Buyer guide · Updated 2026-11-07

Cheap VPS for an n8n Lead-Triage Assistant

See how a small team can use a self-hosted n8n workflow to triage inquiries with AI and human review, with current VPS pricing and purchase terms.

✓

Quick answer
A cheap VPS can keep an internal n8n lead-triage assistant available to your team: a signed-in staff member submits a short form, an AI Agent labels the request and prepares a reply, and Gmail pauses for a person to approve the message. One current plan to evaluate is DigitalOcean Basic with 1 vCPU, 2 GiB RAM, 50 GiB SSD, and 2,000 GiB transfer, listed at $0.01786/hour with a $12/month cap (USD). n8n publishes no universal hardware minimum, so treat this as a pilot candidate to monitor, not a fit guarantee. Source check: November 7, 2026.

Original workflow illustration. n8n documents each of these building blocks, including a human approval step for AI tool calls.

A useful small-team job: triage a new inquiry before replying

Imagine a two-person consulting shop receiving inquiries that range from “What does this service cost?” to “Can you handle this project next month?” A private intake workflow can ask a teammate to enter a short summary, ask an AI Agent to classify urgency and draft a response, then put the draft in front of a human before anything is sent.

This workflow is intended for internal use by a small team. Keep the form behind n8n user authentication and collect only the minimum information needed for triage. Do not submit payment details, health information, passwords, or other sensitive data to an AI model. If an inquiry contains sensitive data or creates a legal or contractual commitment, route it to a person instead of relying on an AI classification.

How the workflow can be arranged

Form Trigger: collect a lead name, contact address, short request, and a few predefined routing choices. Choose n8n User Auth so only signed-in users on the instance can view or submit the form. n8n documents this access restriction; it is appropriate for a team intake tool, not a public customer form for people without n8n accounts.
AI Agent: send only the minimal form fields to the model. Ask for a small, structured result such as category, urgency, missing information, and a suggested reply. Connect a chat model and at least one tool, as required by n8n's AI Agent node.
Human-reviewed email action: connect a Gmail action as an AI tool and configure n8n's documented human-in-the-loop review for actions that need oversight. A reviewer should verify the external recipient and proposed text before approving any send. Start by routing test output to an internal reviewer; if approval is unavailable or the model's output is incomplete, stop and hand off to a person.
Test with production-like examples: use fictional or redacted inquiries first. Check misclassification, blank fields, prompt injection in the submitted text, and the approval recipient before letting the team use real leads.

This is an automation pattern, not a claim that the model will classify leads accurately. Keep business decisions, pricing commitments, and external replies under human review until you have evidence that the workflow handles your specific cases safely.

One current VPS plan to evaluate

Purchase field	DigitalOcean Basic Droplet: 1 vCPU / 2 GiB
Price and resources	$0.01786/hour, capped at $12.00/month (USD); 1 vCPU, 2 GiB RAM, 50 GiB SSD, and 2,000 GiB transfer. Official price page checked November 7, 2026.
Intro and renewal	No introductory term or discount is shown. DigitalOcean lists this as its ongoing usage-based Basic rate and monthly cap; no separate renewal price is published. Recheck the current rate and chosen configuration at checkout.
Billing	Droplets are billed per second with a minimum charge of 60 seconds; bundled Basic plans have a monthly cap. The $12 cap is compute only and does not include optional backups or other services.
Data-center locations	DigitalOcean currently lists Basic Droplets in NYC1, NYC2, NYC3, AMS3, SFO2, SFO3, SGP1, LON1, FRA1, TOR1, BLR1, and SYD1. Some other regions limit CPU variants. Confirm the exact plan and CPU option in the creation screen.
Backups	Optional system backups cost 20% of the Droplet's monthly cost for weekly backups or 30% for daily backups. At a full $12 monthly cap, that calculates to $2.40 or $3.60 extra; charges can be lower for partial-month use. Weekly backups are retained for four weeks; daily backups for seven days.
Refunds	DigitalOcean says it does not offer refunds; it directs customers with extenuating circumstances to support.

At the full monthly cap, the published compute price plus weekly backup works out to $12 + ($12 × 20%) = $14.40 before tax or other services. With daily backup the corresponding arithmetic is $12 + ($12 × 30%) = $15.60. These figures apply the published percentages to a full $12 month; they are not checkout quotes.

The price page does not establish whether this hardware will handle your AI workflow. Model inference runs through a separate provider connection in this example, so model/API charges and rate limits are separate and must be checked with that provider. Running a large language model locally is outside this plan recommendation.

Who should consider this setup?

Potential fit: a small team that already uses n8n or can administer Docker, wants an internal form and a draft/review workflow, can keep the AI credential secure, and is willing to monitor the pilot's memory and execution behavior. The 1-vCPU/2-GiB plan gives a concrete, currently listed purchase candidate; n8n publishes no universal minimum, so start with non-sensitive sample data and measure your own workload.

Not a good fit: a buyer who expects a managed service to secure and update the server, a public lead form that needs customer accounts or protections beyond the form node's documented options, an organization that cannot send the selected fields to an external model provider, or a team that wants the AI to send unreviewed commitments. If you require guaranteed capacity, a refund, or a disclosed compliance posture, confirm those conditions with the provider before purchase.

Purchase and rollout checks

Confirm your chosen data center and current plan rate before you create the Droplet.
Decide who patches Ubuntu, Docker, n8n, and the reverse proxy; n8n explicitly lists server/container administration, resource management, security, and configuration as self-hosting prerequisites.
Keep workflows, credentials, and the n8n data volume backed up. A provider backup is an optional system image with a published retention period; it is not a substitute for a tested application-level recovery plan.
Document which form fields leave the server for the AI model, how long that provider retains them, and who can approve a reply.
Set a pilot review date. Keep the VPS only if the workflow solves enough repeated work to justify the compute, backup, model, and maintenance costs.

Official sources checked November 7, 2026

n8n Form Trigger — team form creation, authentication choices, and production workflow publishing.
n8n AI Agent — requires a connected chat model and at least one tool.
n8n Gmail Message Operations — message sending and “Send and wait for approval,” including human review for AI tool calls.
n8n Docker Compose guide — self-hosting skills, server requirements, and supported deployment pattern.
DigitalOcean Droplet Pricing — Basic 2-GiB plan resources and USD hourly/monthly rates.
DigitalOcean Regional Availability — current Basic Droplet data-center locations and CPU-option limits.
DigitalOcean Backup Pricing and Backup Quickstart — add-on rates and retention periods.
DigitalOcean refund policy — current refund statement.

Source check: November 7, 2026. The workflow example is original and based on documented n8n node behavior. No classification accuracy, resource sufficiency, or production performance has been tested or guaranteed. Verify provider terms and model-data handling before using real customer information.