Your team, thoughtfully built.

How to Write a Customer FAQ With AI

Practical guides from the team building Staffmor.

To write a customer FAQ with AI, give it your approved policies and the questions customers really ask, and require a source for every answer. Where your policy doesn't cover a question, the right output is a question for you. An FAQ that fills a gap with a plausible answer has just written your policy for you.

The example is a made-up bicycle repair workshop.

What to give it

  • Your policy as it stands: bring a bicycle, tools are supplied, the session lasts two hours and teaches flat repair rather than a full bicycle repair service.
  • The questions people ask: are beginners welcome, where do I park, will you fix every problem?
  • Which documents are current. An old refund answer copied from last year's page is the classic FAQ error.

Take customer names out of the questions. The writer needs the question, not who asked.

The request

The request: "Write FAQ entries from the policy and the customer questions below. Use only what the policy says. Put a source label on each answer. List any question the policy doesn't answer separately, for me to decide. Do not invent a parking, refund or accessibility policy. Return a draft only."

What comes back

  • What should I bring? Your bicycle. We supply the tools. Source: policy, equipment.
  • How long is the session? Two hours. Source: policy, schedule.
  • Will you fix every problem with my bike? The workshop teaches flat repair. It is not a full repair service. Source: policy, scope.

And a second list, for you:

  • Are beginners welcome? Not stated in the policy.
  • Where can I park? Waiting on the venue.

The mistake to watch for

The second list is the valuable half. Parking, refunds and accessibility are the kind of questions a policy tends to skip, and they are where a confident made-up answer does real damage. If the draft says "free parking is available on site" and you never said so, that sentence is now a promise on your website.

Group questions that are truly the same. "What should I bring?" and "Are tools supplied?" can share an answer. "Can I come without a bicycle?" cannot, if the answer is different.

Before it goes live

Have whoever owns each policy read its answers, especially anything about money, access or who qualifies. Then read the finished page as a customer and try the contact link. When a gap gets filled, update that one answer and note where the new fact came from. Add to the FAQ you have. A second FAQ page for the same questions just gives customers two answers to compare.

Run it in your Staffmor office

Open the office on the computer running it with its Start Staffmor launcher. If you already have a paired web office, open it here. Paste this request and the source material into the conversation. Use your approved Codex connection, or ask your open Claude Code session to "Check Staffmor for work." A saved request is not a completed result. Review the answer or file when it comes back and reply with corrections. For a long policy, ask for the answered list and the gap list as two separate results.

Staffmor drafts from what you approve. Deciding policy, and publishing the page, stays with you. New to Staffmor? Set up your office with Codex or Claude Code and an empty folder. Use your own supported AI account and approve the download, office activation and local folder/network access during setup. Claude Code pickup is manual; keep that local session open.

Related: How to set up an AI agent team · How to write instructions AI agents actually follow