How to Use AI for Your Small Business (Without a CS Degree)

You’ve heard the hype. AI is changing everything, and every vendor you work with now has an AI feature to sell you.

Fine. But what does that actually mean for your business?

If you run an established local business, maybe 10 to 100 people, the AI conversation can feel like it’s happening in another language. Six-figure enterprise platforms aren’t relevant. Technical explainers aren’t either.

What you need is a practical way to answer two questions: where does AI fit in how we work, and how do we start without wasting money?

Step 1: Look at Where the Time Actually Goes

Before you touch a new tool, spend a week paying attention to how you and your team spend time. Ask each person to jot down the tasks that eat their day.

You’re looking for three kinds of work:

1. Repetitive tasks. The same steps, over and over: data entry, scheduling, pulling reports.
2. Communication-heavy tasks. Follow-ups, reminders, answering the same questions.
3. Writing and summarizing. Proposals, listings, emails, meeting notes, documentation.

These are where current AI tools are most reliable. Work that requires judgment, relationships, or accountability is where they’re weakest.

Step 2: Check What You Already Own

This is the step most businesses skip, and it’s often the cheapest win.

Much of the software you already pay for now includes AI features:

• Microsoft 365 includes Copilot Chat for business users, with a paid Copilot add-on that works inside Outlook, Word, Excel, and Teams.
• Google Workspace business plans include Gemini in Gmail, Docs, and Meet.
• Your phone system, CRM, practice management, or accounting software may have added call summaries, drafting, or automation features in the last year.

Before buying something new, find out what’s already turned on, what’s turned off, and what it would take to use it well.

Step 3: Pick One First Win

Don’t try to automate everything at once. Pick one workflow that:

• Takes at least 2–3 hours a week
• Is repetitive and predictable
• Doesn’t depend on deep expertise or judgment
• Would noticeably improve someone’s day if it went faster

Common first wins look like this:

Email and communication

• Drafting replies to common customer questions
• Follow-ups for leads or clients
• Appointment reminders and confirmations

Content and marketing

• Social posts and newsletter drafts
• Service descriptions or listing copy
• Review responses

Admin and operations

• Moving information from forms into spreadsheets or your CRM
• Meeting summaries and action items
• First drafts of internal documentation

Step 4: Match Tools to Problems

A common mistake: buying the biggest AI platform available and expecting it to solve everything. It won’t.

Look for specific tools that solve specific problems, and prefer ones that fit into systems you already use. Some that often come up:

• ChatGPT, Claude, Copilot, or Gemini for drafting, summarizing, and research. Pick the one that fits the software your business already lives in.
• Zapier or Make for connecting tools and moving information between them.
• Scheduling tools like Calendly, or the one built into your existing software, to end the back-and-forth.
• Meeting transcription from Teams, Zoom, Google Meet, or a dedicated tool like Otter.
• Canva for quick graphics.

A $20/month tool that handles one workflow well is worth more than a $500/month platform nobody learns.

Step 5: Set Up, Don’t Just Sign Up

This is where most DIY attempts stall. Someone signs up, pokes around for 20 minutes, and never opens it again.

To get real value:

1. Block real time for setup, a couple of hours at least.
2. Start with one use case. Ignore the rest of the features for now.
3. Write it down. If you’re using a chat tool, save the prompts that work for your common tasks so the whole team can use them.
4. Give it two weeks of daily use before judging.
5. Measure something. Time spent, response times, errors. “It feels faster” isn’t enough.

Step 6: Set Some Ground Rules

If your staff are using AI tools, and they probably already are, you need simple rules: which tools are approved, what information can’t go into them, and who checks the output before it reaches a customer. A one-page policy is enough to start.

Step 7: Build on What Works

Once your first workflow is running, pick the next one and repeat:

1. Find the time sink
2. Check what you already own
3. Pick the right tool
4. Set it up properly
5. Measure
6. Move on

Progress here is usually steady rather than dramatic. A few well-chosen changes that stick beat a dozen tools that don’t.

When It’s Not Worth It

AI isn’t the answer to every problem. Hold off when:

• The underlying process is broken. Automating a messy process gets you a faster mess. Fix the process first.
• The volume is too low. If a task takes 20 minutes a month, it isn’t worth a new tool.
• Mistakes are costly and hard to catch. Legal advice, medical guidance, and financial decisions need human review, which limits the time saved.
• The tool isn’t mature yet. Sometimes the right answer is to wait six months.

When to Get Help

DIY works well for simple tools. It gets harder when:

• You need tools to work together across phones, website, email, and your core software
• You handle sensitive data (healthcare, legal, financial, insurance)
• You’ve tried tools that didn’t stick and want to understand why
• Every vendor is pitching an AI add-on and you want an objective view

That’s what I do. I look at your business and the whole stack it runs on, tell you what fits and what doesn’t, and help you put it in place.

The Bottom Line

AI isn’t magic, and it won’t run your business for you. But used deliberately, starting with what you already own, it can take real work off your team’s plate.

Start small. Pick one thing. Make it work. Then do it again.

If you’d like help figuring out where to start, book a strategy session. In 90 minutes we’ll go through how your business runs and you’ll leave with a short, written list of the changes worth making first.