We're a small AI consulting practice working with owner-operators across the Portland metro area and Southwest Washington. This post is the short answer to the question we get more than any other: what are small businesses around here actually doing with AI right now?

Not what the AI agencies are pitching, not what shows up in McKinsey decks. What real businesses with one to fifty employees and no IT team are doing in 2026.

The short version

Three patterns cover roughly 80% of what we see:

Draft-and-review for repetitive writing. Email replies, listing descriptions, engagement letters, social posts. The owner or staffer reviews and edits before anything sends.
Intake and triage automation. AI-assisted form intake, document collection, conflict and recall checks. Cuts hours per week of front-office time.
Recurring-deliverable prep. Quarterly client reviews, monthly bookkeeping summaries, market commentary, board prep. AI assembles, the human interprets and decides.

What we don't see working well: replacing human judgment, customer-facing AI chat that pretends to be a person, anything that ships unreviewed.

What it looks like in specific industries

The implementations look different depending on what you do, but the pattern is the same: AI drafts, a human reviews, the business keeps moving without grinding down the owner or the staff.

Law firms

Solo attorneys and small firms across Lake Oswego, Portland, and Beaverton are using AI for first-draft document work (demand letters, engagement letters, discovery responses), intake and conflict checks, and prompt libraries for recurring matter types. The setups respect bar confidentiality rules by using enterprise-tier accounts that don't train on client data. The attorneys we work with describe the time savings as somewhere between three and ten hours per week per attorney. (More on AI for law firms.)

Real estate offices

Agents and small brokerages are using AI to draft listing descriptions, prep comparative market analyses, run CRM follow-up sequences, and write neighborhood guides. The pricing decisions stay with the agent. The writing and data assembly that used to take an hour per listing now takes ten minutes. (More on AI for real estate.)

Accounting and bookkeeping

Small practices in Tigard, Hillsboro, and Vancouver are using AI to draft engagement letters, automate document collection, and prep recurring client deliverables. The AI assembles the data, the accountant interprets and signs. Year-end and quarterly cycles run noticeably smoother. (More on AI for accountants.)

Dental and healthcare practices

Front-office workflows are the high-impact target: recall and reactivation, insurance verification prep, after-hours triage, and review responses. All set up with HIPAA-compatible tooling, never with consumer-grade accounts. (More on AI for dental practices.)

Gyms and fitness studios

Member communication is the big lever: welcome sequences, check-ins, lapsed-member outreach, content drafting. The personal coaching relationship stays human. The repetitive admin doesn't. (More on AI for gyms.)

Financial advisors

Independent advisors use AI for client review meeting prep, newsletter drafting, and onboarding paperwork. Everything routes through compliance review before sending. (More on AI for financial advisors.)

What's not working

Three patterns we keep seeing fail:

Full AI receptionist replacement. Doesn't work for businesses with real customer emotion in the calls. Works only for narrow FAQ and appointment scheduling.
Generic chatbots trained on your website. They hallucinate, they sound off-brand, and the buyer never gets the answer they actually needed.
Buying an AI tool first and figuring out what to do with it second. The tool sits unused. We see this constantly. The right order is workflow first, tool second.

What to start with if you haven't

The fastest, lowest-risk first step is usually one of these:

Spend thirty minutes on a free consult to figure out where AI fits in your specific operation. If it doesn't, we'll tell you.
Run a $375 paid audit. You get a written report within five business days with three ranked opportunities and a recommendation.
Pick one repetitive workflow that takes you or a staffer hours per week and automate just that one. Don't try to do everything at once.

If you're going to do this yourself, start with ChatGPT or Claude on the team's hardest writing task. Track how long it takes to get a draft you'd send. After two weeks, you'll have your own data on whether it's worth investing further.

Common questions

Is AI ready for small business use in 2026? Yes, for specific, bounded use cases. The tools have stabilized enough that a careful implementation pays back within a quarter. The hype is still real, and the wrong tools or wrong setups still fail, but the underlying capability is no longer the bottleneck.

Will I need to replace my staff if I use AI? Almost never. The pattern we see is one person becoming as productive as two, not one person being replaced. For owner-operators who are doing too much themselves, AI buys back hours per week.

What's a realistic budget for a first AI project? Most projects we run land between $500 and $1,500. The $375 written audit is a way to know what to spend on first without committing.

If you're a Portland-area small business and want to talk through what makes sense for you, book a free 30-minute call. No pitch.

How small businesses in Portland are actually using AI in 2026