Luxury activewear brand, Alo Yoga, has embedded AI into its fast-moving product and merchandising model to overhaul inventory planning, allocation and replenishment, delivering a +6.8% improvement in revenue weighted availability.

Speaking at NRF’s Big Show in New York, Alo Yoga’s VP Casey Hahn joined Gurhan Kok, Founder & CEO of invent.ai, to explain how the partnership is applying scientific rigour to what Hahn described as “the art” of Alo’s brand and retail execution.

Best known for its studio-to-street positioning, Alo operates a retail model defined by rapid product innovation. While the business is approaching its 20th anniversary, it has seen an explosion of growth in the last seven years, with stores now open in London and further bricks-and-mortar locations in Manchester and Leeds.

High frequency product drops

“As well as high growth, high velocity items, something that is special to us is constant drops of newness,” Hahn explained. With new products and colourways introduced every two weeks, as much as a third of its in-store and online assortments change within each cycle.

While this strategy has fuelled strong growth, Hanh shared how this has also created significant operational complexity, particularly as the brand grew to over 70 stores across 12 countries.

Alo found its existing planning systems were struggling to keep pace. “Predicting demand for [Alo’s] product globally is one of the most challenging use cases we have seen,” said Kok.

The business began working with invent.ai around two years ago, initially focusing on allocation and replenishment, areas where Alo was, in Hahn’s words, “hungry for results” and where improvements would be immediately visible.

A moment of truth for AI

Within six weeks, a proof of concept across 20% of Alo’s stores delivered a +10% sales uplift, driven by improved availability, while drawing from the same inventory pool as the remaining 80% of control stores. This delivered a “moment of truth” which established trust in AI-driven decisioning across the organisation.

Since then, the partnership has expanded to include assortment planning, SKU-level demand forecasting by store and trend, and buying optimisation. These initiatives have reduced waste and improved availability, contributing to a +2.9% year-on-year improvement in availability, alongside a three-week reduction in weeks’ cover.

Despite holding materially less stock, Alo has continue to improve customer outcomes. From 2024 to 2025, the business recorded -8.2% fewer abandoned sales, while revenue-weighted availability improved by +6.8%.

Reflecting on the impact, Hahn said: “I can say [the power of] art and science a million times. We have the art of products and trends down, but needed support with our assortment mix and understanding where the demand is coming from.”

“Some people think these results are think its too good to be true,” added Kok, while Hahn highlighted the in-store customer experience benefits, with the business now better able to predict demand and keep items in-stock – even after multiple peak trading days.

As product cycles shorten and customer expectations rise, Alo’s experience highlights how AI is increasingly becoming a core enabler of retail performance, explained Kok.

“With our AI-decisioning technology solution, we empower the planners to achieve real-time responsiveness to market trends, ensuring the right products are in the right stores at the right time.”