Getting started with Big Data


Leveraging Big Data can help every company significantly improve competitiveness and financial results. At the same time, poorly conceived and executed initiatives can lead to wasted investment and organizational distraction. Not surprisingly, the question of how best to reap the benefits of Big Data is triggering extensive deliberations in many companies. What is ‘best practice’ in launching a Big Data strategy?

Big Data is a set of activities for collecting and analyzing various types of data located within and outside the organization. The insights derived from this analysis are used to enhance business performance such as boosting advertising efficiency, improving supply chain responsiveness or improving service levels.

Most large companies are already in the Big Data business. The amount of data collected is growing exponentially thanks to the digitization of virtually every customer and operational interaction. In a typical Fortune 500 firm, terabytes of data are being amassed through regular business activities such as point-of-sale transactions, barcode tracking, web traffic or social media communications. This data torrent – when properly mined — affords management a valuable opportunity to learn about consumer behaviour or internal operations, enabling them to optimize tactics for better performance. At the same time, realizing the Big Data vision presents significant technical and organizational challenges. These challenges can increase the chances that managers will embark on expensive or poorly designed initiatives – or become paralyzed due to complexity.

In our experience, the best way to get into Big Data is to start with a sensible roll out plan and leverage best practices. This plan should consider four key elements:

1.  Data

Any plan should begin with a review of the relevant internal and external data, according to the 4 Vs: volume (the amount of data and its location); variety (types of data, both structured and unstructured); velocity (how quickly the data changes) and veracity (the accuracy and availability of the data). In many firms, data is siloed by function or business line; is not standardized and; it comes in various stages of completeness. Getting quality data can be difficult and time-consuming. It may be desirable to outsource this data integration and clean up to specialist firms who can make it ‘analytics-ready.’

2.  Hypotheses

It is easy to get side-tracked if you dive right into analysis without any strategic guideposts. Not all insights are equally important. Like other major initiatives, it is essential the Big Data effort links to business priorities and metrics. One way to do this is to start with a limited number of pilots based on specific hypotheses that directly impact strategic goals. Successful pilots can generate early wins that justify further investment, and can produce important insights around the business, as well as test out first generation capabilities.

3.  Analytics

To effectively and efficiently mine the data, the team should carefully choose the appropriate analytical methodology or model for each business problem. The analytics will vary whether the goal is workflow optimization (e.g., minimizing inventory levels, delivery times) or predictive analytics (e.g., anticipating consumer behaviour, forecasting events). However, managers can easily over-speculate on solutions, choosing costly and complicated tools that require expensive or scarce talent. Judicious CIOs will take a “great is the enemy of good’ approach to choosing their models and depth of analysis.

4.  Capabilities

Many IT environments are not conducive to quick or easy Big Data deployments. These infrastructures can be a heterogeneous mix of new and legacy hardware & software, lacking in data standardization and centralized control. To exploit Big Data opportunities, firms will need a unique combination of data experts, software tools and management capabilities as well as supporting governance practices. This capability should be developed with practicality in mind. Initially, CIOs could outsource Big Data needs to a cloud-based analytics service limiting upfront investment and accelerating time to value. Over the long term, the organization can look to develop world-class capabilities through employing specialized talent, bespoke software tools and private cloud architectures.

As with other strategic initiatives, a prudent way of getting into Big Data would be to start small and target actionable insights. Ongoing attention should be paid to ensure the learnings are understood by the staff and implemented into existing workflows. Where necessary, new processes or practices may be needed to fully leverage the insights. Learning by doing will prompt managers to connect different analytical models together to address wider problems that span functions and business units.

Firms that are winning with Big Data are often the quickest out of the gate with a practical plan, based on a thorough understanding of their data, staff and IT environment. Big Data will be a game changer for companies who can deploy the right analytics and capabilities against their most pressing business issues.

For more information on our services and work, please visit the Quanta Consulting Inc. web site.

Advertisements

No comments yet

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: