London-based technology company Azoma has announced an $11.2m Series A funding round to expand software designed to help consumer brands manage how their products appear in artificial intelligence shopping recommendations.
Azoma announced the financing on 21 October. Puma Growth Partners led the round, joined by dunnhumby, Twin Path, MaRS IAF, Rank Ventures and Ignite Ventures, alongside individual investors. The announcement date should be distinguished from the completion timing of any earlier investment commitments.
The company calls its activity agentic commerce optimisation. Its tools are intended to improve how product information is discovered, interpreted and recommended by automated shopping services, while helping brands understand where they appear in those systems.
The funding is expected to support platform development, wider coverage of AI shopping assistants, recruitment and attribution capabilities that use retail data from dunnhumby. The presence of a data partner does not mean that every customer has access to identical datasets or that sales attribution has already been independently verified.
AI-mediated product discovery differs from conventional website search because a customer may receive a composed recommendation rather than a ranked set of links. An assistant can draw on product descriptions, marketplaces and third-party material, sometimes presenting a comparison without directing the user immediately to a brand-controlled website.
This creates a commercial challenge for marketers. Product specifications and availability must be accurate across channels, but visibility in an AI response may vary according to the question, location, assistant and underlying information sources. Brands cannot necessarily secure consistent rankings by making a single change to their website.
Azoma aims to help organisations manage that information and measure their performance. Yet a product appearing frequently in generated suggestions is not equivalent to a recorded sale. Companies must also consider price, distribution, stock availability, competing promotions and whether customers complete the transaction elsewhere.
Attribution becomes particularly difficult when a recommendation happens in one service and a purchase occurs through a retailer or marketplace. Connecting these events may require access to data that neither the brand nor the optimisation supplier controls completely. Different methods can therefore produce different estimates of commercial influence.
Brands already invest in search, retail media, product content and marketplace management. AI shopping adds another potential channel, but its importance remains uncertain relative to established methods. Investment decisions should be based on testable commercial outcomes rather than assumptions that all purchasing will shift to assistants.
Retailers and platforms may also change their recommendation systems, limiting what an external optimisation company can control. Providers must adapt technology and measurement methods as the market develops, while being clear about which aspects of performance are observed and which are modelled.
Azoma’s financing reflects investor interest in software specifically designed for commerce rather than general AI content production. The company will face competition from marketing technology groups and other startups seeking to provide product visibility, structured information and performance analytics.
New capital can help expand software and staffing, but the size of the round alone does not establish profitability or market leadership. The commercial test will be whether customers can demonstrate better product discovery and, ultimately, incremental business results after accounting for the cost of the service.
The financing comes as shopping assistants increasingly combine conversational interfaces with links to merchants, but those experiences are not yet standardised. Brands need to understand which sources an assistant relies on and how current its information may be. A product may be described accurately on the manufacturer’s website yet appear with an older specification elsewhere. Improving data consistency can be worthwhile independently of claims about influencing the underlying models, whose ranking and recommendation systems remain controlled by their operators.
The announcement gives Azoma resources to pursue that emerging opportunity. The maturity of AI shopping and the reliability of sales attribution will determine how much of the predicted market becomes a durable source of revenue.
You must be logged in to post a comment.