AI Readiness Checklist for Small Business (2026)

A 27-item checklist across 10 categories, with a simple scoring rubric so you can see if your small business is actually ready to adopt AI, or if you need to shore up the basics first.

How to use it

1. Read each item and mark it done or not done — no half credit.
2. Add up the checked items across all 10 sections.
3. Score: 0–10 = not ready, fix the basics first · 11–20 = pilot-ready, pick one workflow · 21+ = ready to scale AI across the business.

01 · Business clarity — You know what you're trying to fix

AI amplifies the process you already have. Fuzzy goals produce fuzzy pilots that never graduate.

• ☐ Named the top 3 tasks that eat the most hours per week. Every good AI pilot maps to a specific, painful task — not "use AI more."
• ☐ Written a one-sentence outcome for each: "save X hours" or "add $Y revenue." Without a target, you can't tell a working pilot from a fun demo.
• ☐ Owner-level buy-in on which task to attack first. Small teams can only run one AI pilot at a time without stalling.

02 · Data — Your data is reachable, not perfect

You don't need a data warehouse. You need the answers, invoices, notes, and transcripts to live somewhere an AI tool can read.

• ☐ Customer records live in one CRM (not just email + memory). AI copilots need a canonical customer list to enrich or draft against.
• ☐ Past proposals, SOPs, and FAQs are in shared docs (Notion, Drive, Docs). This is the corpus your custom GPT or RAG bot will actually use.
• ☐ Call recordings or transcripts exist for sales and support. Meeting AI (Fathom, Granola, Fireflies) turns calls into training data.

03 · Tools & accounts — You have the accounts you need, and no more

Most small teams don't need more AI subscriptions. They need one paid seat per operator and a plan to consolidate.

• ☐ At least one paid ChatGPT / Claude / Gemini seat per operator. Free tiers throttle context and file uploads — the two things that matter.
• ☐ An automation tool set up (Zapier, Make, or n8n). AI without automation stays a chat window. Automation is how it hits your CRM.
• ☐ Audited existing SaaS for AI features you already pay for. HubSpot, Notion, Gmail, and Canva ship AI you might be double-paying to replicate.

04 · Workflows — You've mapped one process end-to-end

Pick one workflow — lead intake, proposal, onboarding — and diagram every step before you introduce AI.

• ☐ Diagrammed the current workflow with owners and hand-offs. AI is only useful at specific steps; a fuzzy map hides where to insert it.
• ☐ Marked which steps are rules-based vs judgement-based. Rules-based = automate. Judgement = draft-then-edit. Different tools.
• ☐ Chosen 1–2 steps as the pilot — not the whole workflow. Automating one step in week one beats automating none in month six.

05 · Security & privacy — You know what you can and can't paste into an LLM

AI adoption stalls the fastest on a security review. Get ahead of it with clear rules and a paid tier that doesn't train on you.

• ☐ Using a paid business tier that excludes your data from training. ChatGPT Team, Claude for Work, and Gemini Business all offer this — free tiers don't.
• ☐ Written rules for what data types are OK to paste (and what isn't). Staff need a one-pager, not tribal knowledge, before you scale usage.
• ☐ Reviewed vendor DPAs / subprocessors if you handle PHI, PII, or regulated data. HIPAA, GLBA, and California CCPA all constrain which AI vendors you can use.

06 · Website & content — Your site is legible to AI crawlers

ChatGPT, Perplexity, and Claude increasingly cite websites in answers. If your site is a JS shell with thin copy, you don't get cited.

• ☐ Every important page renders real HTML (not just a client-side shell). AI crawlers read HTML the way Google used to — no HTML, no citation.
• ☐ A public /llms.txt or clear content map exists. Gives AI crawlers a curated list of the pages you actually want them to read.
• ☐ Long-form service and industry pages exist — not just a homepage. Answer-engine citations reward pages that answer one specific question well.

07 · People & skills — One person owns AI on the team

You don't need a Head of AI. You need one accountable operator with 2–4 hours a week.

• ☐ One named owner for AI experiments (with time carved out). AI projects without a single throat to choke drift and die.
• ☐ Team has done at least one hands-on prompt-writing session. Prompting is a skill; two hours of practice beats a year of reading tips.
• ☐ Comfortable saying "the AI wrote a draft, I edited it" to clients. AI adoption stalls when staff feel they have to hide it.

08 · Measurement — You'll know if the pilot worked

If you can't measure the pilot, you can't defend the spend or scale what works.

• ☐ Baseline metric captured before the pilot starts. Hours per week, response time, or close rate — pick one and write it down.
• ☐ A weekly 15-minute review scheduled for the first 6 weeks. Short, frequent checkpoints catch drift before it becomes a wasted quarter.
• ☐ A kill criteria decided in advance ("if X hasn't improved by week 4, we stop"). Pre-committing to stop prevents sunk-cost pilots that limp for months.

09 · Budget — You've sized the real cost, not just the sticker

Software is the smallest line item. Setup time and change management usually cost more than the seats.

• ☐ Budgeted setup time in hours, not just SaaS dollars. Most AI pilots need 10–40 hours of process + prompt work up front.
• ☐ Set a 90-day spend ceiling and reviewed it with the owner. AI tools are cheap individually, expensive in aggregate — cap it early.
• ☐ Comfortable with a 3–6 month payback window, not 30 days. Well-scoped pilots pay back in a quarter or two — not instantly.

10 · Next step — You have a first pilot picked

The whole checklist collapses to this: is there one concrete thing you'll try in the next 30 days?

• ☐ One pilot task, one owner, one deadline in the next 30 days. Momentum matters more than perfection. Start small, ship, then expand.
• ☐ Booked a review at day 30 to score it against the baseline. A scheduled review is the difference between a pilot and a hobby.

---

Want us to score your business? Book a free 15-minute intake — we'll walk the checklist with you, score it live, and hand you a one-page AI-readiness plan with the first pilot to run.