AI agent for closing sales

In Meteor it is the sales closing Met. It picks up the prospect who already wrote in or showed interest, answers the usual objections (price, timing, comparison) with your arguments and your product and price Collections, asks for the close and sends the Wompi, Mercado Pago or WooCommerce order payment link. It only treats the sale as paid when the gateway approves it. It is not a cold-calling dialer and it does not blast lists: after 24 hours, on WhatsApp it only picks up again with a template approved by Meta, and messages to a whole segment go through Broadcasts.

The cycle has three stretches: qualifying, negotiating and collecting. In Meteor two Mets cover them on the same CRM. The lead qualification Met asks your questions and gives each prospect a score from 0 to 100 with the reason written out; the sales closing Met answers objections, asks for the close and sends the payment link, and treats the sale as paid when the gateway confirms it. Where it stops matters just as much: it does not handle contract signatures, and anything outside your rules goes to a person with the full history.

It generates the payment link on the gateway you have connected: Wompi (PSE, Nequi, Bancolombia or card, in Colombian pesos), Mercado Pago, or the order’s payment link in WooCommerce when your store has an online gateway. The customer pays on the gateway’s page and the Met checks the status: it only treats the sale as paid when the gateway returns it approved. No card data goes through the Met.

No. The Met closes with the payment, not with an electronic signature. If your sale needs a signed contract, the Met leaves the opportunity ready in the CRM and hands it to your team for that step.

It follows up until there is an answer or a close. Within 24 hours of the person’s last message, the Met can write to them on any channel. After that, on WhatsApp it can only pick the conversation back up with a template approved by Meta, and on Instagram and Messenger it has to wait for the person to write again.

It escalates to a person on your team: it leaves a note with the reason, assigns the chat and steps out of the conversation. The rep picks it up with the full history and the Met does not talk over them.

Every implementation includes discovery, setup and tuning during the ramp-up. After that, each Met’s quality is measured by sampling: an automated judge reviews a sample of its conversations and scores the support it gave, and the activity tab shows what each Met did, with which tool and at what cost. If something is not working the way you expected, we review it with you using that data and adjust until it is running.