Accelerating BI’s Time-To-Value Cycle

Business Intelligence (BI) has great appeal for business managers, but its implementation can be far slower and more complex than most people realize. The need to configure database integrations and set up Extract-Transfer-Load (ETL) processes create a friction that slows down time-to-value. Advances in data analytics, as exemplified by Incorta’s Direct Data Mapping, bypass most of the roadblocks inherent in the traditional BI stack.

The challenge: Answering business questions quickly and effectively

Data analytics seems pretty intuitive. You have data from a variety of sources. Using specialized software, you can analyze data sets and discover patterns and insights that help you run your business smarter. You can use the tools to visualize the data in endlessly useful ways. In reality, it’s not so simple.

The challenge is to use data to answer business questions quickly and effectively. Neither is easy to achieve. Data complexity and the distributed nature of data tends to slow down the analytics process. In terms of effectiveness, what you want are the right answers to business questions—an outcome that is not assured without having access to all relevant data for any given issue.

In practical terms, obstacles to quick, effective analytics arise from having data stored in multiple places. This scenario is unavoidable in all but the smallest organizations. It’s often difficult to access data that are residing in different applications and repositories. Each application typically has stores data in its own format.

Access to data can also be limited by security and compliance requirements. This is a good thing, in general. You don’t want anyone to be able to access sensitive data. Indeed, some pretty serious data breaches have occurred because data managers were lax about access controls and data repository configurations, such as with Amazon S3 “Buckets.” Still, controls over data can make it hard to get the data you need to do proper analytics on a timely basis.

The difficulties with traditional solutions

It may sound odd to be talking about “traditional” BI. The entire stack is only about 20 years old. Yet, as we see in IT, 20 years can be an epoch. The dominant players sell BI stacks are that expensive, cumbersome, and brittle. Most BI solutions in use today require ETL as a starting point. It’s necessary to integrate the analytics toolset with each data source and then perform ETL. This is a time-consuming process. Someone has to perform the work, and that costs money.

From there, the traditional approach to BI calls for data aggregation, usually in the forms of data warehouses (DWs) or data marts. The BI solution also needs to map to a traditional star schema. Data sources have to be “joined” and correlated before analysis can take place. These processes, too, take time and cost money. Even if the work is done in-house, there’s an expense. Employees have to get paid, after all. Then, you have the design of the dashboards themselves. The time cycle needed to construct a working BI solution on a traditional stack can be up to six months, or even longer with a success rate of less than 40%.

If your goal is to use data to answer business questions quickly, six months is a rather long time to wait. The other parameter—answering the questions effectively—is also in doubt in the traditional BI model. This is mostly due to change management challenges. The initial rollout of the BI solution and its dashboards will likely be suitable for effective BI. However, as we all know, the requirements for analytics change frequently due to shifts in the market or changes to business strategy.

Having to go back and redo the ETL, redesign the DW and dashboards, and so forth causes delays. Many of Broadcom’s business units, for example, made decisions based on information sitting in Oracle ERP, Workday, Model N, Oracle Demantra, and Microsoft Excel. It took the chipmaker between six and twelve weeks to generate a new analytical dashboard or report—despite having made a substantial investment in a traditional data warehouse and multiple data marts. The people who did the work the first time around may not be available to make changes. The brittleness of the system becomes an issue as well. Every change to the BI solution becomes a project, which in turn is a drag on resources.

End users also affect the quality and speed of the analytics process. People may not have the time or interest to learn how to use esoteric BI tools. They may not want to become experts in creating dashboards or running data visualization routines in Power BI and so forth.

How Incorta accelerates BI time-to-value

Incorta is a Unified Data Analytics Platform -- meaning that it is an end-to-end platform, from the acquisition of data to deliver on insights to users that works in the cloud or on-premises.  It introduces a new approach that accelerates time-to-value.

The solution can thus aggregate complex business data in real-time. The runtime for analytics workloads drops from hours to seconds. The solution can deliver sub-second query performance (e.g. execute 100M rows in 0.8 Seconds). Self-service users can create new reports and dashboards in days instead of months.

With this streamlined architecture and a new generation of analytics technology, Incorta dramatically accelerates the time-to-value for BI. It enables you to get fast, effective answers to pressing business questions based on data—no matter how diverse the data sets. And, as inevitable change cycles occur, the solution is fast to adapt. As a result, you can stay ahead of business trends by always having access to rapid, accurate BI capabilities.

One more thing, you can get a free cloud trial, free training courses with certificates, and you can attend a weekly live demo with Incorta experts.