< blog
6 min read

Crystal-Clear Data Visibility: How Artlist Streamlined Workflows and Reduced Snowflake Costs in 30 Days

Artlist, a leading SaaS creative technology company, leveraged Seemore Data to enhance its data observability, streamline workflows and optimize Snowflake costs.

Quick wins achieved by Artlist in less than One Month:

  • Super Fast Onboarding: After just 30 min onboarding and data stack integrations, Artlist was benefiting from Seemore Data after a few hours
  • 30% Cost Reduction: 3 pipelines consuming excessive Snowflake resources detected

 

Cost & efficiency highlights:

  • $7,300 Instantly Saved: off the annual cost of just one dashboard
  • $2,000 Saved: by shutting down only one unused table
  • 50% Run-Time Reduction: detecting and optimizing a process that was constantly running

 

“Seemore Data gave us the visibility to understand the lineage behind our data pipelines, so now a non-technical person, such as an analyst, can understand how things are being constructed rather than having to ask the data engineering team.”
Hannan Kravitz, Data Engineering Team Leader, Artlist

Rapid Growth Was Putting Pressure on Artlist’s Growing Data Team

Artlist, a leading Saas creative technology company that offers content creators access to an extensive catalog of over 2.5 million high-quality, royalty-free digital assets, has experienced rapid expansion since its launch in 2016.

With over 26 million users, including top-tier global brands like Google, Amazon, Microsoft, and Calvin Klein, this growth placed immense pressure on Artlist’s growing data team, which is made up of two teams, 23 engineers, and has experienced 800% team growth in 3 years.

Data Engineering Bottlenecks: The rapid growth necessitated tight timelines, often leading to shortcuts that deviated from best practices when using tools such as Tableau, Airflow and Snowflake.

Growth in Complex Data Pipelines: Difficulties in understanding data pipelines, business metric calculations, and data transformations due to a lack of comprehensive data lineage tools.

Dependency on Engineering: Without a data observability tool, non-technical users, such as analysts, struggled to grasp the intricacies of the data constructions, frequently needing to consult the engineering team.

“Rapid growth meant more and more pipelines were being added and more models were built, so we needed some sort of basic understanding of what was going on,” said Hannan Kravitz, Data Engineering Team Leader, Artlist.

The Solution: Complete and Actionable Observability

Seemore Data provided complete lineage visibility and insights, enabling both technical and non-technical users to see and understand the data flows, management and usage processes. This reduced the dependency on engineering for insights, allowing analysts to directly engage with the data pipelines and models, streamlining operations and improving efficiency across the board.

“We had a month with a growth of 30% in costs and knew something was not right,” says Krawitz. “With Seemore Data’s dashboard we were able to drill into the workflow and the specific job and the specific task that was consuming a lot of resources. This would have taken a substantial amount of time if we tried to do this ourselves, because Seemore Data breaks down every step in the entire workflow.”

How the Data Team at Artlist Became More Efficient:

Following a seamless 30-minute onboarding process to all tools on their data stack, Artlist quickly saw substantial improvements from the get go:

  1. Optimized Pipelines:
    • $2,000 (approx) saved annually after unused datatable being refreshed every hour was identified and shut down.
    • $7,300 reduction of the annual cost of one dashboard after Seemore Data suggested adding a clustering column to their touchpoint table. A modified MERGE INTO statement was then used to update rows only when changes are detected.
    • 30% reduction in costs after the detection of three pipelines consuming excessive resources due to improperly written Snowflake queries.
  2. Improved Data Understanding:
    Enabled non-technical users to see data flows and understand business metric calculations without needing to consult the engineering team.
  3. Streamlined Data Operations:
    The Seemore Data dashboard allowed for real-time insights into workflows, significantly reducing the time required to identify and resolve inefficiencies.
  4. Smoother Data Team Onboarding:
    New data team members gain a better understanding of our data pipelines as it is easy to understand data flows using Seemore Data and optimize them thanks to the insights it provides.

 

Beyond Phase One: Extending Usage to Business Stakeholders

The first phase of Seemore Data deployment is focused on the analytics team and giving them the ability to see the data lineage. This allows them to see the flow of data and to understand how business metrics are calculated without the need of the engineering team.

The next phase will involve business stakeholders using Seemore Data to understand how things are constructed without the need for them to ask the analytics team how things are done in terms of metrics building and definitions.

“If they understand that every change can significantly impact costs, then they will understand why we do some push backs to requirements they ask us to do,” says Krwitz. “If they have visibility of the cost and understand the work and effort required, I think it might ease the negotiation around the priorities of tasks.”

To discover how Seemore Data can boost your data observability, streamline your workflows and optimize your data costs visit seemoredata.io to book a demo.

2 min read

Seemore Data Appoints Data Management and Technology Veteran Yuda Borochov to Its Advisory Board

Data Budget
8 min read

Dear Rest of the Company — WTF!? Someone is Draining My Data Budget!

21 min read

Understanding Snowflake Data Transfer Costs: A Comprehensive Guide

Cool, now
what can you DO with this?

data ROI