Blog · AI and automation
How Much Does AI Cost for a Small Business? Price Ranges From Subscriptions to Custom Builds
The short answer
Most small businesses can start using AI for $25 to $60 per person per month in business AI assistant subscriptions. Automating a specific workflow usually costs $1,000 to $15,000 to set up plus $50 to $500 a month to run. Custom AI builds, such as an assistant trained on your own documents and connected to your systems, typically start around $15,000 and can run well past $50,000.
A practical first-year budget for a 10-person company is often $10,000 to $40,000 all in. The bigger cost is usually not the software but the time to set it up, train the team, and keep it working.
What you're actually paying for
"AI" covers everything from a $20 chatbot subscription to a six-figure software project. It helps to split the cost into five buckets:
- Tools: per-user subscriptions to AI assistants and AI features inside software you already use.
- Automation platforms: the services that connect your systems and run workflows, usually priced by tasks or runs per month.
- Usage: pay-as-you-go charges when a workflow or app calls an AI model directly, billed by the amount of text processed.
- Setup and build: the hours to design, build, test, and connect workflows, whether your team does it or you hire help.
- People time: training, adoption, reviewing AI output, and maintenance after launch.
The first three are easy to see on an invoice. The last two are where budgets go wrong.
AI cost ranges by level
On our AI implementation work we group workflows into three levels. Each has a different cost shape.
| Level | What it looks like | Typical setup cost | Typical ongoing cost |
|---|---|---|---|
| Everyday AI assistants | Team uses a business AI assistant to draft, summarize, research, and analyze | $0–$2,500 for setup, policy, and training | $25–$60 per user per month |
| AI features in tools you own | Turning on AI in your CRM, help desk, accounting, or meeting tools | $0–$3,000 to configure | $0–$50 per user per month in add-on fees |
| Human intelligence tasks (HIT) | AI drafts and a person reviews, such as support replies or proposals | $1,000–$5,000 per workflow | $50–$300 a month in tools and usage |
| Full automation (RPA) | Rules-based workflows that move data between systems with no one involved | $1,500–$10,000 per workflow | $50–$500 a month in platform fees |
| AI-intelligent workflows (AIIW) | AI reads, classifies, and decides within limits, such as triaging inbound requests | $5,000–$15,000 per workflow | $100–$1,000 a month in platform and usage fees |
| Custom AI build | An assistant or app built on your own data and connected to your systems | $15,000–$75,000+ | 10–20% of build cost per year for hosting and maintenance |
These are typical ranges for companies with $250K to $25M in revenue, based on what we see in client work. Your number depends on how many systems are involved, how clean your data is, and how much of the build your own team can handle. Vendor prices change often, so check current pricing before you budget.
The hidden costs most budgets miss
- Data cleanup. AI and automation are only as good as the data they read. Duplicate CRM records, missing fields, and inconsistent naming often need fixing first. The AI readiness checklist shows what to look at.
- Training and adoption. Plan for 2 to 4 hours of training per person up front and regular check-ins for the first few months. Tools nobody uses are the most expensive kind.
- Review time. Human-in-the-loop workflows save time, but someone still reviews the output. Budget that time instead of assuming it's zero.
- Maintenance. Workflows break when a connected app changes, a form field is renamed, or a process changes. Expect a few hours a month per active workflow.
- Overlapping subscriptions. Many teams end up paying for three AI assistants that do the same thing. See how much a small business should spend on software for healthy ranges.
- Security and policy. Deciding what data can go into which tools takes a little time now and prevents expensive mistakes later. A short AI use policy covers it.
A sample first-year AI budget
Here's what a realistic first year might look like for a 10-person service business that wants to start with everyday AI and automate three workflows.
| Item | Estimate | Notes |
|---|---|---|
| Business AI assistant for 10 people | $3,600 | $30 per user per month for 12 months |
| Setup, AI policy, and team training | $2,000 | One-time |
| Lead follow-up workflow (AIIW) | $6,000 | Instant reply, qualification, and booking |
| New-client onboarding automation (RPA) | $3,000 | Creates records, sends onboarding emails, updates reporting |
| Proposal drafting workflow (HIT) | $2,500 | AI drafts from a template; a person reviews |
| Automation platform and AI usage | $3,600 | About $300 a month |
| Maintenance and tuning | $2,400 | About 4 hours a month |
| Total, year one | $23,100 | Year two drops to about $9,600 once setup is done |
If those three workflows and the AI assistant save the team 20 hours a week, at a $35 blended hourly cost over 48 working weeks, that's $33,600 a year in time. The first year more than pays for itself, and every year after costs far less. To see where hours are going in your own business, read how much time AI can save a small business.
