Still Think Your Business is Too Small for Big Data?by Robert Buda | Mar 25, 2015 | Oracle DBA
Last modified on August 8th, 2017 at 1:20 amReading Time: 3 minutes
Just because your business isn’t that big doesn’t mean you don’t have “big data” issues or a need for big data-style analytics to remain competitive. “Big data” is a relative term—relative to needs and capabilities for making business decisions, that is. Nearly every organization, whatever its size, will sooner or later reach the point where the volume, variety and velocity of the data it needs to analyze exceeds its storage and/or computational capacity, such that accurate and timely decision-making is impacted.
At that point, you have a big data challenge/opportunity. You can throw your data aside and fail to leverage it. Or you can find a cost-effective way to apply analytics to give your company new insights and answers that will help you compete and grow.
Understanding how to manage and analyze data to meet your evolving needs is critical, because big data analytics aren’t one-size-fits-all. You’ll potentially need to capture and integrate diverse sources of structured and unstructured data across everything from standalone department-level transactional systems to social media feeds to web forms to smart devices to external vendors’ services. Once you’ve maneuvered all this data so that query it, business users can begin to exploit it, both in planned and newly perceived ways.
As you begin thinking about a big data analytics application, top-level planning considerations include:
- Focus on business value first. What questions do you need to answer? Where does the data reside that you’ll need to process. In blogs, Facebook and other social media? In your customer transactions? Do you want to cut costs? Predict consumer buying patterns? Accelerate time-to-market for new innovations? Forecast supply and demand? Let business needs drive the technical approach.
- Analytics capability is the bottom line. How will you determine what data is relevant and how it should be extracted, stored, transformed, etc.? If you think you have “too much data,” the real problem is that your analytics environment isn’t properly tuned.
- Rely on data architecture best practices. “Big” shouldn’t mean “out of control.” Whatever the technology and implementation involved, keep your Oracle DBAs in the loop so that data movement and transformation are effectively planned for and handled.
- Make sure you have the right resources in place to deliver the business value you’re looking for. Many companies don’t have in-house expertise to manage data effectively, for example. Data analytics expertise is also scarce. Augmenting your in-house skills with an outsourced Oracle DBA can yield the most value in the shortest time at the lowest cost.
Once you know what you want to accomplish you can more effectively plan for implementation. Any big data initiative involves the acquisition, transformation and storage of large volumes of data from multiple source systems, which is then analyzed.
Where will you get the source data you need? How will you handle data queries? How will you optimize performance? How will you manage testing, and against what data? Will you need more network capacity to handle the data movement? How do you integrate the new analytics solution with any existing data warehouse or other key data sources? As your new capabilities are used more and more, how do you monitor performance and plan for growth?
These are some of the central implementation questions your team will need to answer before your big data “challenge” can become an “opportunity” for decision-makers. Planning at the IT infrastructure level is, of course, critical—but knowing what business questions you want to answer should be the driving force behind the initiative.
To ensure your big data analytics application can generate reports quickly, in appropriate formats, which provide the insights your business demands, contact Buda Consulting. A free consultation with us can be an ideal way to explore your big data analytics challenges, with an eye toward best-practice data modeling, database design, performance and more.