We hear a lot about Big Data and business data analysis. What’s less clear is the strategic impact of such business data analysis: real companies, getting real benefits, from real business data analysis projects. The good news: such examples do exist.
In aggregate, for instance, it’s difficult to argue with the results of a study carried out by the Massachusetts Institute of Technology’s MIT Centre for Digital Business, in conjunction with management consulting firm McKinsey.
Looking to see if data driven companies were better performers, it analysed the inner workings of some 330 businesses, and reached a remarkable conclusion about the impact of business data analysis.
In short, companies in the top third of their industry in terms of their use of business data analysis and data driven decision-making turned out to be, on average, 5% more productive and 6% more profitable than their competitors.
So what does business data analysis and data driven decision-making look like in practice? Let’s take a look.
Business data analysis: market basket analysis.
What products do consumers tend to buy alongside other products? From what you know of the behaviour of this group of consumers, what can you predict about the behaviour of that group of consumers? If someone likes products X and Y, will they also like product Z?
Market basket analysis is the technical term for such business data analysis. And, using open source tools such as R, or proprietary analytics packages, market basket analysis is fairly straightforward to carry out—once, of course, you have captured your raw data and encoded it in a data warehouse, and preferably a Cloud-based data warehouse.
Why Cloud-based, in particular? Because Cloud data warehouses are the most cost-efficient and effective way of dealing with the volumes of data involved. American retailer Walmart, for instance, is estimated to collect more than 2.5 petabytes of data every hour from its customer transactions.
And the benefits of carrying out market basket analysis exercises? Look no further than retailing giant Amazon. Although the retailer itself is officially tight-lipped on the subject, informed sources have estimated that up to 20% of Amazon’s sales come from its recommendation engine—a recommendation engine directly powered by business data analysis.
Business data analysis: better forecasting.
Business data analysis can also make a huge difference when it comes to improved forecasting.
Fairly obviously, many companies have huge amounts of money tied up in inventory—not just retailers, but manufacturing companies and wholesalers, as well.
And being able to better predict the demand for that inventory has a significant impact on reordering decisions, both in terms of replenishment quantities and replenishment timing. Moreover, it helps to ensure better on-shelf availability, so that when a customer walks in and wants to buy an item, it’s actually there, and capable of being sold.
Which is a problem of especial importance where those inventories are of either perishable goods, or goods with a finite demand lifespan, such as fashion clothing.
Can business data analysis help? Just ask iconic British retailer Marks & Spencer, where a business data analysis application has cut the forecasting cycle down from ten weeks to two weeks—delivering a step change in accuracy and responsiveness.
Business data analysis: profiting from outlier events
Finally, business data analysis can help companies respond to—or capitalise on—out of the ordinary events.
When retailer American retailer Walmart first began utilising business data analysis to influence its stocking decisions, executives soon wondered if business data analysis tools could help it to better respond to the periodic hurricanes that batter America’s Eastern seaboard.
Ahead of such hurricanes, retail outlets routinely run out of certain items, as consumers prepare for the worst. Could business data analysis help the chain to anticipate—and plan for—such events?
The answer was proved conclusively in 2004, when Walmart’s data analysts set to work to forecast what might happen ahead of Hurricane Frances, based on what had happened during Hurricane Charley, a few weeks before.
The strong demand for torches and batteries might have been guessed at. But a seven-fold increase in sales of strawberry poptarts, and an even stronger demand for beer? “We just didn’t know that,” summed up one Walmart executive.
Business data analysis: first steps
So there we have it—three real-life businesses, three real-life applications of business data analysis.
And, in each case, one unifying factor: the companies in question actually took the plunge into business data analysis, and made a start.
To make a start with business data analysis yourself, download our free eBook below