What mate announced
On November 10, the team behind the cashback app Checkmate introduced a shopping agent called mate. The announcement says the service is available in the US and can be reached through iMessage or WhatsApp. Customers are meant to search for products, receive offers, complete a purchase and ask about their package later.
The company says cashback or discounts are available at more than 200,000 US stores. It separately describes direct connections to thousands of stores. Those are different claims: the announcement does not establish a direct technical integration with every store in the larger number. It also does not establish how often the promised savings apply or how reliably orders work in practice.
From a message to a purchase
Imagine someone writes: I need comfortable running shoes by Friday and want to spend no more than $120. A useful assistant would need to bring together the size, availability, delivery date and total price. If an offer includes cashback, the customer should be able to see what is paid at checkout and when a later credit is due.
The announcement describes search by message, checkout and parcel tracking. It does not explain the approval steps for every purchase flow or how a return is handled. Those steps help determine whether a convenient entry point becomes a dependable purchase.
A contact on the phone
A message is a familiar place to ask a quick question. The same thread can remain useful when the customer wants to know where a delivery is. The new service shows how that idea can be combined with offers and a shopping agent built around savings.
A few days earlier, Amazon announced text access to Alexa for Shopping for an initially invited group in the US. Amazon says the conversation context will move between its own interfaces. mate describes a separate service built on a cashback network. The announcements share an entry point; they do not establish a connection between the services.
What changes for merchants
A merchant competes in that conversation on more than a product image. An assistant needs to find the right variant, understand the current price and interpret the delivery promise. Discounts and cashback may matter to a choice, and their conditions should be clear to the customer. The announcement does not explain how mate ranks offers or whether commercial relationships affect that order.
I would start with a frequently purchased product that has variants and follow its data from the customer’s point of view: is the size unambiguous, do availability and delivery costs match, and are discount and return terms understandable? Then test what a particular assistant actually reports. That exercise does not establish that mate supports a given store.
One interface, several steps
In When purchasing moves between interfaces, a traveller selects a suitcase in chat, compares it on another screen and later asks about delivery. The chapter explains why the selection and terms need to survive these handoffs. mate brings several tasks into one conversation; the announcement does not say whether it supports handoffs to other interfaces.
In Frontends That Breathe, I described a storefront that adapts to observed interests during a session. A messaging thread is a different kind of interface. The shared question is how a specific intent leads to a useful next action. The practical guide explains the different roles of a shopping agent, merchant agent and copilot.
Source and further reading
Read the original announcement ↗