recommendation engine
Predictry

Recommendation Engine Startup Predictry Lands $230,000 Seed Funding

Next Story

ThriveOn Wants To Make Mental Health Care Affordable And Accessible

Major e-commerce sites like Amazon depend on powerful proprietary recommendation engines to drive user engagement and sales. Smaller businesses, however, lack that advantage. Predictry wants to give all online stores the same kind of tools. The Malaysia-based startup has raised $230,000 in seed funding from the Malaysian Development Corporation’s (MDEC) Big Data Product and Commercialization Fund.

Though Predictry’s competitors include established players like Strands Recommender, founder S.T. Chua believes there are still plenty of opportunities for his startup as the e-commerce market grows, especially in Southeast Asia. The Crescent, an Islamic lifestyle store run by Celcom, currently uses Predictry to power its recommendations. It is also conducting trial runs with several of the region’s largest e-commerce startups, but can’t disclose the companies.

Chua says that Predictry is able to provide relevant recommendations to shoppers because the company tweaks its algorithm for each site. Its target market is small- to medium-sized businesses, as well as fast-growing e-commerce sites that don’t have the resources to build their own recommendation engines.

“When you see all the other apps out there, for Shopify or Magento, those plugins are actually quite dumb and not sophisticated at all because you have to go through categories or tag products yourself,” says Chua. “For every e-commerce store, it depends on how descriptive you are or how many tags or how you hot link it.”

To provide relevant product recommendations, Predictry’s engine tracks which pages shoppers go to, maps out viewing and purchasing behavior, and, if shoppers opt-in, will also consider information from their connected social networks.

“On The Crescent, if you are looking for a headscarf, you might also want a brooch to wear with it, but we can’t show brooches all the time. That’s too much of a hard sell. We have to give some weight to certain things, but we can tweak algorithms to maximize clickthrough,” says Chua.

Predictry will target a wide range of e-commerce sites, ranging from food delivery services to travel sites. Chua says the startup will use its seed funding for marketing and to hire more developers.

Photo by PicJumbo