It’s no secret that many businesses have high hopes for using big data. With few exceptions, big data analytics has been subject to some massive hype. The promises have been many, from greater capabilities, more efficient operations, better understanding of customers, new product ideas, and so much more. Big data can certainly deliver on all that, but recent surveys and studies have found that living up to that hype has proved challenging. It’s mainly a case of organizations not seeing the type of returns they hoped for. In other words, the payoffs aren’t showing up as expected. For obvious reasons, this has been the cause for concern among businesses as the amount of investment being spent on big data analytics continues to grow. Getting to the root of why they’re not seeing those payoffs then becomes a priority.
In one report from Mu Sigma, it was shown that many executives have become dissatisfied with the results they’re having in the analytics realm. They’ve put a lot of resources and effort into making big data analytics lead to a substantial return on investment, only to see the payoffs come up lacking. There’s no single reason for this failure, but a number of factors have appeared to offer good explanations for payoff woes. Part of the problem stems from placing so much emphasis on the technology being used for analytics rather than the role that decision-making plays in the process. The technology plays a pivotal role, no doubt, but a failure to understand how to properly use it to achieve business goals means much of that technology is going to waste.
Some businesses were quick to embrace big data in the early days when many organizations were still skeptical. Investors at the time had high expectations, but it’s possible that those expectations missed the mark. It wasn’t that they were wrong to have certain milestones and goals in mind, it was more a problem of underestimating the challenges businesses would face. They may have even completely overlooked some possible issues altogether. It all comes down to analytics performance. They may have all the right technology on hand, but it analytics is performed in the wrong way, the results will reflect that.
Perhaps one of the biggest and most pressing issues is the challenge of gathering good high-quality data. Businesses largely understand the need to collect lots of data to find hidden insights, but this idea has mostly been interpreted as gathering as much data as possible. This opens the possibility of collecting data that is of poor quality. If businesses don’t know how to differentiate between the contrasting qualities, it’s unlikely they’ll get the results they were hoping for.
Some organizations are also having difficulty delineating who within the company should be responsible for analytics efforts. Some place that responsibility with CIOs, others with marketing executives, and others create new positions like Chief Data Officer to handle the new duties associated with big data. While there’s no single method that necessarily works the best for all companies, much of the problem stems from inconsistency when applying analytics methodology. One portion of the business may be performing analytics in their own way, while another department may be doing it differently.
Much of what these challenges call for is greater collaboration within an organization. It’s not enough to adopt the best technologies for handling big data, like Hadoop or a flash storage array; they also need to swap techniques and communicate over what methods work best for the particular problems they’re facing. It also requires having data experts on staff who know how best to utilize analytics to solve specific business problems. Without that expertise, organizations will only be stumbling around in the dark.
With these problems and challenges, it’s little wonder why companies aren’t seeing the payoffs they were expecting with big data. But just because they’ve become frustrated by the lack of results doesn’t mean giving up should be considered. Optimism surrounding big data still abounds, and the potential is definitely there. With enough learning experiences, organizations will be much more prepared to handle big data challenges, leading to the results that will mean big things in the future.