The Short Answer
Expect to spend $50 to $500 per month for a small business starting out.
Expect $500 to $3,000 per month for a scale-up with 5 to 20 active workflows.
Expect $3,000 to $15,000+ per month for an enterprise deployment.
Token costs from LLM providers (OpenAI, Anthropic, Google) are usually extra and can double the total bill.
Platform Subscription Costs
Gumloop: starts at $97/month. Business plans from $297/month. Enterprise custom.
Relay.app: free tier available. Pro from $30/month per user. Teams from $70/month per user.
Lindy: starts around $50/month. Voice-heavy usage can push costs above $500/month.
n8n Cloud: from $24/month. Self-hosted community edition is free.
Stack AI: enterprise-focused. Expect $1,000+/month minimum.
Vellum: developer-focused platform. Starts around $500/month.
Dust: team-focused. From $29/month per user.
Make.com: from $9/month. Scales with operations volume.
Hidden Cost: LLM Tokens
Most platforms do not include LLM token costs. You pay OpenAI, Anthropic, or Google separately.
Typical agent task uses 5,000 to 50,000 tokens. At OpenAI GPT-4 pricing, that is $0.05 to $0.50 per task.
Running 1,000 tasks per day adds roughly $1,500 to $15,000 per month in tokens alone.
Cheaper models (GPT-4o mini, Claude Haiku) cut these costs by 5 to 10x. Choose your model carefully.
Hidden Cost: Integrations and Add-Ons
Some platforms charge extra for premium integrations.
Voice features (Lindy, Retell) are usually billed per minute on top of the subscription.
High-volume webhook usage may require higher tiers.
Enterprise features (SSO, audit logs, SOC 2) are typically locked behind top tiers.
Realistic Budgets
Solopreneur: $50 to $150/month total. Pick Relay free tier, n8n self-hosted, or Gumloop starter plus a small LLM budget.
Small business (5 to 20 employees): $200 to $800/month. Gumloop or Make plus LLM tokens plus one specialised tool.
Scale-up (20 to 100 employees): $1,000 to $3,000/month. Platform subscription plus tokens plus human time to maintain workflows.
Enterprise (100+ employees): $5,000 to $20,000+/month. Enterprise platform, high token volume, dedicated ops team.
How to Reduce Costs
Use cheaper models for simple tasks. Most tasks do not need the frontier model.
Cache common prompts. Reuse results where possible.
Self-host if you have technical staff. n8n self-hosted removes subscription cost entirely.
Consolidate platforms. Running three overlapping tools is more expensive than picking one.
Measure ROI per workflow. Kill workflows that do not pay back.
Automate Operations and Get Discovered by AI
Agent Console HQ combines workflow automation with AI search visibility. Your AI agents do the work AND appear in ChatGPT, Claude, Perplexity, and Gemini when your customers search. Run your operations and your acquisition from one platform.
Explore Agent Console HQFrequently Asked Questions
n8n self-hosted is free. For cloud-hosted, Make and Relay both have free tiers. Gumloop and Lindy start at $50 to $100 per month.
Usually not. You pay LLM token costs separately to OpenAI, Anthropic, Google, or another provider. Always factor this into your budget.
At high volume, yes. You trade subscription fees for infrastructure and maintenance costs. Break-even is usually around 5,000 executions per month.
$200 to $800 per month is a realistic starting range, including tokens, once you have a few workflows running.
Yes. Most platforms have generous free tiers that let you build and test workflows before committing. Start free, upgrade when you hit limits.
Linearly with volume. If you double your workflow executions, expect to roughly double your bill. Token costs scale the same way.
A few platforms bundle LLM usage into their pricing (some enterprise tiers). For most, it is separate. Read the fine print before signing.