Big Data Reveals Weather-Related Shopping Patterns
Google BigQuery and Tableau Software's visualization tools show how weather affects shoppers' behavior, help retailers prepare for spikes in demand.
5 Big Wishes For Big Data Deployments
(click image for larger view and for slideshow)
If you're a retailer, how do you plan for the unplanned? A sudden winter snowstorm, for instance, can cause a spike in demand for batteries, canned foods and snow shovels. If you don't have enough of those items in stock, customers may grow irate and shop elsewhere -- and perhaps never come back.
Interactions Marketing, a consumer engagement marketing firm, has conducted a big data test case using point-of-sale transactional data and regional weather information data from multiple sources to gain fast insights into shopper behaviors, the company said.
The test case used Google BigQuery, a Web service for interactive analysis of very large datasets, and visual analytics tools from Tableau Software to quickly examine massive quantities of information. This combination of tools allowed Interactions to slash data-analysis times from about week to just hours or even minutes, says Giovanni DeMeo, VP of global marketing and analytics for Interactions.
"We analyzed the data to find out what patterns existed, and identify what those patterns were," DeMeo told InformationWeek in a phone interview. "We were looking for a query that would demonstrate not just the existence of patterns, but also (how) to monetize ... them for retailers and manufacturers that we work with."
The Interactions study identified repetitive weather events (e.g., snowstorms), classified them by severity, and measured their effects on sales before, during and after the events. It also tracked new sales patterns and shopper activity that potentially could help retailers and makers of consumer product goods (CPGs) plan in-store promotions before these events occur.
By using BigQuery and Tableau's visualization software to analyze "hundreds of millions" of rows of data in a very short time, Interactions found it could quickly share significant findings with its retailing and manufacturing customers.
"What we identify as significant is somewhat arbitrary, but it's based on our history of knowing how we can impact sales," said DeMeo.
Interactions' analysts uncovered which products had the most significant sales increases and decreases during specific weather events, and how shoppers' behaviors changed. They factored in a wide variety of data sources, including geographic location, time of day, day of the week and a retailer's proximity to competitors' locations.
Here's what the data analysis found: A day prior to statistically similar weather events, sales in 28 product categories jumped from 20% to 261% over the same day a year earlier.
"Anything with more than 20% or greater increase in sales, over sale-day [the] previous year, gives us the opportunity to identify those categories, and have enough time to communicate with retailers and manufacturers," DeMeo said.
This allows businesses enough time to create promotional campaigns "to educate shoppers ... even before the shoppers know they're going to need those particular items," he added.
On the flip side, the Interactions study also showed a drop in sales during the event's peak (few of us venture out during a blizzard or hurricane) and for four days after. Interestingly, these behaviors took place not only in regions that experienced the weather event, but also in areas where the event was forecast but didn't occur.
Interactions sees its weather test case as just one example of how marketers can use big data to boost sales.
Other companies are analyzing large amounts of weather-related big data for other purposes. EarthRisk Technologies, for instance, is a San Diego-based startup that makes long-term weather forecasts by taking more than 60 years of historical weather data and crunching it with 82 billion calculations.
Our survey finds that most companies will launch mobile commerce within a year. What's your holdup? Also in the new, all-digital Mobile Commerce Takes Off special issue of InformationWeek: Companies that take PCI responsibilities seriously will find that using a cloud provider and staying compliant can be a major project. (Free registration required.)
6 Tools to Protect Big DataMost IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Big Data Brings Big Security ProblemsWhy should big data be more difficult to secure? In a word, variety. But the business won’t wait to use it to predict customer behavior, find correlations across disparate data sources, predict fraud or financial risk, and more.
InformationWeek Tech Digest, Nov. 10, 2014Just 30% of respondents to our new survey say their companies are very or extremely effective at identifying critical data and analyzing it to make decisions, down from 42% in 2013. What gives?