What AI Actually Does for Small Businesses

There's a lot of noise about AI right now. Some of it is useful. Most of it assumes you're running a Fortune 500 marketing department or a tech startup with three engineers and an innovation budget.

If you're an accountant, a bookkeeper, an insurance broker, a financial advisor, or a consultant running a solo or 2-5 person practice, most of that advice isn't useful. The tools are real. The time savings are real. But the translation is missing.

This is the plain-English version.

Where Your Time Actually Goes

Before getting into what AI does, it helps to be specific about what eats your hours.

For most small businesses, the revenue bottleneck isn't expertise - you have plenty of that. It's time. And specifically, it's the time that goes to everything around the client work rather than the client work itself.

That means: email. Proposals. Follow-ups. Client onboarding materials. Marketing content you know you should be producing but can't seem to find time for. Administrative back-and-forth. Status updates. Explaining the same thing to a new prospect that you explained to fifteen clients before them.

Every hour you spend on that list is an hour you're not billing - or resting.

That's where AI earns its keep. Not by replacing your judgment or your expertise. By cutting the time it takes to do the work that surrounds your expertise.

What AI Is Actually Good At

Writing the first draft of anything

The most immediate win for small businesses is first drafts. Proposals, client emails, follow-up sequences, FAQ documents, service descriptions, newsletter content - anything where you know what you want to say but the blank page takes too long.

AI doesn't write these things for you. You still shape the content, fix the specifics, and make it sound like you. But getting from zero to 80% is the part that takes time. Handing that part to AI and then editing the result is almost always faster than writing from scratch.

The key - and this is where most people underestimate the tool - is giving AI your voice first. Before you ask it to write anything, paste in 2-3 sentences from your own writing: an email you're proud of, a paragraph from a proposal that worked. Add a line about your business and your clients. AI is a fast mimic. Give it something to mirror and the output sounds like you. Give it nothing and it sounds like a press release.

Turning a brain dump into a structured document

Accountants and financial advisors do this constantly: you have a clear picture in your head of what a client needs to understand, but translating that into a readable summary or a memo takes time you don't have.

Talk to AI in plain English the way you'd explain it to the client. Then ask it to turn that into a structured document in the format you need. First draft in two minutes instead of twenty.

Handling repetitive explanations

If you have a piece of information you explain to every new client - what your onboarding process looks like, how your fee structure works, what documents you need from them - AI can help you build a reusable document that covers it once and covers it well.

Set it up once. Send it every time. Stop explaining the same thing from scratch on every intake call.

Researching a topic before a client conversation

Your clients hire you partly for your thinking ability - taking complex information and translating it into something useful for their specific situation. AI accelerates the research portion of that.

Ask it to give you a plain-English overview of a topic you're less familiar with, or a list of questions you should be asking before a client meeting. You still bring the judgment. AI helps you get to the starting line faster.

What AI Is Not Good At

This part matters as much as the previous section.

AI makes things up. This is not a quirk that will be fixed in the next update - it is a fundamental characteristic of how large language models work. They generate plausible-sounding text. When they don't know something, they often produce an answer anyway. For a financial advisor giving a client specific tax guidance, or a broker quoting specific policy terms, or a consultant recommending a specific vendor, that's a serious problem.

AI should not be the source of facts in high-stakes client deliverables. It should be the structure, the draft, the starting point - never the authoritative data.

AI also doesn't know your clients, your history with them, or the subtext of your relationships. It can help you write a follow-up email. It can't tell you that this particular client responds better to directness or that they're going through a difficult quarter. You bring that. The tool helps you communicate it faster.

The Setup That Changes Everything

One of the most useful things a small business owner can do is build what I call a base context: a short document - usually 3-5 paragraphs - that describes your business, your clients, your voice, and how you work.

You load that context once at the start of every AI session. From that point on, the tool already knows who you are and who you're talking to. You stop getting generic output. You start getting answers that are actually calibrated to your practice.

This is architecture. The same up-front thinking that goes into designing software, a house, or anything that needs context to work. Get it right at the start and every decision after that gets easier.

Takes about 20 minutes to build. Saves time every single day after that.

This is one of the things I cover in my AI workshop - along with how to ask for options before you commit to a direction, how to tell AI to push back on you rather than just agreeing with everything you say, and how to work through longer documents without the tool going off track.

The Practical Starting Point

If you've been putting off getting serious about AI tools because it feels like one more thing to figure out, here's where I'd actually start:

Pick one type of document you produce regularly - a proposal, a client update email, a monthly summary. The next time you need to write one, try this: open ChatGPT or Claude, paste in two sentences from a previous version you liked, add a quick description of your business and the client, then ask for a first draft.

See how much time it saves and how much editing it needs. That's your baseline.

From there, you can decide whether it's worth going deeper.

---

If you want to go deeper in person - in a session crafted specifically for small business owners like you - I'm hosting an AI workshop for Concord Chamber of Commerce members on June 10, 2026.

Title: "Finally, AI That Makes Sense: Plain-English from a Tech Expert Who Uses It Every Day"

Wednesday, June 10, 2026, 10:00 AM to 11:30 AM. Free for Chamber members.

Register at klass.link/ai.

The first 2 businesses to book a Small Business Efficiency Checkup after the seminar pay $600 (regular price $750). Details at the session.