Hi. Welcome to Ingram Micro.

Please choose your role, so we can direct you to what you’re looking for.

If you’d like to learn more about Ingram Micro global initiatives and operations, visit ingrammicro.com.

Big Data Consulting For Experts: Six Things Most VARs Don't Know

April 08, 2017

Big Data Consulting For Experts: Six Things Most VARs Don't Know

One of the challenges with big data consulting is that you are supposed to have all the answers. That’s why VARs are paid for big data consulting engagements; because they have the answers that the customer does not. Of course, even the smartest reseller can’t know everything, but if you apply the right attitude and some common sense you can have all the answers, or at least you know where to find them.

Sales of big data projects continue to grow at an incredible rate. IDC research predicts spending on big data will hit $125 billion this year. IDC also predicts that much of that growth is being built on video and audio campaigns that will collect unstructured data for analysis, and on Data as a Service (DaaS) that are going to continue to grow. Wikibon analysts note that the big data market is maturing because of the increasing number of reseller agreements being signed between big data and non-big data vendors, which is making it easier to integrate big data technologies.

VARs are clearly emerging as big data experts, which means brushing up on your big data consulting skills. Here are some of the most common misconceptions that many experts have about big data.

  1. Big Data Is Inherently Valuable – The data itself is not valuable. It is the insight derived from the data that has value. Big data analytics are designed to find the insights and patterns within the data. The data itself has no intrinsic value unless it adds relevant insight to address the big data use case.

  1. The More Data the Better – A common misconception is that big data is really big, and the more data sources you pour into the process the better the outcome. In fact, the more data you add the more complex the analytics and the less reliable the results. You want to match the appropriate data sets to address the big data use case. Use only the key performance indicators and metrics that have an impact on the business and are relevant to formulate data-driven decisions.

  1. Only Big Data Scientists Understand Big Data – You don’t need a big data scientist to extract valuable insight from big data. Data scientists can help you interpret the data but they aren’t the only ones who can gain insight. Dashboards and graphic representational tools make it easier to present big data findings for easy interpretation.

  1. Big Data Is Expensive – Big data is moving into the mainstream, which means off-the-shelf hardware is readily available to support big data projects. The cost of big data hardware and deployment continues to drop.

  1. Big Data Is Only for Big Companies – Any company of any size in any market can benefit from big data insights. The cost of big data tools continues to drop, and big data can run on cheap commodity hardware.

  2. Big Data Can Answer Everything – Big data is not a magic wand and it won’t answer all your questions. Big data is most useful when you have different data types that can be used together to shed new light on a business problem. That doesn’t mean big data is suitable for every question. Some simple operational questions, such as sales projections, can be answered using simple business intelligence tools and the data in the data warehouse. When assessing whether big data is appropriate, apply the three Vs:

  • Volume – is there a lot of data that needs to be analyzed?

  • Variety – is the data in different incompatible formats?

  • Velocity – is rapid access or real-time results going to be of value?

If you can answer “yes” to two out of three of the three Vs, chances are you have a big data project.

These are just a few of the common errors that VARs and others make when planning big data deployments. To succeed at big data consulting, it’s best to start small before you think big. Here are three things to consider:

  1. Every enterprise has a data management problem. You don’t have to have a petabyte or more of data for your data warehouse to be at capacity. Most organizations already have data locked away that is inaccessible, and therefore useless.

  1. Small budgets can get you started with big data. You don’t need to start with high-performance in-memory computing systems and petabytes of data storage to get value from big data. Hadoop development can yield insight on smaller platforms, especially if you need a proof of concept.

  1. You don’t need data scientists to get value from big data. Start small and the big data returns will more than pay for a data scientist’s salary.

Start with a manageable use case as a viable proof of concept and your success in big data consulting is assured.