AI automation with n8n

n8n workflows with an AI step: triage inbound email and forms, draft replies, enrich leads and route results to Slack or your CRM, with an optional knowledge base.

Good fit: Ops and growth teams automating repetitive work across the apps they already use, without building an app.

Infrastructure
$64.61/mo
$58.05 at half the usage · $90.87 at double

How the parts connect

  1. Host for self-hosted n8n

    Not needed

  2. Workflow engine$55.86/mo

    n8n · Pro

    ← runs (self-hosted) from Host for self-hosted n8n

  3. n8n database

    Not needed

    ← workflows, executions from Workflow engine

  4. AI model$8.75/mo

    OpenAI API · gpt-5-mini

    ← classify, extract, draft from Workflow engine

  5. Knowledge store$0

    Supabase · Free

    ← search docs from Workflow engine

  6. Embedding model$0

    OpenAI API · text-embedding-3-small

    ← embed query / documents from Workflow engine

Monthly cost

Estimate

Expected cost at the usage below, and the range from half to double that usage. Worked out from each vendor's current prices and free tiers; your real bill depends on your traffic.

PartPickExpected / moRange
Workflow enginen8n · Pro$55.86/mofixed
Host for self-hosted n8nNot needed—
n8n databaseNot needed—
AI modelOpenAI API · gpt-5-mini$8.75/mo$2.19–$35
Knowledge storeSupabase · Free$0fixed
Embedding modelOpenAI API · text-embedding-3-small$0$0–$0.01
Infrastructure total$64.61/mo$58.05–$90.87
Show the arithmetic behind every number

Workflow engine · n8n · Pro

  • executions = workflow_runs_per_month = 5,000
  • Pro list price 50 EUR/month × 1.1172 USD/EUR (ECB 2026-10-30) = $55.86

AI model · OpenAI API · gpt-5-mini

  • input_tokens = workflow_runs_per_month * ai_steps_per_run * input_tokens_per_step = 5,000 × 2 × 1,500 = 15,000,000
  • output_tokens = workflow_runs_per_month * ai_steps_per_run * output_tokens_per_step = 5,000 × 2 × 250 = 2,500,000
  • input: 15,000,000 tokens ÷ 1M × $0.25 = $3.75
  • output: 2,500,000 tokens ÷ 1M × $2.00 = $5.00

Knowledge store · Supabase · Free

  • vector_gb = knowledge_documents * knowledge_tokens_per_document / chunk_tokens * vector_kb_per_chunk / 1000000 = 200 × 1,500 / 800 × 8 / 1000000 = 0.003
  • db_gb = knowledge_documents * knowledge_tokens_per_document / chunk_tokens * vector_kb_per_chunk / 1000000 = 200 × 1,500 / 800 × 8 / 1000000 = 0.003
  • write_units = knowledge_documents * knowledge_reindex_share * knowledge_tokens_per_document / chunk_tokens * vector_kb_per_chunk = 200 × 0.1 × 1,500 / 800 × 8 = 300
  • read_units = workflow_runs_per_month * search_share * read_units_per_query = 5,000 × 0.5 × 1 = 2,500
  • Free list price $0.00/month

Embedding model · OpenAI API · text-embedding-3-small

  • input_tokens = knowledge_documents * knowledge_reindex_share * knowledge_tokens_per_document + workflow_runs_per_month * search_share * question_tokens = 200 × 0.1 × 1,500 + 5,000 × 0.5 × 50 = 155,000
  • input: 155,000 tokens ÷ 1M × $0.02 = $0.0031

The tools, and why

Workflow engine

$55.86/mo

n8n · Pro

n8n Cloud with enough executions and active workflows for a team

n8n.io ↗

Host for self-hosted n8n

Not needed

n8n Cloud hosts the workflows

n8n database

Not needed

Included in n8n Cloud

Knowledge store

$0

Supabase · Free

n8n's Supabase Vector Store node works out of the box with pgvector

supabase.com ↗

All costs are estimates. Prices are read from each vendor's pricing page and checked every 12 hours (last: Oct 30, 2026); your bill depends on your traffic.

Model quality scores: Intelligence Index by Artificial Analysis.

The usage these costs assume

"Half" and "double" in the ranges scale these numbers.

Workflow runsAbout 150 inbound emails or form submissions a day
5,000 executions/month
Active workflowsTriage, enrichment, follow-ups and a few internal reports
8 workflows
AI steps per runClassify then draft or extract
2 steps
Knowledge documentsProduct docs and saved replies
200 documents
Execution history keptEnough to debug failures; older runs are pruned
30 days
Engineering assumptions (12)
Input per AI stepAn email or form plus instructions and examples
1,500 tokens
Output per AI stepA label with fields, or a short draft
250 tokens
Tokens per knowledge documentA doc page or saved reply
1,500 tokens
Knowledge re-embedded monthlyA tenth of the docs change each month
0.1 share
Runs that search knowledgeDrafting steps search; classification doesn't
0.5 share
Tokens per searchThe query the agent embeds
50 tokens
Chunk sizen8n's default text splitter range
800 tokens
Stored size per chunk1,536-dim float vector plus text
8 KB
Read units per searchA small index
1 read units
Stored size per executionExecution data kept for debugging
50 KB
Database compute sizeThe smallest Neon compute
0.25 CU
Database active hoursn8n keeps its database busy around the clock
720 hours/month

How to build it

  1. 1

    Get n8n running

    Create an n8n Cloud workspace, or deploy the n8n template on the host with Postgres connected and a public webhook URL.

    Uses: Workflow engine, Host for self-hosted n8n, n8n database

  2. 2

    Connect your apps

    Add credentials for email, forms, Slack and the CRM; build the trigger and route steps first without AI.

    Uses: Workflow engine

  3. 3

    Add the AI step

    Insert an AI agent node that returns structured JSON (category, priority, fields) and branch on it.

    Uses: Workflow engine, AI model

  4. 4

    Give it knowledge

    Load docs and past replies into the vector store so drafted replies quote your own material.

    Uses: Knowledge store, Embedding model

  5. 5

    Review before sending

    Post AI drafts to Slack for one-click approval until accuracy is proven; log mistakes and add them as examples.

    Uses: Workflow engine, AI model

Not covered by this estimate

  • An OSI-licensed workflow engine (e.g. Activepieces, Windmill) for the open-source tier: not in the catalog yet

Get this built for you

We'll adapt this blueprint to your product, set it up on your accounts and hand it over working, with the monthly bill you saw here.

Get this built for youSee other blueprints