AI Automation for Small Businesses
AI automation takes repetitive, time-consuming work off your plate, including tasks that used to need someone to read, judge or write. It can sort incoming information, answer routine enquiries, draft follow-ups and take actions across the tools you already use.
Not every step needs AI, and people stay involved where judgment, approval or exceptions matter. If it isn't worth automating, I'll tell you not to build it.
Repetitive work gets expensive quietly
A 30-minute task may not seem like much. But when your team keeps doing the same work like reading enquiries, checking details, drafting follow-ups and updating information, the hours add up. AI automation can take on the repetitive parts, so your team can spend that time on work that actually needs them.
Interactive calculator
What does one repetitive task cost over a year?
Yearly cost of this task
C$10,400/year
How an AI automation project works
01 · Understand the work
We look at how the task happens today, who is involved, what tools are used and where time or extra work is being created.
02 · Decide what should change
I work out what can be automated, where AI can help, what should stay with a person and whether the expected value justifies the cost and risk.
I also check whether an existing feature, integration or simpler automation could solve part of the problem first.
03 · Build & test
For a Build project, I connect the agreed tools, build the automation and test normal cases, exceptions and likely failure points.
04 · Launch & hand off
Your team gets the working system, documentation, training and agreed launch support. People stay involved wherever business judgment, approval or exception handling is needed.
Services & Pricing
AI Opportunity
Scan
C$350
Best for
One repetitive task you want to check.
Scope
One existing task or workflow.
Includes
- Current process & bottlenecks
- Automation readiness
- Existing tools & simpler options considered
- Where AI, rules or human review may fit
- Key risks & exceptions
- Likely time or capacity impact
- Recommended next step
The recommendation may be to automate it, simplify the process first, use something you already have or leave it alone. If it isn't worth automating, I'll tell you not to build it.
Timeline
5 business days
AI Opportunity
Map
From C$3,000
Best for
You know there's wasted time, but not where to start.
Scope
One business area · Up to 3 priority opportunities taken through deeper assessment
Includes
- Current process discovery
- Automation opportunities identified across the business area
- Priority opportunities compared by value, effort and risk
- Tool & integration feasibility
- Estimated time or capacity recovered
- Expected implementation & ongoing costs
- Dependencies between opportunities
- Prioritized 90-day or 6-month roadmap
Timeline
3+ weeks
AI Automation
Build
From C$5,000
Best for
You know what you want automated.
Scope
One agreed automation. If more discovery or scoping is still needed, that work happens before implementation starts.
Includes
- Working automation & integrations
- Business rules & AI capabilities where needed
- Human review & approval points
- Exception & failure handling
- Testing with normal cases & agreed edge cases
- Documentation & team training
- 30 calendar days of launch support
Timeline
4+ weeks
Automation
Care
C$500/month
Best for
You already have a live automation.
Scope
Monthly support for one live automation that I built.
Includes
- Monthly automation health check
- Review of configured failure & exception alerts
- Up to two hours of minor maintenance & troubleshooting
- Usage & third-party cost review
- Monthly Automation Health Report
Additional work is quoted separately. Care covers automations I built, not automations built by someone else. Care is scheduled monthly support. It is not continuous monitoring, emergency support or an uptime guarantee.
Timeline
Ongoing, monthly
Frequently Asked Questions
Does every automation need AI?
No. AI is useful when part of the work benefits from understanding information, generating an output, making a bounded decision or choosing what action to take next.
Other parts may be better handled with regular software, integrations or business rules. If a simpler approach handles a step well, there is no reason to add AI to it.
What's the difference between workflow automation, AI automation and AI agents?
Workflow automation follows a defined process using software, integrations and business rules.
AI automation is broader: it uses AI for the kind of work described in the previous answer, not only fixed steps.
An AI agent is one type of AI automation. Instead of following only a fixed sequence, it can decide which action or tool to use next based on a goal and the information available. An agent may also be one part of a larger automation.
Agents are useful for some problems, but they are not automatically better than a well-designed workflow.
How do I know if something is worth automating?
A good candidate usually happens often enough to matter, follows a process that can be understood and uses information or systems the automation can access reliably.
There also needs to be a practical way to handle mistakes, unusual cases and decisions that still need a person.
Frequency alone does not make something worth automating. The expected value, implementation cost, ongoing cost and risk all matter. That is one of the things I check in an AI Opportunity Scan.
Can you work with the tools we already use?
Yes, and I check that first. If the tools you already use can solve part or all of the problem, I recommend using them before anything new is built.
If a new tool or integration is justified, I explain why, what it costs and what would change before anything is built.
What do you need from me?
I need to understand how the work happens today, see realistic examples and get access to the systems we agree are needed.
You stay involved when a decision affects how the business works, what should remain human, risk, cost or launch. I handle the implementation and testing within the agreed scope.
Are there ongoing costs beyond your fee?
Usually, yes.
Third-party software, API, hosting and usage fees are separate unless the project scope says otherwise. I identify the expected ongoing costs before you approve a Build.
What happens to my business data and accounts?
Before implementation, we review what information the automation uses, which systems process it, who needs access and whether any information requires extra care.
Your business keeps ownership of its data. Where the tools allow it, production accounts stay in your business's name and my access can be removed after handoff.
What happens if the automation makes a mistake?
No automation is error-free.
Depending on the task and risk, the system may use validation rules, human review, exception handling, logs or alerts.
Before launch, we agree on what the automation can handle, when it should stop and when a person needs to step in.
What happens after launch?
A Build includes documentation, team training and 30 calendar days of launch support.
After that, you can manage the automation yourself or use Automation Care for scheduled maintenance and support.
Do you offer a bundle discount?
Yes. When you combine services (AI Search and AI Automation together), or more than one tier in the same engagement, I offer a discount on the combined price.
Ask on a call and I'll put a package together for what you're considering.
Let's Talk
Have one repetitive task in mind?
Start with an AI Opportunity Scan. Find out whether it is worth automating, what the practical options are and what should happen next.