useful to know how to structure prompts and understand the basic JSON format, but you don’t need to be a developer.
In self-hosted mode, the cost is that of the server (between €20 and €40/month for a standard VPS) plus the APIs you use (Hunter.io from €49/month, OpenAI per use). In total, a complete lead generation system with n8n can operate for less than €150/month for volumes up to 5,000 leads. In the n8n Cloud, the Pro plan costs €60/month with 10,000 executions included.
Yes. n8n has native integrations with HubSpot, Pipedrive, Salesforce and most CRMs on the market. Nodes allow you to create, update, search and delete records directly from the workflow, without the need for Zapier or additional middleware.
Flow 1 (capture and enrichment) can be operational in one day if you already have n8n installed and the API accounts active. Flow 2 (nurturing with AI) requires 2 to 3 days including testing. Flow 3 (BANT scoring) depends on the complexity of your qualification process, but a working version can be delivered in less than a week.
Marketing tecnológico en vena. Fanático de las tecnologías Martech que rompen moldes: IA generativa, blockchain, no-code, metaverso, automatización extrema… Convencido de que el futuro no se espera, se construye (y se vende muy bien).
Responsable del marketing más disruptivo y tecnológico.
