How to Get ETL Under Control
What do you do when your ETL load is out of control, your team is maxed out, and turnaround times are defined in weeks instead of days? Since ETL is still 70% of every analytics project, the answer to this question will likely provide a breakthrough in the speed of insight delivery to the business.
According to Eckerson Group’s Wayne Eckerson in his article Why Modern Data Integration, “From the business perspective, modern data integration has become instrumental simply because a lot of data that can be useful to organizations has a very short shelf life. Business processes, initiatives, and other activities must be able to tap into data as soon as possible.”
What if you could have data preparation and business intelligence in a single platform to speed the delivery of intelligence to the business? When the data analyst can simply transform data within the analytics platform, it closes the gap between IT and business. That’s where Yellowfin comes in.
Wasted ETL Cycles
Think about the time that gets wasted in the typical ETL cycle. A set of high-level business requirements are communicated to IT. IT comes up with a plan to acquire, cleanse and transform data for use by the business. A massive effort is undertaken to prepare the data. The data is loaded into a business intelligence tool. A data analyst designs reports and dashboards around the data. And weeks later, the business user finally sees the work that has been done.
Because the organization responsible for ETL is so far away from the business, that first delivery of data is rarely, if ever, the right data. In the old world, the feedback goes back to the ETL team and the requested changes add additional weeks to the data delivery process. The entire ETL cycle is outdated and inefficient.
Closing the Gap Between IT and Business
Yellowfin 7.4 closes the gap between IT and business by putting data transformation capabilities in the hands of the data analyst. The addition of data preparation in Yellowfin 7.4 is not meant to replace ETL; it is meant to augment and accelerate information delivery.
Look at it this way. When the first shipment of data arrives at the Yellowfin docks, the data analyst can immediately begin running assisted data discovery to uncover what is in the data and automatically generate new visualizations and dashboard components. Within minutes, the business users have their first look at the data. If the published dashboards don’t meet the needs of the business, the business users can communicate back to the data analyst via their built-in social communication.
Instead of going back to the ETL team to request some changes, the data analyst can work within Yellowfin to make some changes to the data. The changes are back in the hands of the business within minutes. Once the front-end team gets it right, they can make a prioritized request for more codified and governed changes to be made by the ETL team.
The principle is powerful: the closer you can get the ETL and business, the better you will optimize the work being done by the ETL team. And remember, data preparation is 70% of every analytics project. The key to reducing the cost of ETL and getting ETL back under control is putting data preparation closer to the business user. How do you do that? By putting it in the hands of the data analyst.
This is Yellowfin 7.4.
Data Transformation Made Easy
Is your data poor quality or badly structured for reporting? Are you looking to speed up the delivery of your reports? Join the webinar ‘Data Transformation Made Easy’ as our experts share best practices on data transformation, and explain why preparing and enriching your data needn’t be a complex or laborious task.
You will discover how to:
- Obtain inexpensive and fast reporting while migrating data out of a production database into a Data Warehouse
- Easily improve your data quality by modifying and hacking up the data and then pushing this output into a database of your choice
- Build your own transformations quickly and easily, by managing ETL-like operations