Liv Up

The chefs at food tech startup Liv Up, based in São Paulo, Brazil, partner with local producers to produce healthy, tasty meals, which the company delivers across São Paulo, Rio de Janeiro, and Belo Horizonte. Information technology is a big part of their success story, and something they depend upon for competitive advantage.

As a vertically integrated startup, Liv Up generates data on all steps of production, from farmers to sales. Consolidating all of the data from disparate sources generates the infrastructure necessary to support business and operations decisions, and is essential to improving the experience they provide to customers.

Head of Analytics Luiz Arakaki says, "We run production databases in MongoDB, and it was a challenge to integrate MongoDB's NoSQL structure into our analytics stack, which is based on a PostgreSQL data warehouse. We also have integrations from Zendesk and Google Analytics that we want to load into the data warehouse to build a single source of truth."

Liv Up evaluated half a dozen alternatives for ETL. "Stitch had the best solution for transforming NoSQL to SQL – and the best price," Arakaki says. "Stitch does an awesome job of integrating NoSQL in SQL."

The company began using Stitch integrations to load data into its data warehouse. Once the raw data is in place, "we use Pentaho Data Integration and Python scripts to build a transformed layer into which analytics tools Metabase and Tableau plug in. It took one data engineer about one month to integrate all of our data sources and develop transformations. By using Stitch we save at least eight hours a week in our ETL process, and Stitch is 50 to 80% cheaper than the other tools we considered."

Arakaki says, "We're very impressed with Stitch's reliability and replication frequency." Nevertheless, glitches happen – but when they do, the problem doesn't last for long. "We just had a problem with our MongoDB integration, and the Stitch staff provided free rows to fix it.

"We're very impressed with Stitch's reliability and replication frequency," Arakaki says. "We have our data warehouse synced in almost real time. That's infrastructure that no other startup in Brazil at our stage has."