Your front desk, answered.
After hours first.
Calls, texts and web chat, answered from rules you can read. Start with the calls you miss today.
There are two phone paths. The rules-based line answers from your approved rules in a standard phone voice and makes no model call. The Concierge lines we set up for you run on an ElevenLabs natural-voice agent, which does use an AI model, bounded by your business facts and our conduct rules. There is no public demo line today, and outbound calling is not sold yet. Text and web chat make no model call.
One thread, three rules.
A customer texts at 8:42pm. Pick a rule to see which reply it made.
Nothing leaves without you.
A draft is written, a person on your team decides, and four gates stand before anything is sent.
The same path for every outbound action, whatever the workflow. Pick a scenario, or any box.
Scroll the diagram sideways to see the whole path.
The line item it takes off your desk.
The answering service
Knows your hours, prices and booking rules. Says the same thing at 2am as at 2pm.
The unanswered phone
Calls, texts and web chat land in one thread per customer.
The risky chatbot
Text, web chat and the rules-based phone line make no model call, so they never invent a policy. A natural-voice line is set up to answer only from the facts you confirmed and to take a message for anything else.
It has to earn the switch.
Every channel starts switched off. A channel that fails a backtest of your own history cannot go live. The six controls →
Rules you can read
When a customer says this, do that. Every run leaves a trace.
A safety battery before go-live
Seven standards, four critical. A critical failure blocks activation.
Conversations are scored
Scores appear in your monthly quality report. A confirmed failure becomes a permanent probe.
What you actually see.
No card is taken, and every decision stays yours.
Read the rules it comes with.
Type real questions. Watch the trace.
Nothing reaches a customer while a critical standard fails.
Replies queue for your approval.
Staff-only first, then public.
It answers only what it was taught
It answers what it was taught. A question nobody configured gets a contact captured, not an invented answer.
A request for a human is never ignored
A caller who asks for a human is transferred to your workspace transfer number first, then to the box-wide number; if neither is set, a message is taken. It never hangs up on them. Clinical words, complaints and requests for regulated advice are marked urgent.
It says it is AI on the first turn
It says it is AI and names your business on the first turn of a call or text. The line is always on for every business and cannot be switched off.
Texts go only to people with consent on record
Texts go out only to people with a consent record on that channel. STOP on any channel stops every channel, and a send with no consent is refused and logged.
Phone numbers are set up with our help
Phone numbers are set up with our help today, not bought self-serve from inside the product. That is on the roadmap. The steps are on how your phone gets answered.
Start with after-hours, not your receptionist
Start with after-hours and overflow. We do not recommend replacing a receptionist on day one, and we will say so.
Voice minutes: 500 included, then USD 0.29
Concierge, the phone line, is the Team plan with the phone workflow. It includes 500 voice minutes a month. After that, voice minutes are USD 0.29 a minute (CAD 0.39, EUR 0.25), added to your monthly invoice by hand for now. Business includes 1,000 voice minutes a month, then USD 0.29 a minute (CAD 0.39, EUR 0.25). Included minutes reset each month and do not roll over. Alerts at 50%, 80% and 100% of your included minutes, and at a spending cap you set, are built, and a Usage page shows minutes used with a daily chart. When the minutes and cap run out, new calls forward to you or take a message, your choice; a call in progress is never cut off. Minutes are counted per call, rounded up to the next whole minute, and only for calls the AI handles. Text and web chat cost nothing per conversation.
No customer counts, no invented savings
No customer counts, no invented savings figures. The proof is the trace and the audit, on your own conversations.
Flat. Unlimited seats. The price is the price.
Phone line: 500 voice minutes a month included on Team, then USD 0.29 a minute (CAD 0.39, EUR 0.25); 1,000 on Business, then USD 0.29 (CAD 0.39, EUR 0.25). Minutes do not roll over. Every tier, in your currency →
Free
One workflow, draft-only, no outbound sends.
Team
Live channels, approver roles, consent ledger, data export, verification report on request.
Prefer that we run it for you?
We set up your phone line and look after it. USD 449 a month, setup included, so there is no separate setup fee. It includes 500 voice minutes a month, then USD 0.29 a minute, added to your invoice by hand for now. You also get a monthly report on your calls and enquiries, in your portal. Pay yearly and get 2 months free: USD 4,490 a year, arranged with us. If you are not happy, tell us within your first month and we refund that month's fee in full (one refund per client, and not on top of a pilot). Trades businesses can try it first with a 14-day pilot on their forwarded calls, up to 200 voice minutes, with no card and nothing charged at the end. Want your follow-ups, Google listing and books run too? See We run your office. The do-it-yourself Concierge stays at USD 149 a month, with optional setup at USD 299.
Try it on your own questions.
Nothing sends, pays or files until your team says so.