Big data and retail: saviour or sham? | Technology | theguardian.com

Big data and retail: saviour or sham? | Technology | theguardian.com

It seems like there isn’t a day that goes by without yet another report proclaiming big data as an absolute necessity for retailers and small-to-medium business (SMB) owners, but these claims risk ignoring both the existing challenges of the industry, as well as the bigger picture.

In fact, some analysts go as far as to claim that big data is nothing short of a sham for most retailers and SMBs.

In retail, big data generally refers to the use of sets of information from inside and outside a company that can then be analysed and used to improve profits. These data sets include everything a customer – or potential customer – puts into the internet about a particular brand or about how they consume, and all the information a company collects internally relating to sales patterns and customer relationships.

A March report from the McKinsey Quarterly, a report from IBM and Planet Retail and a Bart Baesens’ 256-page epic on Analytics in a Big Data World all claim that retailers and small-to-medium business owners need big data to be successful in the digital economy.

However, Matt Young, founder and principal of Sydney-based consulting company Turning Leaf told Guardian Australia that SMBs were being sold a pup.

“I work with businesses that are being tempted by people promising big data solutions,” he said. “For most SMBs, big data is a sham.”

The challenge for retailers and small business owners isn’t big data, Young said, but how to manage lots of little data.

“It is much better to stick to timeless basics than rely on some consultant and his ‘black box’ magic,” he said.

Matt Peterson, founder of Shiny Things, a company that makes education apps for the iPhone and iPad, said he deliberately chose not to collect data on the 1.3 million customers that have downloaded his education apps since 2012.

“The first reason is that it’s a big responsibility,” he said. “We are focused on the education market and we can’t afford to make a mistake with our customers’ data. These are kids we’re dealing with.

“The second is that it requires big resources for collecting and securing data as well as gaining insight into it, that requires huge resources and very separate areas of expertise. We have 12 people in our office, so a big data program or project is not something we can afford today.”

Peterson said it was easy for SMBs to get hooked on the idea that data tells them everything, “when really, if you’re talking to customers face-to-face as regularly as you can, that is probably enough to give you a successful product”.

Peterson said that he had not seen big data being used effectively by SMBs.

“There are so many signals you can be distracted by,” he said. “More often than not you’re getting the right answer to the wrong question”.

“The truth is we just don’t know how to deal with information yet,” he said.

Even Facebook, arguably the king of big data for marketing purposes, isn’t getting it right. Peterson said that the social network continued to deliver ads for wedding dresses to his wife even after she was married.

If Facebook can’t get it right, the chances of retailers achieving success are pretty low, he said.

“At the end of the day you could have invested a cool half million collecting all this data and realise it’s useless and now your business is going nowhere.”

On the other hand, Milan Direct, an online furniture store, has seen a large improvement in sales thanks in part to its big data solutions.

Dean Ramler, the company’s CEO and founder, said that big data enabled the company to make decisions based on facts, not on “gut feel” by looking at what people were searching for online using free tools, such as Google Trends.

“A couple of years ago we saw a real peak in search patterns for timber outdoor furniture so we added a line of outdoor timber furniture to our line based on that hit from day one,” he said.

Ramler wasn’t afraid to admit he used big data to see what other high-end brands were making so he couple replicate it and sell top quality replicas on the site.

“We use a lot of internal data,” he said. “Our big data solutions allow us to hold as little stock as possible while fulfilling orders as they come through. So we’re not constantly number crunching. If a product doesn’t sell fast enough we ditch it.”

However Ramler said that for most retailers and SMBs, big data was more of a threat than a solution.

“I don’t think these big data solutions are relevant for all businesses,” he said. “It is especially irrelevant for small business.”

Ramler said big data should be a solution a business grows into, rather than something it starts with.

There’s even scepticism at the larger end of the retail market. Coles corporate affairs spokeswoman Anna Kelly said that while big data could play a part in retailers’ decision-making, it was just one of many tools.

Kelly wouldn’t be drawn on exactly how the supermarket uses big data, but said Coles’ Flybuys programme was an example of where big data comes into play.

“Flybuys uses the information provided to operate the program to deliver rewards and offers, such as bonus points and My Weekly Specials,” she said.

“No matter the scale of the business, it is important to understand your customer,” she said. “Big data may play a part, but it is one tool of many.

Craig Reardon, from web-design company The E Team, said that the vast bulk of SMB operators didn’t know about big data.

“All but the most tech-savvy SMB operators see developments like this as a threat rather than opportunity because of the time (cost) involved,” he said. “Time they just don’t have.”

“You have to remember that this is a sector that – with a reported uptake of up to 50% – still hasn’t fully embraced ordinary websites, let alone big data.”

Sponsored content

This content has been sponsored by IBM, whose brand it displays. All content is editorially independent.

Read this article:

Big data and retail: saviour or sham? | Technology | theguardian.com

Share this post