nexy

Use case

A receptionist that answers your phone

It answers when you cannot. What you have switched off is not in its head at all — not known-but-hidden. It says plainly what it cannot do, never pretends to be a person, and takes a message you can act on.

It picks up when you cannot, answers only the things you have switched on, and takes a message for everything else. What is switched off is not in its context at all, rather than known-but-withheld — and it says out loud what it is and what it cannot do, so nobody is misled into thinking they reached you.

What it looks like

It says what it is in the first sentence. What it may say is a set of switches, not a hope.

How to do it

  1. Say in plain words what it may tell people: your hours, whether you cover an area, that there is a call-out charge, when you are back.
  2. Say what it must never do: no prices quoted, no appointments agreed, nothing about a particular customer.
  3. Say who things get routed to. “Damage always goes to Joep” is an ordinary sentence, not a form.
  4. It answers, says what it is, and handles what it can. Everything else becomes a message with the details you need.
  5. You get the message with the useful parts separated out: what, how urgent, where, who, when they called.

Why it works

Off means absent, not hiddenA switched-off subject is not in its context at all. There is no persuading it into saying something it was never given.
It never pretends to be a personPartly honesty, partly practical: somebody who knows they are leaving a message leaves a better one.
Saying ‘I cannot’ is usefulA caller who is told plainly that it cannot commit to a date stops pushing for one and gives you their number instead.
A message with the right fieldsWhat, how urgent, where, who, when. Not a recorded ramble you have to listen to twice.
Reachable without being interruptedThe point is not screening people out. It is that the phone can ring while you are under a sink.

Where it stops

An incoming channel is where the limits matter most, because the person on the other end did not choose this.

No prices and no promisesIt does not quote and it does not agree a time, unless you have explicitly switched those on as fixed facts.
It will not claim to be humanNot even if asked directly. There is no setting for this.
It knows nothing about your customersIt cannot look up an order, a balance or a case unless you have deliberately given it that.
It will not keep somebody talkingIf they want a human, it says when you are reachable and takes the message.

Questions people ask

What does an assistant that answers my phone actually say to callers?

Only the things you have switched on, in plain language: your hours, whether you cover an area, that there is a call-out charge, when you are back. Everything else becomes a message. The first thing it says is what it is — that it answers the phone for you and is not a person — so nobody is under the impression they reached you.

How is this different from an answering machine or voicemail?

Voicemail takes a monologue and gives the caller nothing. This answers the questions it is allowed to answer, which is most of what people ring about, and for everything else it asks the specific things you need — what, where, how urgent, which number — so the message that reaches you is actionable rather than a two-minute recording you have to listen to twice.

What stops it from saying something it should not?

The fact that the thing is not there. A subject you have switched off is not in its context at all — it is absent rather than known-and-withheld — so there is nothing to coax out of it. This is a different arrangement from a system that knows your prices but has been told not to mention them, and it is the reason it can be trusted with a phone line at all.

Will it pretend to be a human if someone asks?

No, and there is no setting that allows it. If a caller asks directly, it says what it is. Beyond honesty there is a practical reason: somebody who knows they are leaving a message leaves a usable one, with their number and the actual problem, instead of trying to negotiate with something that cannot agree to anything.

Can it book appointments or quote a price?

Not unless you have given it a fixed fact to repeat, like a standard call-out charge. It will not negotiate, it will not agree a date, and it will not commit you to a job. What it does instead is tell the caller plainly that it cannot do that, which stops them pushing and gets you their details.

Does it know anything about my customers or my orders?

Nothing, unless you have deliberately given it something. It cannot look up an order, a balance, a case or a history, and it will say so. Most people start with the opposite of a knowledge base: a handful of facts they are happy for a stranger to be told, and a clear instruction that everything else becomes a message.

Can I use this for a business with several people rather than just me?

Yes, and that is where the routing matters. You say in ordinary words who things go to — damage always to Joep, invoices to the office, anything urgent to whoever is on call — and each caller's thread goes to that person. It stays one phone number from the outside, with a separate thread per person on the inside.

Will callers be annoyed at not reaching a person?

Less than they are annoyed by voicemail, in practice, because they get an answer to their actual question most of the time and a clear statement of when a human is reachable the rest of the time. What makes people angry is uncertainty — not knowing whether the message was heard. Being told plainly what happened next removes that.

Related use cases

New here? How Nexy works explains the whole thing in a minute, and the index lists every use case.

Stay posted on Nexy

We add a new use case most days and test what actually works. Get the short version by email, or a notification the moment something ships. No account, and one click to stop.

A receptionist that answers your phone and only says what you allow