California-based Instacart has acquired Israeli computer vision company Arpalus, which has developed shelf intelligence technology purpose-built for grocery retail.

The accuracy of online orders is only as good as the underlying inventory data; undetected out-of-stocks and catalogue gaps are among the most persistent sources of customer dissatisfaction in online grocery, driving substitutions, cancellations and eroded trust. The acquisition should enhance Instacart’s ability to serve consumers, shoppers, retailers and brands with a more complete picture of on-shelf inventory and product availability.

Arpalus’ technology turns a quick video scan of a store shelf into an accurate, real-time picture of what is actually there. Using computer vision models built for the complexity of real grocery store environments, including low or unreliable wifi, inconsistent lighting, and thousands of visually similar products packed tightly together, Arpalus can identify individual items on shelves with more than 95% accuracy, on average.

The technology runs on any smartphone or camera-equipped device, which means Instacart’s network of 600,000 shoppers can generate real-time shelf intelligence at scale using the app they already use on every order. Arpalus models can also guide shoppers in the moment, prompting them to adjust camera angle or distance to ensure every item on the shelf is captured accurately and efficiently.

Instacart has spent more than a decade building comprehensive grocery-specific intelligence built on more than 1.6 billion lifetime orders, real-time inventory views from nearly 100,000 stores across North America, and proprietary machine-learning models purpose-built for grocery. At the centre of that system is a community of 600,000 Instacart shoppers who visit large-format stores generating over ten million unique daily data points.

Every order placed, item picked and substitution made feeds directly back into Instacart’s models, creating a continuously improving picture of real-world store conditions that takes years and significant scale to build. This results in an industry-leading found rate. With Arpalus’ shelf intelligence now part of this system, Instacart can drive that rate higher, moving closer to a future where every item a customer orders is seamlessly delivered to their door.

Arpalus’ computer vision technology extends to Caper Carts (company.instacart.com/enterprise-platform/connected-stores/caper-carts) equipped with external cameras as they move through store aisles, updating in-store on-shelf inventory insights in real time. Combined, Instacart and Arpalus are expected to drive more fulfilment efficiencies that create a better customer experience and new earning opportunities for shoppers, while improving inventory accuracy for retailers and brands. Data will be used in accordance with all applicable laws, including privacy laws, and contractual requirements.

“We believe the future of grocery retail is a unified experience powered by Instacart intelligence, where what happens in store connects seamlessly to ecommerce in real time,” said David McIntosh, chief connected stores officer at Instacart (www.instacart.com). “The Arpalus team has spent years building exceptional shelf intelligence technology, solving the problem of understanding what’s actually on store shelves, at any given moment. With our leadership in physical AI for grocery retail and by activating our network of shoppers, we can feed even more accurate shelf information back into our models, delivering better outcomes for customers, shoppers, and our retail and brand partners.”

Ofir Zilberberg, CEO of Arpalus, added: “Arpalus was founded to solve one of retail’s most fundamental challenges: bridging the gap between what happens in-store and the decisions driving business performance. We built an AI platform that gives retailers real-time visibility into inventory and store execution, enabling faster and smarter decisions at scale. Joining Instacart is a transformative milestone. By combining Arpalus’ AI innovation with Instacart’s scale and retail reach, we are accelerating the future of intelligent retail and redefining how stores operate.”

Richer shelf data strengthen Instacart’s platform across online and in-store environments. For consumers shopping on the Instacart Marketplace, or on their preferred retailer’s owned-and-operated web site powered by Storefront Pro (company.instacart.com/enterprise-platform/e-commerce/storefrontpro), more accurate inventory data mean a more reliable shopping experience. For retailers, it directly enhances Store View – Instacart’s real-time computer vision being piloted by retailers including McKeevers and Sprouts – improving shelf availability and accuracy at the store level.

Arpalus (www.arpalus.com) is an AI-powered retail intelligence company that helps retailers and CPG brands improve product availability, inventory accuracy and store execution through real-time computer vision. Founded in 2019, the company serves retailers and consumer brands, and its proprietary technology is protected by patents in the USA and Europe.