Don't rely on analytics that examine the past to increase your ecommerce conversions and sales in the present. Implement a self-learning system that delivers relevant content to customers in real time.
Everyone in the ecommerce industry lives and dies by their analytics, and rightfully so. The information is extremely valuable when creating marketing strategies and making critical business decisions. And while such tools can tell you the amount of traffic and conversions realized for a certain period, often that information is provided too late for companies to reach out to individuals during their actual visit.
Trying to correlate real-time needs from time-late analytics, while commonplace, can also prove to be detrimental to the success of companies relying on the internet to drive revenue. Here are some examples of how leveraging historical site traffic data for current outreach efforts can mislead both marketers and customers alike:
Myth 1: Previous visits indicate current interests
Many sites offer frequent visitors the chance to receive special offers and information in exchange for providing personal information and allowing the company to track previous purchases. When the individual returns to the site, the e-tailer will promote certain items based on their historic and personal information in hopes of closing a repeat sale or upselling that individual. Problems often occur with this assumption. The customer may have been shopping for a Mother's Day gift on their most recent visit, but is now seeking something for a friend's birthday. Neither purchase has anything to do with the other. So while the e-tailer may think they are providing valuable content, they are only frustrating the individual.
Myth 2: Keyword frequencies provide a solid indication of relevance
While keyword search frequencies are a good start, they are by no means exact in their correlation to relevant and meaningful content for individuals. More important than its rate of recurrence is the relationship between keyword searches and the responses generated. For example, many people who search for laptop cases end up purchasing other items, such as a backpack or other travel accessories.
Myth 3: Identifying similar users can deliver more valuable content to the visitor
The general assumptions made from this practice fail to identify current interest or even overall preferences. What's more, this technique can eliminate many ratings and reviews that are more pertinent to the visitor's current needs.
Consumers can be fickle, and for good reason. Ecommerce companies need technology systems that are flexible and dynamic enough to keep up with what they seek today as much as what they did yesterday.
Platforms that focus on the now
Companies should, therefore, consider complementing their existing analytics programs with personalization solutions that examine the current traffic patterns and clickstream data of each individual visitor in real time, in order to provide the most relevant and personalized web experience possible. Such tools deliver more pertinent product, content and promotional offers to multiple channels and don't require any personal information, third-party cookies or special interaction with the individual.
This process is also the most effective way to deliver relevant content because consumers' present behaviors on the site are the most accurate representation of their interests and needs at that time.
As a result, these self-optimizing personalization solutions make it possible to enhance a site's product cross-sell/ up-sell merchandising and content targeting capabilities while preserving the visitor's right to privacy. Such platforms also provide an enhanced web experience for the customer by not requiring them to fill out annoying pop-up surveys or lengthy, intrusive online registration profiles. Additionally, the more advanced systems do not track visitors' behavior from site to site.
Target and uplift realized
Case in point -- a leading American catalog merchant and online retailer implemented a third-party personalization solution that focused on the traffic patterns and clickstream behavior of visitors just prior to the 2007 holiday season. In doing so, the company's ecommerce site has witnessed a more than 15 percent increase in converted online visitors and a more than 18 percent rise in direct revenue contributions from automated product recommendations and merchandising.
The system used neural network technology to identify linkages between behavior patterns and content selections. This enabled the platform to automatically and autonomously extract the attributes that are relevant to the goals of the application domain. Because it eliminated the need for expert knowledge to define and extract model attributes, the system readily adapted across different catalogs, languages, cultures and media types. It also complemented the merchant's mainstream analytic data, and thus provided higher click-through rates on content as well as increased conversions, site stickiness and repeat purchases by generating relevant product recommendations and personalized promotions based on current surfing habits.
Other retailers have claimed that such systems account for 20 percent of their online revenues, with the average order value from visitors who acted upon the automated recommendations being over 60 percent greater than those who didn't. Such testimonials speak clearly to the value that these self-optimizing solutions that leverage current traffic patterns and click stream data bring to an organization's bottom line.
Next page >>