December 29, 2009, 12:00 AM

Panning for Gold

(Page 2 of 2)

While it’s easy for retailers to feel overwhelmed by the sheer volume of analytic data available to them, generally they can improve results by focusing their attention on measures of traffic quality rather than simply measures of traffic volume, experts say. Knoch of Adobe Omniture, for example, has worked with retailer clients on very specific quality metrics, such as average revenue per visit off various keywords.

Since different keywords-“warm sweater” or “wool sweater,” for example-could drive searchers to the same product page, it behooves retailers to know which one of the terms converts best, to validate that they’re spending their time on optimizing that right keywords and to guide search engine optimization efforts going forward.

Another metric that offers insights on how well a keyword and landing page perform together is bounce rate, the rate at which visitors view one page of a site and then leave. The free Google Analytics package offers bounce rate reports, making it easy for retailers to determine the efficacy of their search engine optimization work, says Eric Peterson, CEO and founder of consultancy Web Analytics Demystified Inc. “Essentially, the higher the bounce rate for words and pages, the worse a job your SEO efforts are doing in driving qualified traffic,” Peterson says.

Measuring bounce rate revealed a few surprises to The retailer found a high bounce rate on the landing page for grommets-rings that reinforce holes, such as those that computer and cable wires are threaded through-that it had optimized for natural search. Shoppers were landing on the page from natural search but often leaving quickly.

That landing page consisted of a results page generated dynamically by CableOrganizer’s internal search engine, and the high bounce rate suggested the page wasn’t effective. tested the idea that the landing page wasn’t adequately showcasing its broad assortment of grommet products.

“After careful analysis, it appeared our internal search results did not give an immediate bird’s-eye view of our grommet line, which resulted in high bounce rate among customers looking for a wider selection,” says Mildred Munjanganja, vice president of marketing and general manager of

In light of that information, the retailer changed the landing page it optimized for “grommets,” making it the category page for that product instead of the results page generated by the site search engine. This immediately offered consumers more choices on the page. It also added “best matches” for grommets in the right-side navigation. “The result was a reduction in bounce rate,” Munjanganja says.

Visitor engagement

Retailers looking to better qualify traffic so as to focus SEO efforts most profitably should be looking at multiple measures of visitor engagement. Generally, as served up by most advanced analytics packages, that includes separate measures such as session duration, page views per session, visits per visitor and customer satisfaction in addition to conversion.

Peterson recently developed and published as an open source project-free to any retailer that wants to try the formula-a measure of visitor engagement that rolls up these and several other parameters-such as the rate at which a visitor returns to the site over time and the visitor’s willingness to directly contribute feedback-to calculate a single metric gauging visitor engagement.

This metric can be calculated by retailers using the data capture and analysis capacities offered by several advanced analytics packages, according to Peterson. “The most important use of the visitor engagement metric is as a basis for visitor segmentation,” he says. “The ability to differentiate referring sources, campaigns, keywords, and pages by the measured level of visitor engagement opens the door to whole new opportunities for marketing.”

According to Peterson, this measure of visitor engagement could be applied to more deeply examine SEO search efforts that aren’t driving immediate online sales-perhaps 96% of search traffic, based on average retail conversion-but may yield other insights of value. Online marketers experimenting with this metric include a major big box retailer and a large travel services company, Peterson says, though he would not disclose their names.

Set in context

Though the possibilities for analyzing site traffic and visitor behavior have expanded as web analytics technology has developed, experts caution that the calculations will do little to improve search engine optimization unless they’re set in the context of defined business objectives supported by dedicated resources.

“Retailers don’t always have enough time to fully engage in their analytics, so a platform that pulls all the relevant data together into a coherent framework is key,” says Squire of Coremetrics. “But analytics alone doesn’t get you there. You have to have your business goals in place as well as decision-making systems based upon the data.”


Click Here for the Web Analytics Products & Services Guide

comments powered by Disqus




From The IR Blog


Philip Masiello / E-Commerce

5 essential e-commerce tools for consumer research

Online retailers have many ways of learning about what consumers want and how they shop, ...


Andrew Ruegger / E-Commerce

How online search data can improve offline retail results

Search data represents the largest, unbiased free source of consumer data in the world.