AI for small business
Where to start without a big budget: one process, a measured pilot, then the next one.
Where should a small business start with artificial intelligence?
Start with the task that eats the most hours and follows clear rules. To find it, ask your team what they do every day that feels like copy and paste. The answers tend to repeat: answering the same customer questions, moving orders from email into the ERP, logging invoices, preparing similar quotes or updating a spreadsheet.
A good first task meets three conditions: it happens many times a week, it follows logic you can explain, and a mistake can be caught and fixed without serious harm. If you can also measure how long it takes today, you have a clear savings target. Our AI readiness assessment shows where you stand.
02What are the cheapest and most profitable automations?
These are the ones we recommend most often as a first step in businesses of 5 to 50 people:
- Customer replies on WhatsApp or web: an agent handles common questions, opening hours, prices and order status, and hands the rest to a person.
- Sorted email with drafts: AI labels incoming mail, prepares replies to the usual questions and flags urgent messages. See AI email automation.
- Supplier invoice capture: automatic reading and posting to your accounting software, or delivery to your bookkeeper.
- Quotes from a request: AI reads what the customer asks for and fills your template with products and prices from your price list.
- Connecting your tools: a web order creates the invoice, alerts the warehouse and updates the CRM without anyone copying it.
How much does it cost to implement AI in a small business?
The cost depends on the number of processes, channels, integrations and volume, and on whether you need an automation or an AI agent. We start with a free 30-minute call to review your business, then send a written proposal with a fixed budget per phase, agreed before we start.
On top of that comes AI model usage, paid directly to the provider, which in a small business is usually modest. You can work out the return for your case with the automation ROI calculator, or tell us about your case.
04How does the phased method work?
We split the project into four phases so you do not have to bet the whole budget at once. In the diagnostic we review processes and tools and set a savings target in writing. In the pilot we automate a single task and measure it with real data for a few weeks. If it hits the target, we move to scaling and extend to more tasks or departments. The last phase is operation: maintenance, adjustments and monthly follow-up.
Each phase pays for itself before the next one starts. If a pilot does not deliver the expected result, you find out early and at low cost. The detail is on our method page.
05Is there funding for small businesses adopting AI?
It depends on where you operate. Some countries and regions run grants, tax relief or subsidised advice for small business digitalisation, and some cover AI tools.
Amounts, eligibility and deadlines change with every scheme, so we do not quote figures that may be out of date. Check current programmes with your local business support body, chamber of commerce or government small business portal. If you find one that fits, we help you scope the project to meet its conditions, although the application itself is handled by your company or your accountant.
06Example: a 15-person distributor
An electrical supplies distributor in Birmingham receives around 60 orders a day by email, WhatsApp and phone. Two admin staff key them into the ERP by hand and answer customers asking whether their order has shipped.
In the diagnostic we measure about 25 hours a week on those two tasks. The pilot automates reading orders from email: the AI identifies customer, item codes and quantities, loads them into the ERP and flags doubts. After six weeks, 80% of email orders go in untouched. In phase two, a WhatsApp agent that checks order status is added. The example is illustrative; every company starts from different volumes.
How we do it, step by step
- Diagnostic
We review processes and tools and set a savings target in writing.
- Pilot
We automate a single task and measure it with real data.
- Scaling
If it meets the target, we extend to other tasks or departments.
- Operation
Maintenance, adjustments and a monthly review of results.
| DIY with ChatGPT | Closed SaaS tool | Phased project | |
|---|---|---|---|
| Upfront cost | Almost none | Low | Fixed budget per phase |
| Connected to your systems | No | Partly | Yes |
| Fitted to your process | No | Little | Yes |
| Measurable saving | Hard to measure | Varies | Written target |
Frequently asked questions
How much does artificial intelligence cost for a small business?
It depends on the number of processes, channels, integrations and volume, and on whether you need an automation or an AI agent. Projects are budgeted per phase with a fixed price agreed before starting, and AI model usage is paid directly to the provider.
Do I need a technical team to use AI in my company?
No. We do the technical part and your team only needs to know how to use the result: review what the AI flags as doubtful and tell us when something changes. We train whoever will work with the system.
Which is better, ChatGPT or a custom project?
They complement each other. ChatGPT works for individual tasks such as drafting or summarising. When you want AI to work with your data, inside your tools and without someone copying and pasting, you need a connected automation or agent. We explain it in how to implement AI in business.
Can I get a grant or subsidy to pay for AI?
It depends on the programmes open in your country or region and on whether your company and the project meet their requirements. Check with your local business support body. If one fits, we help you define the project scope.
What happens if the automation does not work?
That is why we start with a small pilot and a written savings target. If it does not reach it, we adjust it or stop there, with a limited investment. We do not move to the next phase until the previous one has paid for itself.
Keep exploring
AI consulting services
AI consulting that doesn't stop at a slide deck: strategy, pilot, scale and operation with one team.
Business process automation with AI
AI workflows that read, classify and log data across your tools, with human review on exceptions.
AI agent development
Assistants that answer on WhatsApp, web or phone and act inside your CRM, calendar or ERP.
n8n automation agency
n8n workflows hosted in the EU, with AI, connected to your CRM, ERP and email, and maintained by us.
WhatsApp AI agent for business
WhatsApp chatbot for business on Meta's official API: answers, books, qualifies and hands over to your team.
Supplier invoice automation
AI invoice processing: reading, validation, approval and posting to QuickBooks, Xero, Sage or Odoo without typing.
AI email automation
Inbox sorting, AI drafts, data extraction into your systems and follow-ups in Gmail and Outlook.
AI customer service agent
Multichannel agent that resolves queries from your knowledge base, triages tickets and escalates cases.
AI for accountants
Invoice capture, reconciliation, document chasing and deadline reminders connected to QuickBooks, Xero or Sage.
AI for ecommerce
Support on orders and returns, product listings at scale and recovered carts for online stores.
AI for restaurants
Phone and WhatsApp bookings, reviews, allergen questions and supplier orders for restaurants.
AI readiness assessment
10 questions, 3 minutes: find out which AI level your business is at and what to do next.
Automation ROI calculator
Calculate hours freed, annual savings and payback months for automating a repetitive task.
AI agent
An AI agent is a program that uses a language model to understand a goal, decide which steps to take and carry them out with external tools, such as a calendar, a CRM or email, until it finishes the task or hands it to a person.
n8n
n8n is a workflow automation tool with a visual editor that connects apps, APIs and AI models. It uses a fair-code license, runs in the cloud or on your own server, and Jan Oberhauser created it in Berlin in 2019.
LLM (large language model)
A large language model (LLM) is an AI model trained on enormous amounts of text to predict the next word, which lets it understand questions and write, summarize, translate, classify and reason about text.
Pick a time yourself
30 minutes by video call. Free, no commitment.
Prefer us to get in touch? Leave your details.