Selection dimensions that matter
- Capability — does it actually solve your specific problem, with evidence from real deployments?
- Total cost — build fee, monthly operating, model passthrough, hidden labor.
- Security and compliance — SOC 2, BAA if needed, DPA, data residency.
- Portability — IP ownership, exit clauses, data export.
- Reliability — uptime, latency, error handling.
- Operating support — escalation, response time, who actually answers when it breaks.
- Vendor stability — funded? Profitable? How long have they been operating?
The 12-question SMB checklist
- 1. Show me a live deployment at a real client. Not a sandbox.
- 2. What does the contract say about IP ownership of my workflows?
- 3. What is the no-fault termination clause?
- 4. Who from this conversation will actually do the work?
- 5. What model and infrastructure costs do I pay on top of your fee?
- 6. What is your data processing addendum?
- 7. How fast do you respond when something breaks in production?
- 8. What is your eval coverage and current pass rate?
- 9. Are you SOC 2 Type 2 certified?
- 10. Will you sign a BAA if my workflow touches PHI?
- 11. What is the all-in 12-month cost?
- 12. What scope is outside your competence?
Red flags during selection
- Pricing opacity until NDA signed.
- Refusal to provide a customer reference call.
- Contract longer than 90 days for SMB scope.
- No SOC 2 or BAA when required.
- Vague answers on data handling, retention, and training.
- No documented evals or quality measurement.
- Sales team different from delivery team.
How to score vendors
Build a simple weighted scorecard before any vendor demos: list the dimensions above, weight each by your priorities (security may matter more than capability for regulated workloads; capability may matter more than cost for revenue-generating workflows), score each vendor 1-5. The scorecard prevents demo-induced bias and produces a defensible record if the choice is questioned later.
What it means for your business
Most bad AI engagements trace back to vendor selection, not vendor execution. The half-day spent on a scorecard saves the six-month wind-down you avoid.
Related terms
- AI Readiness — AI readiness is whether an organization can actually deploy AI safely and usefully. Definition, dimensions, and a practical SMB checklist.
- AI ROI — AI ROI is the measurable financial return from an AI deployment. Definition, calculation, and the common traps that fake the numbers.
- AI Implementation — AI implementation is the end-to-end process of deploying an AI workflow from scoping through production. Phases, timeline, and SMB common pitfalls.
- Build vs Buy AI — The build-vs-buy decision for AI depends on scope, talent, time horizon, and total cost. A practical decision framework for SMB owners.
- Enterprise AI vs SMB AI — Enterprise and SMB AI projects share technology but differ in budget, scope, timeline, and vendor type. Comparison framework for SMB buyers.