Clicky

PromptCloud | How to implement a Big Data strategy: a step-by-step guide
 

How to implement a Big Data strategy: a step-by-step guide

How to implement a Big Data strategy: a step-by-step guide

Big Data is apparently the most overused corporate buzzword of the year 2013. In an analysis of 5000 conference call transcripts, Factset found that the term ‘big data’ was mentioned in 841 corporate calls, up 43% from the previous year’s figure of 589. Svetlana Sicular of Gartner suggests that big data was at the peak of inflated expectations in 2013 and is falling into the trough of disillusionment in Gartner’s hype cycle (see figure on the left). But we believe that businesses so far (especially in the year 2013) have been testing out big data to measure real business value and are now equipped to prove the business case behind adopting big data at a large scale. The following steps can help any business in carefully traversing the path of big data adoption and equip them with a predictable road map to measurable outcomes.

B2B-Marketers-Why-Organic-Content-Is-Your-Weapon-Of-Choice

Identify the business challenges

Using big data without concrete business problems is like sailing without a compass. It’s recommended to start with identifying the business problems at hand where additional data can be useful either by improving the existing processes, reducing costs or improving productivity. In order to put an effective big data strategy in place, decision-makers should first answer this question: What is it that we cannot do without big data, and how is that affecting us?

Prioritize the topmost business problems

Once you’ve identified the biggest problems that can be solved using big data, figure out the order in which you plan to tackle these problems. Big data is not the answer for every other business need. Identify the three or four biggest challenges that big data can help you solve so that you easily avoid the trap of trying to achieve everything and then ending up sub par.

Find relevant data sources for your business

The next step is to find how data can help you in solving the problems at hand. You must establish what kind of data can help you in dealing with these problems. Take help from line managers to gain first-hand insights on what challenges they typically face and what kind of data can help them in dealing with these problems and in doing their jobs better. Also describe what formats of data can be fed into your analytic systems to avoid any data integration troubles.

Involve the business

By now its obvious that there’s a case for big data in your organisations. But one of the most important concerns of the management executives is the real business value of big data. Establishing monetary outcomes is an important step before implementing a big data strategy. Though open-source tools such as Hadoop are easy to install and hence can be done very affordably in-house, successful deployment may take multiple iterations for which a buy-in from the top management becomes vital.

Start small, grow big

Rome wasn’t built in a day. Big projects require gradual, progressive steps to come to fruition, and the same is applicable to setting up a big data infrastructure. Starting small helps in many ways, such as: showing what benefits data can accrue to your firm; and in panning out a careful approach towards big data by taking care of the small details and in creating fall-back plans at each step as you move gradually.
It’s recommended to define an initial level of achievement through a proof of concept, and try to build on it afterwards. Companies trying to boil the ocean while starting off with their big data initiative are certainly going to have problems in proving return on investment to their stakeholders.

Know what data to use and what to exclude

Big data is really big, some large retailers have a terabyte of data on each of their customer (these companies have millions of customers). As the data-sets involved are really big in terms of scope and complexity, it becomes all the more important to differentiate good data from useless data.
It’s important to avoid analyzing data that’s not relevant to the business problems at hand. Identify data that leads to relevant insights that help you encounter real business problems. For instance, asking questions such as: what kind of data will help in identifying trends in customer defection will enable you to connect business value to big data.

Final thoughts

Following the above steps will provide a degree of cohesion to your big data implementation strategy and help you in starting out with big data adoption. A successful big data strategy is all about asking the right questions in the context of your business challenges, and then following iterations to derive key insights regarding what is most useful to your business. As the world of big data is evolving, its vital to maintain flexibility at your end, by remaining fluid to the changing outside dynamics. Approach big data with an open mind, embracing the new, undiscovered insights that big data may reveal about your business.

Image credits: Flickr.com/Iron Vision

Related Posts

No Comments

Post A Comment

Ready to discuss your requirements?

REQUEST A QUOTE
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.

Price Calculator

  • Total number of websites
  • number of records
  • including one time setup fee
  • from second month onwards
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.
  • This field is for validation purposes and should be left unchanged.