You say 'machine learning', employees just hear 'Terminator'

Watchfinder IT director explains how to ease your business into data science

Luxury used watch retailer Watchfinder explained how to bridge the thought gap between "machine learning" and Terminator when trying to sell data analytics techniques to C-level executives.

Speaking at a Computing roundtable focusing on retail data analytics, IT director Jonathan Gill explained how he's attempting to guide the company - which is experiencing a 50 per cent year on year growth in its sector - into machine learning-led analytics.

Gill and his team have been experimenting with the Machine Learning service on Microsoft's Azure Cloud platform along with Office 365's Power BI offering.

Gill reasoned that getting to grips with machine-assisted data crunching will not only allow ongoing experimentation, but also save time.

"I'd rather that than have three senior managers come in and say, ‘Hey, I've had a thought, can I have all this data?' and then we've got to go and churn out a load of data rather than making new features or expanding things," explained Gill.

But by providing non-technical employees with IT-assisted data interrogation methods, Gill believes it will be possible to "just say, ‘Here you go - go and do some experiments and tell us what you think' and I think that's sort of the way we need to go. So rather than us providing the data we just provide the raw, and give them the tools to interrogate it."

In the meantime, Gill has identified a problem.

"If you say to people ‘there's this whole machine learning thing' all they're thinking is Terminator. As soon as you say ‘artificial intelligence' they say ‘I don't want a computer making decisions for me'.

This is why Watchfinder has decided to begin the process by bringing in a consutant data scientist.

"You're looking at a six-to-10 week consulting period, which is obviously going to have a cost attached to it, but if you can get the value out the other side to increase the bottom line, it's money well spent.

"I've pushed quite a lot for the last five years to say that the data tells the truth. What you feel is happening can generally be proved," he said.

Gill said that, when the consulting period is over, the company will consider whether to take on data scientists as permanent members of staff.