How to keep AI costs under control
- Start with the tools you already pay for. Many CRMs, help desks, and office suites now include AI features. Turn those on before buying anything new.
- Pick one or two workflows, not ten. Choose the ones with the most hours and the clearest rules. Where to start with AI walks through how to choose.
- Do the ROI math before you build. Estimate hours saved, multiply by the hourly cost, and compare with setup and running costs. Use the method in how to calculate the ROI of automating a workflow.
- Choose the lowest level that works. A human-in-the-loop workflow often gets 80% of the benefit for a fraction of the cost of a fully autonomous one.
- Set a monthly usage cap. Most automation platforms and AI APIs let you cap spending. Use it, and review usage monthly for the first quarter.
- Assign an owner. Someone on your team should own each workflow, watch the results, and flag when something breaks.
Should you build, buy, or hire help?
| Option | Best when | Trade-off |
|---|---|---|
| Do it yourself | Simple workflows, a tech-comfortable team member, and time to learn | Lowest cash cost, but slower and easy to stall |
| Buy an off-the-shelf AI tool | A common problem with a proven product, such as meeting notes or phone answering | Fast to start, but you adapt to the tool |
| Hire a freelancer or agency to build | A clear spec and someone to manage the project | Can be cost-effective, but quality and handoff vary |
| Work with an implementation consultant | Several workflows, unclear priorities, or a team that needs training and change support | Higher up-front cost, with one person accountable for results |
Frequently asked questions
How much does AI cost for a small business?
Most small businesses start with business AI assistant subscriptions at about $25 to $60 per person per month. Automating a single workflow usually costs $1,000 to $15,000 to set up plus $50 to $500 a month to run, and custom AI builds typically start around $15,000. A first-year budget for a 10-person company is often $10,000 to $40,000 in total.
What is the cheapest way for a small business to start using AI?
Turn on the AI features in software you already pay for, such as your CRM, help desk, or office suite, and give your team a business AI assistant with a short AI use policy and a few hours of training. That can cost less than $50 per person per month with little or no setup fee.
How much does it cost to automate a workflow with AI?
A human-in-the-loop workflow, where AI drafts and a person reviews, typically costs $1,000 to $5,000 to set up. A rules-based automation that moves data between systems usually costs $1,500 to $10,000. A workflow where AI reads, classifies, and makes decisions within limits usually costs $5,000 to $15,000. Ongoing platform and usage fees run about $50 to $1,000 a month depending on volume.
What are the hidden costs of AI?
The most common hidden costs are cleaning up data before AI can use it, training the team and driving adoption, the time people spend reviewing AI output, maintaining workflows when connected apps change, and paying for overlapping AI subscriptions that do the same thing.
Is AI worth the cost for a small business?
Usually yes, when it is pointed at specific, high-volume work. If AI saves a team 20 hours a week at a $35 blended hourly cost, that is worth about $33,600 a year, which covers a typical first-year budget of around $23,000 and costs far less in year two. Calculate the return for each workflow before you build it.
How much does a custom AI chatbot or assistant cost?
A custom AI assistant trained on your own documents and connected to your systems typically costs $15,000 to $75,000 or more to build, plus about 10 to 20 percent of the build cost each year for hosting, usage, and maintenance. Many small businesses can get most of the benefit from an off-the-shelf tool configured with their own content for much less.
Know what AI will cost and save before you spend
The free Profit Check estimates how many hours manual work is costing you and what automation could save, in about 3 minutes. The AI Operations & Profit Audit ($7,500, about 3–4 weeks) ranks your AI and automation opportunities by margin impact, with a 30/60/90-day plan and a realistic budget.
Take the Free Profit Check See audit optionsCost ranges are general estimates for companies with $250K to $25M in revenue and vary by vendor, industry, data quality, and scope. AI tool prices change frequently; check current vendor pricing before budgeting. This article is not financial advice.