Little wins key to big data success

Big data remains one of the major forces in technology in 2014, rich in potential with insight to allow informed business decision making. Yet it does have its critics – people have questioned the level of ROI it delivers and also whether it’s as suitable for smaller organisations as it can be for bigger enterprises.

We believe this is as much a perception issue as anything. There are many grey areas around ROI, but perhaps part of the problem with big data ROI is that people are expecting game-changing returns from it, right from the off. Is there not just as much value from smaller, incremental wins? 

Big data ROI
When an organisation invests in new technology of any kind, there is an inevitable rush to see an immediate return on that investment. Big data is no different. Hadoop, data scientists, the computing power required to manage large volumes of data….these are all reckoned to be needed in order to deploy big data. We’ve written before about the costs associated with big data and it’s fair to say that CFOs want to know what they are getting for their money.

This means the people that implement big data projects are under immediate pressure to justify the investment. But expecting a significant return that soon is unrealistic. Perhaps over time it will deliver that jaw-dropping ROI, but in reality a number of smaller wins can be just as valuable and are far more realistic.

Perhaps it has been the way big data has been marketed and presented, but it is rare to see major, ground-breaking initiatives as a result of analytics. It does allow for many smaller enhancements or improvements and provided people accept that fact, it is still a worthy and valuable demonstration of big data ROI.

Big data, SMEs and bare metal
Having looked at the ‘R’ of ROI, let’s turn now to the ‘I’. Big data does not have to mean a major and significant investment, and is in fact as available to start-ups and SMEs as it is bigger businesses.

Even for a major enterprise, buying a dedicated Hadoop infrastructure outright doesn’t necessarily make sense. It would cost hundreds of thousands of pounds just to deploy a farm of 20 servers; then there’s the cost of acquiring the right technical skills, supporting the systems and keeping the technology current. That’s clearly unrealistic for most.

Rather than compromise with a ‘make do’ virtualised Hadoop scenario in the cloud, a start-up with big-data requirements can easily buy affordable, fit-for-purpose bare-metal services. This allows them to power up a dedicated infrastructure as needed, and only pay for what they actually use.

This means they can get started with big data in a sustainable, low-risk way, requiring much less initial investment. Big data has massive potential and if organisations can focus on smaller incremental wins and realise that major investment isn’t required, then there is no reason that ROI can’t be demonstrated.