To remain viable and competitive in the global marketplace, organizations must adopt a data-driven mindset. By harnessing your data, you can make better, faster decisions to enhance customer satisfaction, improve your business and IT operations, and more.

Thus far in this blog series about the big data journey, we’ve seen the importance of understanding your business challenges and your analytical maturity level. Also, we explored how rapidly growing data volumes demand effective management .

In this last post of the series, it’s time to roll up our sleeves. We’ll take a look at how you can take a big data solution through its paces.

Get ready: Understand your returns

Before making any project-specific or technology-related decisions, IT and business leaders must have a meeting of the minds. Clinching agreement on the business objectives of your big data project helps garner support and reduce project risk.

Return on investment (ROI) is a key criterion that should align with organization-wide goals and initiatives. Knowing the project’s ROI helps business leaders decide whether they should invest in a big data project.

ROI also gives you a way to measure the value of your project. Potential returns can be thought of as business outcomes. Examples include faster time to market, improved operational efficiency, increased revenue and better product quality. These business outcomes should be translated into specific metrics or performance indicators. For example, you may choose to define operational efficiency as how well you meet your service-level agreements (SLAs).

You also need to factor in costs related to hardware and software acquisitions, training, overhead and facilities. Don’t include sunk costs — which have already been incurred and cannot be recovered — but do measure avoidable, variable legacy costs.

Get set: Assess your environment

An assessment can help you understand your current business environment and how best to execute a big data project within that environment. The assessment should involve evaluating the fundamental components of a big data project:

  • User profile. Understand who your primary users will be. What do they want to know? How does big data impact them? What are their skills? Examples: salesperson, financial analyst, data scientist, IT architect
  • Technology. Review existing technologies in use and identify the right technologies needed to deliver business outcomes. Examples: SQL database platform, Apache™ Hadoop® platform, in-memory appliances, database accelerators
  • Skills. Review existing skill sets and determine what other skills are necessary to support the project. Examples: Java, Apache™ Pig, R, managing data from heterogeneous sources
  • Measurement. Select criteria that help you define project success. Examples: Defect rate, gross margin change by product, speed of customer service, churn rate

You should also consider potential use cases. For example, your marketing team might need to identify how a customer feels about a product and track that over time. Information gathered from sentiment analysis can be used to indicate both intent to purchase as well as dissatisfaction with a product. Or perhaps your IT team wants to gain experience with Hadoop to see how it can complement your existing data systems.

Prioritize the use cases by how well they align with your organizational goals and whether you have the technology and skills needed to execute on the use case. To improve your probability of success, select one use case that offers the best business opportunity and focus on addressing it before starting on the other use cases. When you tackle your first use case, you will learn things as you go. This smooths the way for successive iterations.

Go: Test before you invest

Now, you’re ready to slip into the driver’s seat. Because of the diverse and complex nature of big data technologies, a test drive helps you zero in on solutions that meet your specific business needs.