The wealth of data available from online ad campaigns is creating a dilemma for advertisers: How do they make sense out of all the information, from user behavior to number of exposures of an ad?
The latest upgrade of ad serving software by Atlas DMT, an operating unit of interactive agency Avenue A Inc., is designed to solve this problem with an analytics profiler that gives marketers deeper insight into data collected from online campaigns.
The Atlas Digital Marketing Suite 3.0, released at the end of June, features a new architecture called Web Analysis Results Profiler, or WARP. Based on a proprietary programming language and making use of distributed computing, WARP processes large amounts of data and returns reports that let marketers drill even deeper into the results of their interactive advertising programs.
One of many analysts
Atlas DMT is one of many companies that provide analysis of Web advertising campaigns. Virtually all ad serving companies, including DoubleClick Inc. and 24/7 Real Media Inc., provide detailed reports and analysis of ads served.
There is also a growing number of technology companies, such as Poindexter Systems Inc., New York, and Paramark Inc., Sunnyvale, Calif., that provide real-time analysis of ads served, said Rudy Grahn, analyst at Jupiter Media Metrix Inc.
"The Atlas tool is pretty robust in terms of tracking ad campaigns," Grahn said. "This sort of application is probably aimed at the top end of Internet marketers."
However, Grahn said, "Iâm not convinced online advertising as an industry has matured to the point where every advertiser will be ready for granular information. Most are swinging and missing at the macro stuff."
Meaning within the glut
Joe Redden, director of product management for Atlas DMT, Seattle, said the new product was developed to address a glut in information.
"One problem we ran into â¦ is the wealth of data can get so large itâs hard to provide meaningful data and analyze results," Redden said.
Earlier versions of Atlas DMT, first released in April 2001, also analyzed customer data collected through online advertising but couldnât provide the deep analysis possible with the new version.
Redden said the software is distinguished by three new features. The first is a query-based interface that lets marketers ask questions about the performance of ad campaigns in simple language, which is then translated into programming language.
For example, a marketer might be interested in learning how many people visiting a Web site had purchased a product at least three times and how many ads they had been exposed to at the site. The marketer could then adjust ad levels accordingly.
Second, the software uses distributed computing, which means several computers handle requests. Distributed computing is used for heavy-duty analysis involving large quantities of data.