Headline FeedsE-Commerce Times Talkback
|
|
See Full StoryIn today's difficult business climate, companies that sell their wares
online have found themselves in a race to boost the effectiveness of
their Web sites. Toward that goal, these enterprises increasingly are
employing Web analytics -- a technology that monitors and reports
site use patterns. But what is the best way to use data gleaned from
such analysis to improve customer loyalty -- and, by extension, sales?
The 80/20 Rule - From 20% comes the reward!
You receive 80 percent of your return from 20 percent of your activities. Apply this principle to your agency, it makes sense to focus your web marketing efforts on this elite group. Identify your most profitable group of customers. First, rank your accounts by the revenue. Next, prioritize customers toward categories that deal with relationships and resources. Now, see which customers fall into both the "high revenue" and "exceptional" categories. Those are your most profitable customers based on revenue, relationship and resources. From here, you can develop a profitable customer profile based on the traits these customers share. You'll also want to find more opportunities for selling to the existing top-tier customers and moving as many middle-category accounts as possible up the scale.
Don Williamson
You receive 80 percent of your return from 20 percent of your activities. Apply this principle to your agency, it makes sense to focus your web marketing efforts on this elite group. Identify your most profitable group of customers. First, rank your accounts by the revenue. Next, prioritize customers toward categories that deal with relationships and resources. Now, see which customers fall into both the "high revenue" and "exceptional" categories. Those are your most profitable customers based on revenue, relationship and resources. From here, you can develop a profitable customer profile based on the traits these customers share. You'll also want to find more opportunities for selling to the existing top-tier customers and moving as many middle-category accounts as possible up the scale.
Don Williamson
Hi there,
After working in the field of Web Analytics for 3 years now, I've come to realise that what is explained above is not an exact science.
Analysing customer activity is not possible unless considerations are made when a web site is being written - and I've yet to see a web site where this is the case. Too often the emphasis is on getting something in front of the customer rather than what are we going to do when it is there. The major disadvantage of web logs is that they can only track the delivery of http components (e.g. HTML pages, JSPs, GIFs) but they do not by default track site content. e.g. You may know that someone has bought something because they went to "thanks_for_buying.jsp" but what did they buy? And who are they anyway? Tracking activity over time per individual people is crucial to any successful e-commerce venture.
The only way that this can possibly be done with web logs is to analyse the URI query strings in URLs, and even then it is difficult.
A far better option is to utilise application servers that are designed for the purpose. Application servers have the ability to track specific events on the site (e.g. A particular person buying a particular product). There are products out there that do this and in turn provide the capability for one-to-one on-line marketing (e.g. BroadVision). With a successful CRM strategy these products can be leveraged to provide valuable analytical information about a company's sales and customers.
Web logs are only part of the picture. In my experience web logs are more useful for analysing site performance (e.g. Time for server response, Server Cluster Load Balancing) than customer activity.
Any comments? I would be very interested in anything you may have to say.
Thanks for reading.
Regards
Jamie Thomson
jamie.thomson@int21.com
After working in the field of Web Analytics for 3 years now, I've come to realise that what is explained above is not an exact science.
Analysing customer activity is not possible unless considerations are made when a web site is being written - and I've yet to see a web site where this is the case. Too often the emphasis is on getting something in front of the customer rather than what are we going to do when it is there. The major disadvantage of web logs is that they can only track the delivery of http components (e.g. HTML pages, JSPs, GIFs) but they do not by default track site content. e.g. You may know that someone has bought something because they went to "thanks_for_buying.jsp" but what did they buy? And who are they anyway? Tracking activity over time per individual people is crucial to any successful e-commerce venture.
The only way that this can possibly be done with web logs is to analyse the URI query strings in URLs, and even then it is difficult.
A far better option is to utilise application servers that are designed for the purpose. Application servers have the ability to track specific events on the site (e.g. A particular person buying a particular product). There are products out there that do this and in turn provide the capability for one-to-one on-line marketing (e.g. BroadVision). With a successful CRM strategy these products can be leveraged to provide valuable analytical information about a company's sales and customers.
Web logs are only part of the picture. In my experience web logs are more useful for analysing site performance (e.g. Time for server response, Server Cluster Load Balancing) than customer activity.
Any comments? I would be very interested in anything you may have to say.
Thanks for reading.
Regards
Jamie Thomson
jamie.thomson@int21.com
Jamie, I agree with your assessment. Analyzing web logs only yields information about anonymous behaviors without any knowledge of pre or post conditions. Web log data has to be analyzed within the context of a properly design research endeavor.
Best wishes,
Dave
Dave Wagner, President / Principal Consultant
Website Paramedics -- Internet Marketing Management & Effectiveness
Email: dave@websiteparamedics.com
Visit: http://www.websiteparamedics.com
Join: http://groups.yahoo.com/group/webmarketingeffectiveness
Best wishes,
Dave
Dave Wagner, President / Principal Consultant
Website Paramedics -- Internet Marketing Management & Effectiveness
Email: dave@websiteparamedics.com
Visit: http://www.websiteparamedics.com
Join: http://groups.yahoo.com/group/webmarketingeffectiveness
