About
The route from recommendation to cart is a graph. Somebody should map it.
“In AI commerce, the money doesn't go to whoever the AI recommends. It goes to whoever controls the route from the answer to the checkout.”
Aqxle started as an analytics company. Then we ran one experiment — 168 beauty shopping questions in ChatGPT, five times, every link followed to where the shopper lands — and the answer was too important to keep behind a product. AI is already selling. The credit, the signal, and often the sale itself leak along the way, and almost nobody in the chain can see it.
So Aqxle started writing, publicly. The AI Commerce Leakage Index is its first public edition: a standing measurement of what happens between the moment an AI names a product and the moment someone gets paid for it. Weekly essays explore the same territory; monthly deep dives measure it.
The method is simple to state and hard to fake. No panels or revenue estimates. Every figure is a per-run average with its range, from links followed by script to their final landing pages.
The founder
Bharad Ramesh
Founder of Aqxle, ex-WPP Media head of research & investment analytics. Bharad has spent his career at the intersection of media, commerce, and measurement — and now studies what happens when the recommender is a machine and the receipt belongs to someone else. He takes on a small number of advisory engagements.
Member, ARF Analytics Council · 4As expert network · CIMM start-up program
Press
Charts and data from the AI Commerce Leakage Index are available on request, with credit line: "Source: Aqxle, AI Commerce Leakage Index, Beauty, Edition One (September 2026)."