Your comments, answered in your voice

justask reads the comments and messages arriving on your accounts, drafts three ways to answer each one, and sends the one you pick. Direct messages it answers on its own.

Request early access See how it answers

Reads and replies. It never posts to your feed.

The justask inbox: a complaint, and three ways to answer it

The inbox. A real comment, and the three handlings it was given.

Comments

Three ways to answer, and they differ in what they do

A comment is read by everyone who reads the post, so answering one is a decision, not a typing task. justask gives you three handlings and marks the one it would choose. Most tools give you three rewordings of the same reply, which is one option printed three times.

Take it to DM Move detailed handling to a private conversation

We have seen your issue and will message you directly to sort this out.

Ask one question Need one specific fact to proceed

Could you share your order number so we can locate the shipment?

Acknowledge and close Brief empathy, no further action publicly

Sorry to hear the candle arrived cracked and did not meet expectations.

Direct messages

A message answers itself

Nobody is waiting at 2am to pick between three drafts. A direct message is already private, so justask writes the reply, checks it against your brand rules, and sends it. You see what went out and can follow up in the same thread.

Every automatic reply says it was sent automatically. That is on by default, because Meta requires it and because a customer who finds out later feels tricked.

When it cannot answer

It says so, out loud

A refund, a double charge, anything touching money it cannot verify: the agent does not guess and it does not go quiet. It tells the customer a colleague is picking this up, then puts the thread in a queue with a button that hands it to you.

The alternative, which most tools ship, is a flag on an internal dashboard while the customer sits in silence.

Setup

Three steps, then it is watching

The agent learns your voice from the site you already have and the replies you have already written. There is no tone-of-voice questionnaire.

01 Brand system

Point it at your site. It reads what you sell, what you promise, and how you write, and turns that into rules an agent can be held to.

02 Connection

Authorise Instagram in one window. You are already signed in, so it is a single approval and nothing to copy across.

03 Inbox

Comments and messages arrive, already read and already drafted. You choose, or you let the direct messages run themselves.

Limits

What it will not do

A tool that answers customers in your name is only worth having if the boundaries are real. These are enforced in code, not promised in a policy.

It will not post to your feedNo scheduling, no content generation, no posting. It answers what arrives. That is the whole job.
It will not invent a policy you do not haveNo refund, delivery date, discount or price is ever stated unless your brand rules say it in those words. A draft that quotes one is blocked from sending on its own.
It will not take instructions from a strangerA comment that tries to talk to the agent is treated as data, never as a command, and never gets an automatic reply.
It will not answer quietly when it should not answerEvery automatic reply is disclosed, and every thread it hands over is one a person is told about.

Connect one account and read the first three drafts

It reads your site, watches one account, and shows you what it would have said. You send nothing until you want to.

Request early access