“we needed a tool that would take away the need to support countless ETL processes on one hand and manage all of the day to day data and storage optimizations on the other hand so that me and my analysts can focus on building our models and providing value to the organization.”
Honeybook was founded in 2013 with the main goal to provide these entrepreneurs an easy way to manage all their jobs seamlessly, so they can spend more time on their craft, and less time on the administrative and marketing pains of running a small business. In their four years of operation Honeybook has become the biggest and the most widely recognized player in the creative entrepreneur space, raising funds from investors such as Norwest Venture Partners, AngelList founder Naval Ravikant, Medium CEO Ev Williams, Bebo’s Michael Birch, and VCs such as Aleph, and Hillsven.
Honeybook has established a loyal consumer following, due in large to its invite only collaborative platform that streamlines all of the elements and processes of event planning—from booking and collecting payments to collaboration—through an elegant, intuitive interface. Honeybook connects markets and people, around multiple transactions from proposals to payments.
The company has also been focused on developing its mobile and tablet applications. Utilizing countless different sources of data to build a holistic view of the company’s ecosystem and understanding the underlying forces that drive the revenue stream, Honeybook relies on a team of highly sophisticated data scientists to provide insights to top management.
Honeybook has tried out numerous different ETL solutions to pipeline their data into their vanilla redshift cluster and then visualize it on third party or in house built BI tool. When Honeybook moved to Panoply, they replaced the Vanilla Redshift cluster with a Panoply Powered AWS stack and immediately transferred all of their ETL management to Panoply’s pipeline dashboard. “When we were running our own Redshift cluster we were constantly dealing with maintenance tasks of performance optimization and scaling” says Abhi Sivasailam, Honeybook’s lead data scientist. “we needed a tool that would take away the need to support countless ETL processes on one hand and manage all of the day to day data and storage optimizations on the other hand so that me and my analysts can focus on building our models and providing value to the organization.”
Honeybook uses Panoply to streamline JSON in realtime and store daily logs and tables, generating around terrabytes of analytical ready data per day. The company also uses Panoply’s archiving mechanism built on S3 to lower storage costs. Panoply’s self optimizing warehouse enables the analytics team and data scientists to actively seek value as schema and query self-evolve towards different data use cases and everchanging analytical use cases.
Features powered by Panoply also include:
• Automatic Data Modeling allows Honeybook to send arbitrary data in any standard format, without the overhead of maintaining schemas and altering data types. Panoply’s platform collects that data, parses it, and determines the best data model to use. This supports a continuous development cycle, where code and data changes are push to production on a daily basis, without requiring any reconfigurations. • The Schema Optimization process optimizes Honeybook’s schemas based on query patterns. Panoply’s platform continuously examines the queries that are executed on the data warehouse, and generates a statistical analysis that maps the importance and performance of each column. Panoply modifies the configuration of the schema: from data types, to compression, sorting & distribution. This makes Panoply a self-evolving warehouse and shaping itself to best fit the needs of the analyst.
• Materialized Transformations is Panoply’s alternative to the aging ETL paradigm. Panoply is an ETL-less platform, which behaves as both a data warehouse & a data lake. Data is cleaned and enhanced, but is otherwise left intact, allowing analysts to query the raw data the way they see fit. Transformations are still required in order to create meaningful domain-specific aggregated datasets. Panoply’s SQL-based Transformations enabled analysts to design their own logic, in real-time.
• Data Archiving is an important feature for many organizations. This is usually done after generating the aforementioned transformations & aggregations, thus keeping the insights from the data available without having to pay for raw-data storage. But it’s really not just about cost, reducing the data-size with Archiving also provides a huge performance improvements for your daily queries.
• Users & Permissionshas traditionally been a huge headache for teams that care about the privacy and security of their data, usually requiring complicated processes & configurations to manage the access & permissions of the different users within the organization. Panoply’s fully-encrypted data warehouse makes managing users access as easy as sending out invites to colleagues. Honeybook is grateful for Panoply.io’s architects that helped with their analytical architecture development.
Using Panoply.io Honeybook was able to save countless of man hours in analytical infrastructure development and ongoing optimization and get up and running in a matter of days. However, most important to Honeybook is the opportunity cost. “With Panoply.io” says Abhi “our data scientists and analysts can focus their energies on value added challenges instead of constant upkeeping of our analytics infrastructure.”