There’s still a long way to go before the average end user can get business benefit out of big data analytics, data solutions and services firm MTI has warned Computing.
Speaking to us this week, the firm’s Isilon business unit manager, Andrew Oliver, said there remained a big gap between what vendors say analytics can deliver and what users actually experience.
“I have my doubts about how many people are deriving business benefit from analytics data so far,” said Oliver.
“I think we’re only now moving into a space where people can get something useful out of data they are just now starting to collect.”
Computing research would seem to support Oliver’s view. In a recent study of over 500 UK organisations, only 13 per cent of respondents reported that they are deriving “true business value” from big data. A further 19 per cent said they are now turning data into “useful information” having run proofs of concept first.
“A big area is telemetry data – taking it from machines to establish trends, faults or problems as they develop. It’s great in theory, but not really happening yet,” said Oliver.
Oliver considers use case examples “bandied around” by vendors as “fairly obscure” so far.
“I’ve spent a lot of time working in central government and defence areas, and if someone turned round and said ‘We discovered a fault in a fighter jet and discovered that in production and fixed it, and it saved millions and millions of pounds – that would be a good use case, but the examples we hear are never as interesting or candid as that,” he said.
Oliver believes, like many, that the data science skills gap is preventing organisations from deriving true value and actionable insights from big data.
“You always have a disconnect between the technology people and those who do need to run queries against data, and end users, who understand the problems and know what they want to solve. But the bridge between those is missing,” said Oliver.
“Data scientists are key to this. Without them the whole thing is not going to work. But I guess what we’re trying to do is engage more than just IT and infrastructure people in how to build physical layers – we’re trying to move the conversation on so we talk to the people who actually have to use it, too.”
MTI is trying to create more of a “flow”, presenting a complete solution rather than “one component”, said Oliver. It removes siloed work practices, and means data scientists have fewer barriers to break down when attempting to talk across the business.
“Rather than just be a vendor, I see this as adding a more consultative approach, and trying to engage people at a ‘What is your problem?’ level rather than just providing the platform. Without that, they’re not going to derive all they can from the investment,” said Oliver.
“There’s a lot to talk about with analytics – using, capturing and storing the type of data that has traditionally been discarded because it would be too expensive to capture and store it, but now people are just starting to do that,” said Oliver.
“There are a number of initiatives, a number of people are talking about it and it resonates with most of the customers we speak to.”