Beginner Big Data

Beginner Big Data

http://www.wpwkly.com/whitepapers/The-Big-Data%20Holy-Grail.pdf

Key takeaways:
Harness big data
Manage information
Analyse high performance
Deploy big data analytics

This unstructured data is also known as
“dark data,” emphasizing its resistance to
analysis using traditional tools.

Here are a few examples of specific ways  organizations can put big data to work:
• Combine analysis of SKUs and Social Media feedback to spot buying trends.
• Combine analysis of inventory levels and customer loyalty program data to craft  high-redemption coupon offers that  maximize profit and clear out inventory.
• Use email/call-center transcript analysis to improve customer satisfaction
• Examine social media chatter for sentiment analysis
• Analyze collections data to target delinquent accounts with the highest  probability of payment

Your big data strategy should factor in three  key components: information management,  high-performance analytics, and flexible  deployment.
First Key:
Information Management
The focus of Information Management (IM) is the ability of your organization to capture, organize, store, and deliver the right information to the right people.

Best Practice: Foster a data-driven culture

Second Key:
High-Performance Analytics
Best Practice: Know what business goals your data will support

Third Key:
Flexible Deployment
Best Practice: Focus on late binding for data analysis
Most data analysis uses an early binding approach, where data is fit into a specific schema and bound to business rules at the outset. This is possible for simple, well-defined, structured data. But, if
you’re searching free text, machinegenerated data, or diverse data sets, it isn’t practical, or even sometimes possible, to understand and model the data at the early stages. And, in most cases, you don’t want to because early binding can irrevocably alter the data, tailoring it for a specific set of analytics