Quantivo Targets Marketing Analytics At Web ScaleQuantivo Targets Marketing Analytics At Web Scale

Big Data Talent War: 10 Analytics Job Trends

You've probably heard Gartner's prediction that marketing spending on technology at many companies will surpass IT spending by 2017. Consider, too, the buzz around cloud-based applications and big data analytics. These are three of the hottest trends in enterprise technology.

Enter Quantivo, a cloud-based analytics vendor that specializes in Web data analyses--particularly marketing analyses. Quantivo's platform is used for a variety of data sources, but the company's sweet spot is delving into high-volume Web data from Web analytics systems like Adobe Omniture, Google Analytics, and WebTrends.

"The big Web analytics packages can do summarizations and give you the number of visits at the page level, but we're helping our customers develop profiles of customer behavior," Quantivo's CTO, Paul O'Leary, told information. "They're looking to do things like develop seasonal-buyer profiles, and then take those segments and figure out what else they buy and how often they buy."

Quantivo customer Manheim Auto Auctions studies segments of customers made up of wholesale car buyers (dealers and distributors) who frequent online sales events. Manheim needs to know how they're responding and how they're likely to respond. The company uses WebTrends for general-purpose Web analytics, but until it implemented Quantivo last year, it couldn't get the total picture of online activity.

[ Want more on SaaS-based BI and analytics? Read 12 Brawny Business Intelligence Products For SMBs. ]

"We had the transactional records if somebody bought or bid online, but we had no visibility into the Web activity data that didn't result in a transaction," explained Stephanie Straeter, Manheim's director of business intelligence, in an interview with information. So Manheim couldn't tell if a dealer almost bid, but then didn't. If the bid-confirm page served up errors--a common occurrence in the complex and fraud-prone auto business--Manheim didn't know whether, say, 20 errors were generated by 20 different customers or by one fraudster repeatedly attempting one bad transaction.

Embraced by Manheim last summer in a four-week implementation project, Quantivo now gives the firm insight down to granular segments and individual registered users.

"We can tell customers who is looking at their vehicles, so if Ford or Mercedes has an online sales event, we can say which franchisees visited the event, who bought, and who didn't," Straeter said.

If a segment responded or didn't respond to a particular offer or product line, Quantivo helps Manheim understand other actions those customers might take. "You just select the segments and variables and press a 'likelihood' button in Quantivo," Straeter said.

Manheim has a conventional, on-premises data warehouse for the more than 122 physical-location auctions it holds throughout the world (including more than 70 events in the U.S.). But adding Web clickstreams to that warehouse for deeper analysis was out of the question due to the sheer volume and complexity of the data.

"Our website generates more data in one month than we currently have in the entire data warehouse, plus it's in a Weblog format that's a big string of data, we'd have to build rules on how to add it up, slice and dice it," Straeter explained. Quantivo has prebuilt data-loading and transformation routines set up for the leading Web analytics packages.

Quantivo's platform is also ready to scale, according to CTO O'Leary. Clickstream analysis is not typically a petabyte-scale problem, he admits, but O'Leary says even small-but-active sites can generate hundreds of millions of records per month. Companies tap the SaaS provider when they see that the volume of data, the rate at which they're ingesting data, the turnaround time, the complexity of the analysis or all of these challenges combined are more than they care to handle on their own, he said.

"The beauty of this service is that I don't really have to know anything about the technical requirements behind the scenes," Straeter said. "We did it outside our IT group because they had their hands full with other projects."

Larger companies with deeper staffs and investments in analytics packages from the likes of IBM SPSS and SAS can muster deeper and more predictive analytics in house, but O'Leary says even departments of such companies take the software-as-a-service (SaaS) route with Quantivo because they need fast deployments and don't want to engage with IT and pay for a share of the data warehouse.

Quantivo's competitors include Birst, GoodData, and PivotLink among general-purpose-BI SaaS providers while ClickFox is more of a direct competitor.

Quantivo's services start at $2,000 per month per data source, with up to 10 million new records per month, 50 million records in total, 20 users, and up to three schema changes per year. Additional data sources, schema changes, and daily, rather than weekly, data loads entail additional fees.

Every company needs a social networking policy, but don't stifle creativity and productivity with too much formality. Also in the debut, all-digital Social Media For Grownups issue of The BrainYard: The proper tools help in setting social networking policy for your company and ensure that you'll be able to follow through. (Free with registration.)