Castore Taps Ocula for AI Product Content

Castore uses AI agents to optimise product content, improve discoverability, and reduce errors across its digital commerce platforms.

Castore has chosen Ocula Technologies as its AI product content partner. The performance sportswear brand, which has partnerships spanning Oracle Red Bull Racing, Rangers Football Club, and some of the world’s biggest sports teams, will use Ocula’s AI agents to optimise product copy, metadata, and content across its digital catalogue.

“For a brand built on the principle that Better Never Stops, the bar for product content is high. Every SKU needs to be discoverable, compelling and consistent across search, marketplaces and AI shopping engines. That’s exactly what our agents are built for. More to come,” said Thomas McKenna, CEO and Co-Founder at Ocula Technologies.

Castore operates 35 Shopify stores globally and was managing pricing updates through overnight manual CSV uploads. During peak trading, updates took around 5 hours to complete, error rates ranged from 30 to 35%, and teams were routinely working late into the night just to keep systems stable.

ALSO READ: Why Events are a Missed CX Opportunity in B2B

After deploying Patchworks’ iPaaS, it has:

  •     Reduced pricing execution time from five hours to around two minutes
  •     Cut pricing error rates by over 90%, now sitting at 1 to 2%
  •     Achieved zero error pricing days during peak trading
  •     Eliminated overnight manual pricing work entirely
  •     Stabilised API load across 35 Shopify storefronts without licence expansion

“We were in a place where a 30% to 35% error rate was considered acceptable simply because the process was so long and so manual,” said Andy Richley, Head of Tech at Castore. “When you are spending most of the night just trying to get pricing through, you stop asking whether the process itself makes sense.”

According to Jim Herbert, CEO at Patchworks, Castore’s results highlight a broader shift in how retailers are reassessing integration performance as a strategic capability rather than a background function.

“Like many fast-growing retailers, Castore had gradually pushed more execution workloads into its core systems of record,” Herbert said. “Pricing data was being validated, transported and executed through the same platforms, creating unnecessary coupling and exposing peak trading operations to risk. The issue was not governance, but execution.”

ALSO READ: Sports Fans Are Always On, but Engagement Strategies Are Struggling to Keep Up

Herbert added, “By separating pricing governance from pricing execution, Castore was able to leave its core systems doing what they do best, while moving high-volume data movement and orchestration into a retail-first iPaaS.”

The impact during peak trading was immediate and sustained. Pricing execution times were reduced from approximately five hours to two to three minutes, with worst-case execution times of 6.5 minutes during peak periods. Pricing error rates fell from 30 to 35% to just 1 to 2%, with zero-error days achieved during peak trading. 

Operationally, concurrency remained stable throughout peak periods, with no emergency licence expansion required. Formal approval workflows reduced pricing fraud risk, and customer service issues caused by mispriced promotions were significantly reduced. 

Manual overnight pricing work was eliminated entirely, allowing data and ecommerce teams to shift from execution to oversight. Trading calls evolved from reactive firefighting to confirmation of stability.

“What surprised me was just how straightforward it became,” Richley said. “Once the data was prepared properly and approved upstream, Patchworks moved it into Shopify incredibly fast. That was the moment we knew we had the right execution layer.”

ALSO READ: Agentic AI Is Moving CX From Pilots to Performance

- Advertisement -spot_img

Featured Articles