I'm not quite ready to +1 the idea though. It's intriguing, but seems to open a user up to abuse. Unsolicited messages (email, SMS, IM, DM, Facebook, etc.) are a popular way of phishing (tricking individuals into giving away private information). One tip security pros repeat over and over is not to click on unsolicited links.
While you say customers will only provide payment information to a company they trust, what is to keep a crook from impersonating a trusted vendor and tricking a customer into paying them instead of the actual company?
I'm all for convenience when done safely, and perhaps your business model accounts for this - I'm just curious what you have done to prevent SMS-based payments from turning into a source of fraud.
If a crook spoofed a phone number (pretending to be a company), the customer would still need to initiate the order. Meaning, if they received a text out of the blue from a company saying "thanks for ordering pizza now put your credit card in" without having actually asked for pizza, it would be quite weird for them to put their CC info in.
If a user's phone was lost/stolen, there's a chance someone could text in an order and at that point it is up to the company to do things like verifying information for orders over a certain price point (along with other security hurdles).
As for unsolicited messages, it is illegal for a company to do that (it still happens, I get spam phone calls on a daily basis) and we give an easy way to opt-out (simply reply back any of our unsubscribe keywords like "stop") and they wont' be able to message you anymore through Sonar.
Hope that helps! Would love to hear your thoughts.
