For freelancers & solo operators
Typical goal: lift personal throughput without adding headcount.
Typical scope: AI-assisted writing, research, and client communications; template libraries for repeat deliverables; a lightweight inbox and scheduling assistant.
Representative tool stack: ChatGPT / Claude, a structured prompt library, n8n or Make for automations, Notion or Obsidian for reference, email-client integrations.
Typical deliverable: a documented personal AI workflow with runbook, reviewed after 30 days of live use.
For micro-businesses (1–10 people)
Typical goal: reduce admin overhead and accelerate customer replies so the owner can spend more time on revenue work.
Typical scope: shared prompt and template library, AI-drafted customer responses with human review, automated reporting for weekly or monthly metrics, structured intake forms that pre-fill your tools.
Representative tool stack: Claude or ChatGPT Team, Airtable or Google Sheets, Zapier / n8n, a vector store for internal knowledge, email and CRM integrations.
Typical deliverable: two to three operated workflows, team training, and a 30-minute runbook walkthrough for non-technical staff.
For growing service teams (10–50 people)
Typical goal: stand up repeatable AI-assisted operations ahead of the next hiring round, so new staff inherit a working system rather than ad-hoc tooling.
Typical scope: knowledge-base retrieval for support and sales, structured quality-review pipelines, an internal agent for frequently asked internal questions, cost monitoring and guardrails.
Representative tool stack: hosted LLM APIs, a managed vector database, internal dashboards, role-based access controls, usage and cost telemetry.
Typical deliverable: a documented internal AI platform with access controls, operator runbooks, and an on-call escalation note for the owning team.