Join us for RedisConf and Hackathon, April 20-21

Register Now!

GoMechanic

CASE STUDY

GoMechanic needed a fast data layer to search across 10 million spare parts and power our customer service chats. RediSearch on Google Cloud was not only four times faster than our existing database but also extremely robust, easy to scale, and inexpensive.


— Prasenjit Singh, Vice President of Engineering, GoMechanic

Company: GoMechanic
Industry: Specialty Retail, Automotive

The Customer:

GoMechanic is India’s largest network of technology-enabled car service centers, offering a seamless experience for car service, repairs, cleaning, and spare parts and accessories across the sub-continent.

The Challenge:

GoMechanic needed a fast and accurate search for their spare parts catalog page that holds more than 10 million items. The team wanted an autocomplete feature to provide a better search experience for their customers. Additionally, they wanted their customer service reps to be able to search in real-time across multiple simultaneous chats.

The Solution:

GoMechanic adopted RediSearch to power their spare parts retail site and their customer service chats because of its amazing performance, flexibility, and simple developer experience.

The Benefits:

RediSearch increased performance of spare parts display, search, and SEO by more than four times. The performance improvement for the customer support chat search using RediSearch will allow their reps to deliver superior experiences to their customers and future-proof their infrastructure as the company scales.

“We need to load our spare parts and accessories retail sites and our customer service chats instantaneously, and RediSearch does that with ease. We tried many solutions, but Redis Enterprise on Google Cloud delivers the best performance for its cost.”
— Nagendra Kumar, Senior Manager Technology, GoMechanic

Expanding to new use cases

GoMechanic is India’s largest network of technology-enabled car service centers. Founded in 2016, the company started with automobile maintenance and repairs, but has since expanded into selling spare parts and accessories.

While preparing to launch the spare parts website and app—which lists more than 10 million items—GoMechanic found its existing database was unable to provide the performance and experience it needed. The engineering team was looking to replace it with a database that had native full-text search and autocomplete features so customers could search spare parts without knowing the exact product name, or its spelling.

Additionally, the customer support team wanted a robust search engine for its own use. Support agents handle up to 40 conversations at once, and customer-support conversation threads are linked to the customer’s phone number, explains Prasenjit Singh, Vice President of Engineering at GoMechanic. To leverage previous conversations, agents use keywords to search through chats. GoMechanic wanted both searching and loading to be extremely fast. “We didn’t want an agent who had 40 live chats going to wait five seconds after that page refreshed for all the chats to load,” he said.


Redis Enterprise is critical for scaling

Redis Enterprise is critical for scaling

The team uses Redis Enterprise on Google Cloud, making management and configuration dead simple. “We need to load our spare parts and accessories retail sites and our customer service chats instantaneously, and RediSearch does that with ease. We tried many solutions, but Redis Enterprise on Google Cloud delivers the best performance for its cost,” says Nagendra Kumar, Senior Manager Technology at GoMechanic.

GoMechanic adopted RediSearch for both the catalog search and customer-support chat search. “We loved Redis’ flexibility. We started out with RedisJSON, and it was super easy to convert—it probably took us only a few days to switch from their existing database to Redis. Now we use RediSearch, and it’s working fantastically,” Singh says.

This is key for GoMechanic’s developers, Kumar adds. The development team uses Python, and since all the Redis modules have python bindings, the developers are immediately productive and can start building applications with just two to three lines of code, he says.

After implementing RediSearch, the response time for the spare parts site—including display, search, and SEO—improved by more than four times. The performance improvement for the customer support chat search could be critical to the company’s growth. GoMechanic currently conducts 1,000 chats a day, and hopes to eventually turn them into a source of leads.

“When we reach 5,000 chats a day, it’s actually going to become a medium where we can start converting leads into paying customers. At that point, every bit of performance is going to become very, very critical,” Singh says. “We don’t want to lose out on a potentially paying lead simply because the chat loaded a couple of seconds later than it should have. Redis Enterprise has shown us that we can be confident it will scale with ease as we grow.”


Expanding to new use cases

The GoMechanic team plans to move its car accessories e-commerce platform to RediSearch as well. Many people in India are searching for car accessories on their phones, often with 2G or 3G mobile service. With RediSearch, the team hopes to speed page loads for users with slow connections.