Omnicore’s unique business proposition is providing its clients with an end-to-end e-commerce solution. Naturally, all the different business aspects - website, marketing, and logistics, require data and reports for accurate operations and performance measurement. Any delay or errors in data availability can lead to longer delivery times which is a deal-breaker for Omnicore’s clients. Samvel is responsible for orchestrating these data flows and ensuring that the right data is available to the right people at the right time.
Before I joined Omnicore, there were 52 different reports created on spreadsheets using Jasper Studio and were sent to a list of employees and partners via email. Analytics processes were largely ad-hoc so it was difficult to establish and measure the KPI of different processes. To make things worse, the website’s database was full of data recursion, overwritten statuses, and non-typed history transaction log errors.
When establishing a data & analytics foundation at Omnicore, Samvel started using Power BI’s inbuilt, open-source, and commercially available data connectors to load data. But, these connectors weren’t always reliable and Samvel frequently faced data accuracy issues causing reporting errors.
Loading data using Power BI’s connectors was not at all reliable or stable and we couldn’t afford that. For example, one of the most important reports, which was being used by our logistics team to detect new deliveries from partners, couldn't be unreliable as it would lead to longer delivery times and consequently, harm customer experience.
To resolve these issues, Samvel realized the importance of having an OLAP system in place and decided to build a data warehouse that will act as a single source of truth for all their reporting and analytics needs. He also realized the need for a data pipeline tool that will connect all the sources with this data warehouse and automate the data flows.
After settling on BigQuery as a data warehouse, Samvel started looking for ways to feed data into this warehouse. He came across Hevo as a potential solution and decided to activate the free trial to test the platform before coming on board as a user.
Hevo is extremely user friendly. I can build integrations with a lot of tools without time-consuming procedures like installing & configuring databases, writing scripts, researching ETL solutions, etc. Additionally, for a non-python developer like myself, the drag and drop interface is a great way to apply advanced data transformations. The best part about Hevo is that it is continuously improving, with each release they expand the list of connectors, add new features, and improve existing connectors and features.
Using Hevo, Omnicore has built pipelines from their databases - MySQL & PostgreSQL and SaaS sources - Google Sheets, OKI TOKI, Jira, Google analytics to their Google BigQuery Data Warehouse and Google Cloud Storage. Business users then use Power BI and Google Data Studio to create dashboards and reports.
Switching to a modern data stack helped Omnicore achieve reliable data ingestion and resolve data availability issues. Hevo’s pre-load transformation feature helped Samvel easily clean the data coming from their source systems and load the analytics-ready data in the data warehouse. On top of BigQuery, Samvel is now able to use the DirectQuery option in Power BI to fast-track its reporting and analytics.
Thanks to Hevo, all our ad-hoc Power BI reports are now streamlined and the number of reports came down to just 3 from 52 and they get refreshed 8 times a day instead of a weekly basis. As a result, we can now provide our business teams and customers with more meaningful insights in real-time!
Using Hevo's free historical data load Omnicore loaded 37 million events during their initial setup and now, they losslessly aggregate over 5 million events per month in real-time with 100% accuracy. Samvel estimates that setting up a data warehouse with a data pipeline system would’ve taken over a month, with Hevo, they were able to get it up and running in under 2 weeks saving almost $10K in initial set up cost.
Hevo is an excellent solution for companies undergoing a multi-directional growth, have numerous data sources, and are planning to or have a data warehouse in place. With Hevo, data is always reliable and on time!
After successfully setting up Omnicore’s data stack, Samvel is now planning to create a cross-project administrative portal, implement end-to-end analytics principles, create live KPI dashboards on LTV, CAC, Marketing funnel, ARPU, and more. Hevo is elated to play a key role in helping Omnicore deliver a superior customer experience and is excited to partner with them for their next series of data challenges.