Factorialist

Avoid These Common Data Mistakes

More and more business relevant information is available that can significantly inform your enterprise’s operations so that more effective function is possible. This data is available almost immediately and can be processed as it streams. It can answer vital questions on trending events and give more personalized content to your clients and customers during online visits.

Data’s Emerging Role in Business

Industries are collecting big data for front-end and central purposes and are inserting data extraction points throughout their business operations. Sales information is integrated with social media comments and cellphone locations to generate geographically and topically relevant consumer data. Primary goals are to cut costs, generate sales and increase customer loyalty.

Transport services and mobile workforces are including sensors on the vehicles to provide an information base that can be used to optimize routes. Rating feedback on thousands of products can be transparently trawled to inform manufacturers and service providers of consumer preferences.

It’s quite apparent that this information is more than a luxury. Your competitors will be using it, too, and you certainly don’t want to fail at making critical decisions because you can’t keep abreast of the latest trends both inside and outside your company.

Ford used text mining algorithms to determine whether a 3 blink turn indicator was to be included in their Fiesta model. Ford decided that conducting full scale market research was too costly and instead employed text-mining algorithms to extract and summarize 10,000 comments related to this potential feature which stops signaling after 3 blinks. Ford ultimately decided to implement this feature and considered this data extraction and analysis technique successful. They had a specific question they needed to answer and gathered the data in an economical way, while succeeding in answering their question and implementing their answer.

Big Data Needs Big Planning

Companies are finding that big data doesn’t necessarily translate into easy success.

As many as 40 percent of big data projects fail. Some common big data blunders are:

Acquiring Data without a Purpose

Before collecting data it’s essential to define information goals. Data alone is not a goal but a means to achieving business objectives. Splunk Inc. employed 100 engineers for one year on a big data project that didn’t produce useable results because there had never been a well-defined objective. Founder of Radius Intelligence, Darian Shirazi calls these situations haystacks without needles because people believe that data can solve problems and they don’t do the upfront analysis and policymaking necessary for gaining worthwhile results. The primary question that needs to be asked is: What answers do you need to find?

Lack of Human Resources

There is a growing deficiency in available talent for big data implementation. As a result, more and more educational institutions are providing certificate programs and degrees in decision and data science as well as machine learning. Non-experts will also be able to crunch big data with tools from WordPress and Tumblr, which will also provide the opportunity for those who are not computer literate to create their own websites.

Biting off more than you can chew

Michael Chui, a principal of McKinsey Global Institute, is certain that big-bang implementations rarely succeed. Complex problems may require complex and expensive solutions, whereas tackling several smaller yet easier to complete projects may ultimately yield more revenue for your company.

Data Loss

Setting up a set of procedures to prevent and detect data loss in endpoint, in-motion and storage states is essential. When vital enterprise data has been corrupted and there is no backup, this can severely impact your business. Consider a cloud backup plan for automated backups.

Failure to Maintain Data Integrity

Data is represented in numerous and ambiguous formats. Names can appear to be the same but represent completely different information elements. Failure to correctly identify data when setting up an analysis will result in erroneous results.

Develop a Winning Big Data Strategy

Developing a winning big data strategy requires that you always solve a specific problem. Choose a problem that you can solve or a question you can answer in a reasonable amount of time and at a cost that will make sense. Next hire the right people. If you can’t hire them, then get your current employees trained so they can do the job. Make certain that your implementation plan includes backup and security. Design your strategy to help your business succeed rather than for some nebulous reason. Implement your strategy right away and reassess it periodically. Big data is growing rapidly.

Image credit: cooldesign on Freedigitalphotos.net

Exit mobile version