Case Study: Ritual’s Move to Real-Time Analytics to Personalize the Multivitamin Experience

March 31, 2021

,

Ritual is a health-meets-technology company reimagining the products we use every day, starting with the multivitamin. With an in-house team of scientists and researchers, Ritual invests in creating high-quality, science-backed multivitamins formulated to help fill common nutrient gaps in diets at different life stages and making those nutrients readily traceable though the first visible supply chain of its kind.*

Taking an evidence-based approach, Ritual has designed multivitamins with key high-quality, traceable nutrients in bioavailable forms and even tested its flagship product, Essential for Women 18+, in a university-led clinical study. The team has also developed Made Traceable standards, a published standard that shows where the ingredients come from, why they’re there and how the finished product is validated.

Ritual started in 2016 with a single reimagined multivitamin for women and has since launched products for different stages of her life and seen tremendous growth, crossing the threshold of over 1M multivitamin bottle sales in 2019. As more women trusted Ritual to help support their everyday health, they started asking for multivitamins for their partners and kids. In 2020, Ritual expanded beyond women for the first time with the launch of Essential for Men 18+, quickly followed by a multivitamin for teens and a sugar-free gummy multivitamin for kids.

Ritual has bundled multivitamin subscriptions, allowing subscribers to not only buy for themselves but also for their family. The team made it easier for users to add new products to their subscription with personalized recommendations.

Essential Prenatal, Essential for Men 18+ and Essential for Women 18+

Three of Ritual’s product lines- Essential Prenatal, Essential for Men 18+ and Essential for Women 18+.

Personalized Multivitamin Offers And Bundles

As Ritual expanded into new product lines, a key to effective monetization was personalized experiences.

Personalization is no new concept to the e-commerce industry. 80% of shoppers are more likely to buy from brands that offer personalized experiences. And, personalization is shown to increase sales by 20% (Bloomreach). The team at Ritual looked to personalized offers and bundles to increase the average order value and the lifetime value of the subscriber.

Ritual piloted personalized banner ads in their online portal, a spot where subscribers could add new products to their existing subscriptions. Prior to personalization, there was a generic promotion for the different product lines in the portal. The team at Ritual believed that they could make more relevant offers and increase the conversion rate with personalization.

The seamless integration of personalization in the online portal prompted Ritual to expand to personalize the cart checkout experience and email campaigns. As customers went to checkout, they would receive personalized offers to add product lines to their subscription. Post checkout emails and delivery notifications would also include personalized offers.

Ritual’s cart checkout experience featuring personalized upsell offers

Ritual’s cart checkout experience featuring personalized upsell offers.

Evaluating The Data Stack For Personalization

The front end of the site needs to query a data backend for personalization; Ritual explored Rockset to serve personalizations.

Ritual uses cloud data warehouse Snowflake for ad-hoc analysis, periodic reporting and machine learning model creation. The team knew from the outset that Snowflake would not meet the sub-second latency requirements of the site at scale and looked to Rockset as a potential speed layer.

Rockset is a real-time indexing database that delivers millisecond-latency search, aggregations and joins on terabytes of semi-structured data. The system is designed to make real-time analytics fast, flexible and easy- removing the need for data preparation, index management and operations.

The team at Ritual started a free trial of Rockset and was impressed at the ease of use. The data team could continue using a SQL interface, the language the team was fluent in, and engineers could access developer workflows including APIs and CI/CD for user-facing personalization. With a data team of one, Ritual was keen on a solution that would be quick to get off the ground and could scale with the team over time. The team went live on personalization in a single week.

While the immediate need was personalization, the team had a larger vision for real-time analytics in their online experience. Rockset provided both sub-second query latency as well as data latency in seconds, opening the door for real-time personalization as part of Ritual’s online pre-purchase experience.

Implementing Personalization With Rockset

Ritual adopted one-to-one targeting using affinity-based personalization. An affinity profile ranks customer preferences based on their site activity such as product views and past purchases. Given the ranked preferences, the model is designed to recommend a promotion or bundled offering.

The data science team at Ritual builds models in Snowflake and updates the affinity model with the latest rankings data nightly into Rockset. A nightly batch job drops the file from Snowflake into S3 using DBT. Rockset automatically pulls from the latest affinity list to serve personalized recommendations.

The flow is completely automated so a one person team can manage it. Rockset’s built-in connector to S3 automatically picks up new data as soon as it lands in the bucket. Rockset indexes the data in a Converged Index™, which enables queries on new data to be available almost instantly and serve sub-second latency personalized offers.

Ritual’s data stack for serving personalized cart recommendations, email campaigns and banner ads

Ritual’s data stack for serving personalized cart recommendations, email campaigns and banner ads.

The data team builds queries using the scores from the model and additional relevance logic. The queries are shared with the engineering team using Rockset’s Query Lambdas which are named, parameterized queries that can be executed from a dedicated REST endpoint. Query Lambdas enable teams to build applications backed by Rockset as opposed to querying raw SQL directly from the application.

The backend and front-end engineering teams use the Query Lambdas to trigger personalized offers, never interacting directly with Rockset. It’s a simple handoff to the engineering team without the data team needing to build or manage their own APIs. The Query Lambdas can also be versioned so that the data team can test changes in their logic without impacting queries in production.

Kira Furuichi, the Manager of Data Science and Analytics at Ritual, found Query Lambdas to be a big win with the engineering team. "Using data to create custom, relevant site experiences has been made simple with Rockset. My engineering team is wowed by the query speed and the ease with which they can consume data APIs created on Rockset."

Real-time Analytics On The Horizon

Personalization was a key component of Ritual’s go to market strategy for new product lines and building relevant, customized experiences continues to be a focus area for the company. Ritual foresees using real-time analytics to continue customizing content experiences around its prenatal and postnatal multivitamins and also incorporating real-time personalization to product pages to further increase conversion rates.

The ease of use was paramount to Ritual’s success with personalization. "Ritual was able to move much faster on customizing our subscriber experience once we had Rockset in our stack. We were serving personalized offers from Rockset within a week. We found a service that supports Ritual's personalization today and opens a door to incorporating more signals in the future," says Kira Furuichi.


* These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease.