AI for Small Business Playbook

How to Use This Template

Replace [[Tokens]] with your real operation. Start with the 2-3 highest-ROI items that match your available time. Keep humans in the loop for customer-facing and financial decisions.

What to Provide

Section 1: Use Case Matrix by Function

Marketing: content, social, email, local SEO, ads copy, customer stories.

Operations: scheduling, inventory notes, vendor comms, process docs, reporting.

Support: reply drafts, FAQ updates, onboarding sequences, feedback analysis.

Finance & admin: invoice descriptions, cash flow narratives, expense categorization drafts, basic forecast scenarios (always verify numbers).

Your top 3 (rank by time saved × impact):

1. [[Function + specific task]]

1. [[Function + specific task]]

1. [[Function + specific task]]

Section 2: Tool Stack by Budget

Free / very low cost: ChatGPT or Claude free tier, Canva Free, Google Workspace or Microsoft 365 free tiers, native platform AI (Meta, Google, email tools).

$20-100/mo: Paid LLM + Canva Pro + email or CRM with AI + simple automation (Make or Zapier starter).

Higher: Full small business platform (Shopify + AI apps, QuickBooks + AI features, dedicated marketing platform).

Rule: Pick tools your team will actually open every day. Prompting skill is portable.

Numbered evaluation for any new tool:

1. Can it do the top use case in under 10 minutes on day 1?

1. Does pricing scale with our actual volume?

1. Is there a clear export path for our data?

1. Does the privacy policy match what we promise customers?

1. Can one non-technical person run the core workflow after 30 minutes?

Section 3: Ready Prompts and Workflows

Customer reply draft:

"Reply to this customer message [[paste]] for [[business type]]. Be helpful, concise, warm. Offer next step. Use our voice: [[3 traits]]. If you need more info, ask one clear question."

Social content batch:

"From this [[event or offer or update]], create 5 social posts for [[platforms]]. Mix formats (question, tip, story, offer). Include image description for each. Voice: [[traits]]."

Process or SOP from notes:

"Turn these notes [[paste]] into a clear 1-page SOP with: goal, who does what, step-by-step numbered actions, what good looks like, and common pitfalls."

Basic financial narrative (verify everything):

"Using these numbers [[paste summary]], write a short plain-language explanation for the owner: what changed, why, and one suggested action. Flag any assumptions."

Invoice or proposal line item helper:

"Turn this scope of work [[paste]] into 4-6 clear line items with short descriptions suitable for an invoice. Use plain language a non-expert client will understand."

Section 4: Guardrails for Small Business

• Never let AI make final decisions on money, legal commitments, or customer promises without human approval.

• Verify prices, dates, policies, and inventory before sending or publishing.

• For customer data, use only what is necessary and allowed by your privacy policy.

• Keep a simple log of AI-assisted customer communications for the first 90 days.

Numbered daily guardrail habits:

1. Human reviews every customer-facing AI draft before send.

1. Spot-check 1 in 5 factual claims in AI content.

1. Never paste full customer lists or payment details.

1. Keep receipts and notes for any AI-assisted financial summaries.

1. Pause automation immediately if error rate rises.

Section 5: 30/60/90 Adoption Plan

Days 1-30: One use case, full human review on every output. Measure time and quality.

Days 31-60: Add 1-2 more. Create 4-6 prompt templates. Train the whole team or key person. Add light automation for one flow.

Days 61-90: Review results. Drop what didn't work. Scale the winners. Document in a shared "AI at [Business]" page so new people ramp fast.

Success: Clear hours back per week and maintained or improved customer experience.

Numbered end-of-90 review questions:

1. Which use case delivered the most measurable time back?

1. What is our actual monthly tool spend vs value created?

1. Any customer issues traced to AI output?

1. One new use case worth testing in the next quarter?

Pro Tips and Pitfalls

• Batch similar tasks for prompt efficiency.

• Feed your own past good outputs back as examples.

• Common pitfall: shipping un-reviewed AI text that sounds off-brand or contains wrong details.

• Another: chasing too many tools instead of getting really good with one LLM + your daily apps.

• Third: forgetting to verify inventory, pricing, or legal language on customer outputs.

Start small, prove the time savings on real work, then expand. AI works best for small businesses when it removes the repetitive writing and research so the owner and team can focus on customers, product, and the work only humans can do. Use it as a smart assistant, not a replacement for judgment.

Additional Guidance and Checklists

Numbered pre-use checklist (run every time):

1. Confirm the goal and constraints are still accurate.

1. Update any [[Token]] values with current data.

1. Run the core prompt or workflow.

1. Human review for accuracy, tone, and compliance.

1. Log time and outcome for later tuning.

Common failure modes and fixes:

• Output too generic: add 2-3 specific examples from your real work to the prompt.

• Wrong length or format: specify exact word count, bullet vs paragraph, or section headings in the prompt.

• Outdated info: add "use only information from the attached sources or current date" and paste fresh context.

• Tone drift: paste 2-3 examples of the exact tone you want.

Pro Tips (continued):

• Version your best prompts with dates and results.

• Keep a "bad outputs" note so you stop repeating the same mistakes.

• When in doubt, simplify the prompt and add concrete examples rather than more instructions.

Final reminder: AI accelerates. You decide, edit, and own the result. Use the numbered steps and checklists above on every significant piece of work until the pattern is automatic.