Anvyl is a software startup that offers its B2C customers a global, end-to-end view of their supply chain to streamline the production of physical products. Their unique Intelligent Supplier Relationship Management Platform uses data analytics to provide visibility, automation, and insight into complex supply chain processes.
Lack of end-to-end visibility made data hard to find
Anvyl provides its customers with unprecedented data visibility into their supply chain processes by centralizing data, automating processes, and facilitating collaboration across supply chains. But when the startup’s client-base began to grow rapidly, Anvyl needed an immediate solution to the inconsistent data cataloging practices that were making it hard for internal teams to serve their customers.
As the company’s data volumes expanded, disparate naming conventions and term definitions made it difficult for internal users—including developers, testers, and designers—to find and understand specific datasets. They had to manually search over 200 columns in their Looker BI platform to serve their customers, often leading to incomplete or duplicate datasets and confusing reports.
“Without an end-to-end view of our data pipeline, deleting views or tables within Looker could easily break critical dashboards for users downstream. To mitigate that, we stopped deleting data in Looker. With all that extra data, users trying to generate reports would have 200+ columns to choose from. All those options meant many users weren’t sure what information they actually needed in their reports,” says Christine Liu, the Data and Analytics Lead at Anvyl.
Their small team was working with a huge volume of unclassified data and limited insight into which data was valuable to internal users and clients. Improving data discovery on their own would require them to manually classify, catalog, and record metadata for their rapidly growing data volumes—a task Christine says would have taken them up to a year.
Anvyl's and analytics team needed a solution to help bridge the gap between how the technical teams recorded and classified data, and how other internal teams naturally searched for that data.
Without Select Star, we were blind to how our team was using data. Now we can discover what's between the dbt transformation layer, BigQuery job pipelines, and what's in our Looker.
Select Star creates consistency across the data pipeline
Select Star, an intelligent data discovery platform, provides Anvyl with deep insight into their data and standardizes governance best practices—without Anvyl’s data and analytics team having to dedicate time to customizing the software. They use Select Star to:
Stabilize data when making changes
Select Star’s data lineage view provides Anvyl with insight into how changes in one table will affect others by automatically detecting and displaying where data is pulled from. This end-to-end visibility saves Christine’s team hours that would otherwise be spent fixing broken tables and dashboards when errors upstream impact dependencies downstream.
“Before Select Star, we had no way to know what information our users were reviewing in Looker. We saw immediate value because Select Star’s popularity score gave us visibility into who was viewing a dashboard or column,” Christine says.
Reduce and remove duplicated data
The integrated popularity score—calculated automatically from the query log history—helps the Anvyl team understand what data, dashboards, and columns are valuable to users. Using popularity alongside data lineage makes it easy for them to understand what can be removed and to guarantee it’s safe to do so.
Standardize KPIs and metrics
Better visibility into Anvyl’s data helps Christine’s team find and rectify gaps in how end-users understand and define terms. Clearly defining KPIs and metrics ensures internal users operate under the same understanding, streamlines data cataloging, and makes data easier to search for. Revealing data models, measures, and dimensions helps each team understand how KPIs are calculated and provides full visibility into what data goes into a particular metric.
It's been amazing to see how Select Star has supported our needs. We're receiving substantial value without putting much effort into customizing the system.
As a rapidly growing startup, Anvyl needs a scalable data governance and management solution that is simple to maintain as its data gravity continues to expand. Built-in onboarding support helps team members who are not familiar with Anvyl’s data to leverage insights through easy self-serve data access.
Increased visibility and standardization have transformed how Anvyl accesses and leverages its data. Select Star provides Anvyl with visibility into its most valuable data by identifying the datasets that are most popular among users. This in turn enables Anvyl to visualize the full data pipeline and gives its users the information they need to find their data quickly and easily.
The data and analytics team saves up to a year on manual tasks
Select Star’s automated metadata ingestion and data cataloging have made Anvyl’s software even more robust. Anvyl’s internal teams have new insights into how to strengthen their products with better reporting, and their customers can quickly find the data they need to make critical business decisions.
With improved data discovery and searchability, the Data and Analytics team now spends less time fielding questions from users, including both customers and internal teams. Data transparency and visibility are now available for every aspect of their customers’ product development and production, even as their client roster continues to grow.
Select Star’s scalable solution is equipped to grow alongside Anvyl and support their data cataloging needs every step of the way.