"Big Data" this word can be said to have been completely "cloud computing" of the wind covered down, and now a variety of industry conferences as if not to mention the big data can not keep up with the times. Gartner has recently reported that, although the world's major business users have strengthened the big data (Big Data) business investment, but 60% of the enterprises on the big data investment returns to question. What makes companies skeptical of this technology that they're flocking to?

60% say it's too early to talk about returns

Gartner reports that big data investments by the world's largest enterprises totaled $4.3 billion in 2012, with the vast majority of those investments targeting software running on company servers. This total investment is expected to increase to $34 billion during 2013.

But while organizations are investing more in big data, most are not confident that these investments will pay off. In a survey of more than 800 business and IT executives, 60 percent of respondents said it is too early to tell if big data investments will deliver a good return.

Behind the big data gloss

What is big data? There is no standard definition of "big data" until now. Someone on Wikipedia described Big Data as follows: data is growing so fast that it is difficult to navigate using existing database management tools, and difficulties exist in acquiring, storing, searching, ****enjoying, analyzing, and visualizing data.

As an inevitable trend of future development, there is no doubt that big data has an extremely far-reaching significance for enterprises. In the past two years, including IBM, HP and other storage vendors in pursuit of the concept of "big data", they proposed in addition to providing customers with basic storage solutions, but also to promote a series of "big data" to the enterprise analysis solutions, tap the value behind the data. The concept of "big data" has been proposed by them.

But while various articles have been written describing the bright future of big data, there have been few reports on the actual results of big data project implementation.

The first thing we have to realize about the rate of return that companies are questioning is that the "rate of return" is not obvious in some industries. In financial services, big data can lead to better and more efficient service, leading to more favorable business strategies. Media companies can sell more advertising space. E-commerce companies can sell more products.

But these companies have a ****similarity that the average entrepreneurial company doesn't have: the ROI is clear enough for these companies to remove the barriers to entry into the Big Data space. And is big data attractive enough for most organizations? Most likely not. The value of Big Data would have to be very high, cheap and mature enough to entice organizations to buy it.

How to put big data to the end

Some industry insiders point out that there are two main constraints on the development of big data: first, the technology that can unearth big data is not yet mature; second, the cost is too high. When doing big data, storage should be very cheap, although storage is much cheaper than many years ago, but still very high.

The rapid growth of unstructured data has increased the difficulty of data processing. At the same time, many companies are still in the research and development phase of big data. And as a result, a lot of uncertainty has been added within many organizations. Big data technology must be made easier and project management skills more widely available before big data can truly become mainstream.

From a specific technical point of view, the return on investment in data is the value of the data divided by the cost of the data, first of all, we need to reduce the cost of the data to improve the value of the data. There are many ways to reduce the cost of data, the most important is to transfer low activity data to low-cost storage. And increase the value of data to collect more and more comprehensive data, the recent hot socialization software can play a role here. Second, have a data governance team and process for data quality. Finally, to have a good ability to analyze data, "data visualization" is the current trend.