5 Ways to make Big Data work for you

5 Ways to make Big Data work for you

Information is being continuously generated by people, processes and machines. It comes from many different sources, and can quickly become overwhelming – yes, we are of course talking about Big Data. This data is an asset, but takes a lot of computing power – and sense – to get the most out of it.

Here are 5 ways to make Big Data work for your business:

1. Trends & Insights

The first step is to find out what information you have, and establish your ability to access and use it to support decision making and day-to-day operations. Information will likely stem from multiple sources, both within and outside your organization.

Raw data from sensors, machine logs, clickstreams, websites etc. presents a particular challenge. Yet you don't need to spend a fortune to reap the benefits.

Google Analytics, the company's free program, can be used to monitor website traffic levels. Dashboards present your data in context, so you know which information to look out for. You can create them from scratch, or select from the Dashboard Gallery: audience engagement, traffic sources (referrals), conversions, mobile visitors, etc.

Likewise, Google Trends will relay the top related search items, when you input a keyword or phrase. You can use the results to create focused content that will attract new visitors to your website. And Google Custom Search (free, but there's a more feature-rich version for $100 to $250 per year) can be used to identify the top search terms people use once they're on your site.

If you have a sign-up list, email marketing segmentation analysis can determine what your various subscribers are interested in. The tools needed come with most email marketing programs (like Constant Contact, Vertical Response, and AWeber.com). Some can even “learn” what recipients are interested in by noting which emails they’ve opened and what they’ve clicked on in the past. This will help you design more effective email communications.

For social media, there are tools that monitor Facebook, Twitter, blogs, images, or videos. These can keep track of what members are saying about your organization, a competitor, or terms relevant to your business, as well as help engage with customers, respond to queries, or solve any problems they might have.

2. An Enhanced Customer View

"Customer" here could mean retail clients, patients in healthcare, or suppliers in manufacturing. In any case, internal and external sources of information can be used to assess customer sentiment – how they prefer to shop, what they'll buy next, etc. –and the knowledge used to empower your staff to interact with them.

Ideally, data should be combined in context from all the applications and repositories containing customer information (CRM, ECM, supply chain, order tracking, email etc.), to give a complete view of the customer. At the point of interaction, this should be available to your staff without their having to search through multiple systems.

3. Security & Intelligence

Analysis of Big Data can uncover hidden relationships, detect patterns and prevent security threats. Technologies like stream computing and enterprise-class Apache Hadoop analytics enhance traditional security and intelligence analysis platforms by accessing data from unstructured and/or streaming Big Data sources.

For example, fraud may be detected by correlating real-time and historical account activity, to uncover abnormal user behavior and suspicious transactions. Various sources, like the internet, mobile devices, email, machine-generated data and social media can be examined for evidence of abnormal or criminal activity. By analyzing network traffic, new threats may be discovered early enough for organizations to react in real time.

4. Combinations

Computers, network devices, sensors, meters and GPS devices generate various kinds of machine data, which comes in large volumes, and requires unique visualization capabilities depending on the data type and the industry or application.

By combining machine data with existing enterprise data (like customer or product information), organizations can gain insights into their inner workings, customer experience, and transactions. Operational efficiency can be improved. Lapses in service or stock shortages can be avoided.

This kind of operational analysis also benefits from good reporting. It's likely that there will be some key performance indicators that you watch to determine whether your operations are running properly. Things like profit and loss, average revenue per employee, or labor costs. Augment the analysis, by regularly generating reports and studying your numbers, so you can make strategic decisions.

5. Store and More

The traditional data warehouse isn’t built to analyze multi-structured data, or gain new business insights from it. Big Data use requires an integrated set of technologies specifically designed for working with high-volume, high-variety and high-velocity data.

The key here is to build on your existing data warehouse infrastructure.

A pre-processing hub may be used in areas where organizations want to leave some of their data at rest, before deciding what should be moved. Infrequently accessed or old data can be offloaded from warehouse and application databases, using information integration software and tools. It can be held in low-cost storage, yet remain accessible.

Being able to process and act on information as it's happening can further reduce storage in the warehouse. Stream computing can process and analyze streaming data, without having to store it first. Stream analytics can bypass the data warehouse by optimizing it and enabling new types of analysis.