ShiFt Answer Bank
What questions should I ask an AI receptionist vendor?
The written questions to put to any AI intake vendor before you go live.
Ask seven questions and get the answers in writing: Does the caller hear it is AI in the first sentence? What does it do when someone asks for legal advice? Can it state a fee or promise the firm will take the case? What happens to a declined prospect’s information? Who runs the conflict check? Where are recordings stored, who reads them, and are they used to train models? Did a lawyer approve the script, and can you prove it? A vendor that answers all seven on paper is one you can defend.
- •The answers matter less than whether they are in writing — insist on written responses.
- •A “Don’t know” from your own team is a question to resolve this week.
- •The same seven questions work on any vendor, so you can compare like for like.
The seven questions that matter
Marketing pages describe features; they rarely tell you what the AI actually does on an intake call. Seven questions cut through that. Disclosure: does the caller hear it is AI in the first sentence of every call, text, and chat? Advice: when a caller asks “can I sue,” does it redirect to an attorney, scripted and logged? Promises: can it quote a fee, predict a result, or say the firm will take the case? Prospective clients: what happens to a declined caller’s information under Rule 1.18? Conflicts: who decides — a person at the firm, or the AI? Data: where are recordings kept, who can read them, and are they used to train models? Supervision: did a lawyer approve the script, and is there a sign-off log?
Why “in writing” is the whole point
A confident answer on a sales call is not evidence. If a client or a grievance committee later asks how your intake handled a duty, a screenshot of a feature list will not help you — a written answer from your vendor will. So the useful move is to send the questions and keep the replies. When a vendor cannot put an answer in writing, that gap is exactly what you needed to find, and it is far cheaper to find it now than after a declined prospect complains. This is also how you compare two vendors fairly: same seven questions, same written standard.
A letter you can send today
The Supervised Intake Standard is the seven-test framework behind these questions, and the Bar-Safe AI Check turns it into a two-minute score that ends with a copy-ready vendor letter. Run it against your current provider, send the letter, and file the answers. ShiFt publishes its own answers to all seven on the standard page, so you have a worked example to compare against.
Questions answered
Full answers
- What is the single most important question to ask?
- Whether the vendor will answer in writing. A confident verbal answer is not evidence. The seven questions cover disclosure, advice, promises, prospective clients, conflicts, data, and sign-off — but written responses are what you can actually defend later.
- Do these questions work on a human answering service too?
- Yes. The Supervised Intake Standard is written so you can score any vendor — an AI tool, a human answering service, or your own staff — against the same seven tests and compare the answers directly.
Explore ShiFt
Related questions
AI answering and lead response
What is an AI answering service?
An automated system that answers inbound calls, captures intent, qualifies the caller, and books or routes the next step.
Read answer →
What is AI call answering for service businesses?
How AI picks up calls when you are on site, after hours, or busy, and qualifies the caller before a person needs to get involved.
Read answer →
What is after-hours lead response?
The system that captures and qualifies buyers who contact your business at night, on weekends, or when staff are unavailable.
Read answer →
What is missed-call recovery?
How to respond to callers who did not reach a person, restart the conversation, and move qualified buyers toward booking.
Read answer →
Attribution and revenue tracking
What is conversation intelligence?
Analysing what happened inside a call, intent, urgency, qualification outcome, objections, and connecting it to attribution.
Read answer →
What is closed-loop ads-to-revenue tracking?
Connecting every marketing source all the way through to revenue, not just clicks or leads, so every channel is measured by the money it made.
Read answer →
What is revenue-centric attribution?
An attribution model that replaces vanity metrics with source-to-revenue reporting, showing which campaigns produced customers and money.
Read answer →
Qualification and speed to lead
What is AI-powered lead qualification?
Using AI to score, route, and qualify leads the moment they arrive, separating high-intent buyers from low-fit inquiries before a human gets involved.
Read answer →
What is speed to lead?
The time between a buyer's first enquiry and your first response, and why it is one of the strongest predictors of whether a lead converts.
Read answer →
What is an AI lead response platform?
A system that captures inquiries, responds instantly, qualifies buyers, books next steps, follows up, and attributes every outcome to source.
Read answer →
Find your biggest revenue leak
The free leak check takes 2 minutes and shows you exactly where you're losing money.