Global online eyewear company sees exceptional results with Algolia AI Search

Founded in 2003 by visionary entrepreneurs Tibor Laczay and Julia Zhen who thought being able to see shouldn’t be so expensive, Zenni Optical was one of the first online-only shops for prescription glasses. 

Its formula for success in providing affordable and accessible eyewear was going directly to consumers — and it was a formula that worked. From a small, homegrown business in Novato, California, Zenni has become a widely recognized name, employing nearly 2,000 people. 

With more than 1/5 of all glasses sold in the U.S. being purchased through Zenni, it has become the number-one destination for online prescription eyewear. And, through its global distribution, Zenni is responsible for selling more than 50 million eyewear frames around the world. 

Continuously innovating, the company is known for providing high quality at an affordable price point. In addition to its well-known online site where it sells quality glasses at a fraction of the cost, Zenni also has begun to work with eyecare providers to offer unbranded lower-cost, high-quality eyewear to their patients.

Obsessed with customer service

In its mission to help people “see better and live better,” Zenni recognizes the key to its growth is around customer service, and with that in mind it strives for a healthy amount of passion for the customer experience. 

“We definitely want to grow the company and serve our customers better,” says David Ting, CTO and Global General Manager at Zenni Optical. “So, using a well-used term, we are trying to be ‘customer obsessed.’” 

That demands continuous innovation around its products and price point, especially given high inflation, and constant work at improving every facet of the company. To operate quickly across a global distribution organization, it embraces collaborative technology. 

“We call it async work mode. A lot of discussions happen on Slack or within Google Docs. We try to include everyone’s opinions because they matter, but we also try to move fast.”

Improving search performance

The drive to achieve fast performance doesn’t end internally. To meet customer expectations, Zenni’s technology team knew they needed customer search results to come back to them quickly. 

The company was using two legacy search platforms, Endeca and Sitecore, and was looking for a solution that would operate as a secure source for patient data and could be cloud-hosted. And, of course, it needed to be fast. 

“There were two things we wanted to address,” Ting says. “We wanted to move from two search engines to one, and it had to be performant.” 

“We wanted the results pages to be served up very quickly. That’s one of Algolia’s strengths; infrastructure uptime and performance with Agolia are superb. And, since it’s a pure SaaS solution, we don’t have to deploy anything, removing management complexity.” 

“All of that made Algolia extremely attractive to us.” It took Zenni just two months to deploy Algolia Search to its custom-made e-commerce stack, which it uses everywhere on the site for both browse and search purposes — except for visual search. Algolia is used to power search for both Zenni’s website and mobile e-commerce experiences. 

Zenni worked with Algolia partner AAXIS Digital during the implementation who, Ting says, “did most of the heavy lifting.” He calls the migration project a three-way partnership between Zenni, AAXIS, and Algolia, noting that it's been a major success. 

Behind the scenes, Zenni’s merchandising team uses Algolia a great deal, Ting suggests, for SKU and collections optimization, or to boost products. He expects, however, that some merchandising activities will become eliminated or simplified as the company takes more advantage of neural AI technology from Algolia that it is currently testing.

A unique industry needing powerful AI search

To improve customer experience, Zenni needed a search solution that could handle the specifics of searching for eyewear. For example, eyeglass frames are extremely nuanced products: a customer might be looking for a turquoise frame and call it “spotted,” while another might have no idea what to call it. 

With about 5,000 SKUs (and more than 200 being added every year), discoverability is a challenge, requiring the ability to filter by color or different qualities and variations. But, it also adds complexity. 

From synonym handling to material filtering, Algolia has been adept at handling the needs of eyewear search, improving customer experience while upping search speed and relevance. 

To take things to the next level, however, Zenni has been beta-testing Algolia Neural Search. The new technology is built on a new type of AI, combining advanced natural language understanding (NLU), AI-driven vector search, adaptive learning, and query suggestions to provide even more nuanced search capabilities. 

“Because of the complexities of eyewear variations, getting synonyms right is extremely difficult,” Ting adds. For example, there are people who call clear frames — what we call translucent color — white. Having those searches added and relevance scored would be huge for us, because manually managing it across a large multilingual SKU is extremely challenging.” 

“That’s why we’re looking at Neural Search to simplify and calculate those synonyms in a relevant way that improves automatically with user input. We’re testing it, and very excited about not having to manually manage all that.”

Outstanding results: Seeing is believing

Since deploying Algolia, Zenni has seen great improvements in metrics related to the search experience on its site. 

In less than a year, its new search bar experience, taking advantage of autocomplete and query suggestions, has driven a 44% increase in search traffic and a whopping 34% increase in search revenue. 

Perhaps most impressive, the company has seen revenue per session increase by 27% since implementing Algolia. CVR (conversion rate) has similarly gone up 9% and Search AOV increased by 16% in that same period (February to September 2023) over the year before. 

Ting says the entire Algolia team helped make the journey easy, saying they are a true partner that provides white glove service. “We’re really excited to continue to work with Algolia and continue to improve the customer experience together.”

As for Algolia’s biggest benefit, Ting suggests it all comes back to speed and reliability. “Algolia is up and fast—and everywhere. That’s actually the killer feature for us right now.” 

“We’re working a lot on the personalization side and improving relevance rankings because of nuances around glasses, but performance is the top feature. If it works as advertised, then Neural Search will be the new killer feature beyond Algolia’s strength in reliability.”