AI Pulse · Business Models & Winners

Cashfree has moved its Relay AI agent from beta to general availability for all merchants. The tool is aimed at the messy work that sits around getting paid, not just processing a transaction. It automates failed-payment retries, abandoned-cart recovery, cash-on-delivery confirmation and dispute filing.

For smaller businesses, that work often lands with finance or operations staff who are already stretched. Cashfree says Relay can reduce payment-operations time from roughly 60 hours a week to less than 45 minutes. That is the company’s claim, not an independently verified result, but it points to a clear use case. AI agents are starting to take on repeatable operational tasks where delays directly affect revenue and cash flow.

The more interesting business-model decision is the pricing. Relay is free at launch, with Cashfree planning outcome-based pricing after it has collected adoption data. In plain terms, it wants merchants to use the product first, then charge when it can better link the agent’s work to measurable results. Expect to see more AI products follow this path, especially where the value can be tied to recovered revenue, saved staff time or fewer manual errors.

For operators, the question is not whether an agent sounds impressive. It is whether it can reliably handle a narrow process, show what it did and prove the financial impact. This is the kind of thing we build into an AI command centre, connecting operational activity to the outcomes that matter.