|
|
Personalise Your Trade - That's Where the Trick Lies
Posted On November 7, 2007 by Priyadarshan Roy filed under Miscellaneous
Have you ever felt the need for a customized newspaper, where your favourite sports page is Page 1 and where politics is relegated to the last page? Or where you have important financial news finding place on the second page instead of the seventh? You're not alone if your answer is a yes. Every single person has his/her own tastes and prefers to have customized information. In fact, that's what personalization is all about.
Unfortunately, the fact that print medium is non-interactive makes it very ineffective when it comes to personalization factors. However, with the Internet boom, personalized services takes yet another dimension.
If you have ever been a registered user or consistent visitor to either Excite.com or Firefly or Amazon, you would notice changes in the look of the website after every visit. Excite.com studies a visitor using a concept called discovery learning.
In essence, what the website tries to do is to define what the surfer needs by studying the visitor's previous experiences on the web.Every month, a number of websites are jumping onto the personalization bandwagon. As a developer or an entrepreneur, it makes a lot of sense for you to personalize your website. There are a couple of constants.
First, successful per-sonalization almost al-ways requires you to pay close attention to user behaviour. And you must also consider privacy issues. In India of course, privacy issues are not that important since the legal system has a lot of loopholes and internet related legal infrastructure is still not in place.
However, if the site is global, user trust is critical because personalization generally entails the collection and use of personal data.
To help you determine if personalization is right for your site, we'll demystify the jargon, work through the basic issues, sample successfully personalized sites and tell you what software is available to help you make your site more responsive to individual users.
What is Personalization?
Website personalization is still so new that there's little agreement on terminology. Personalization (sometimes called customization) generally refers to making a website more responsive to the unique and individual needs of each user.
This can be accomplished in a number of ways, some of which require the user's active involvement (typically through filling out a form or following a decision-tree set of questions).Other approaches operate behind the scenes, without relying on user input-by using cookies, for example, or by looking at an IP address and serving up content based on the user's browser. (In some cases, web servers can even figure out geog-raphic locales by resolving IP addresses into domain names, which must be registered to a person or a company and a corresponding address. However, the address is typically that of the ISP, and that may or may not be enough to specify location.)
For users, the desired result is relevant content appearing more prominently on the site, while less relevant content may be buried in menus.
In addition to basic website personalization techniques such as rules-based filtering, new app-roaches like collaborative filtering and learning-agent, technology can further refine the process of giving each individual user exactly what he or she wants.
Rules-based Filtering
Rules-based filtering poses users a series of questions (or asks them to meet a set of criteria), and then delivers content appropriate to their responses. The criteria can be anything from making sure a user lives in a state served by the company sponsoring the site, to making sure the site offers products that meet specific needs. Examples of rules-based filtering questions include: What is your zip code? Are you male or female? Do you use a Mac or a PC? Are you a graduate?
This approach differs from basic personalization by providing users with specific content based on their responses. Users are led down a path by answering a set of yes/no or multiple-choice questions of increasing specificity and relevance.
Collaborative Filtering
Collaborative filtering (also called group filtering), meanwhile, is designed to serve relevant material to users by combining their own personal preferences with the preferences of like-minded users. An example can be a website which sells film music. Imagine User A likes O P Nayyar and Lata Mangeshkar, and User B likes S D Burman and Hemant Kumar.
The site suggests the names of all CDs purchased or referred by User B to User A, since their tastes are similar. However, to do so the site uses filtering techniques using questionnaires or web-based rating tools, which users use.
No single personalization technology or model works best: It depends on what you are trying to accomplish. Collaborative filtering and rules-based filtering are not mutually exclusive. In fact, it's often desirable to offer users the opportunity to input information, while at the same time leading them where you want them to go.
Most collaborative filtering technologies let web builders weigh certain criteria more highly than others, adding some rules-based rigidity to help control where users actually go. The bottom line is to enable one-to-one marketing-uniquely targeting specific offers and products to specific customers.
Inference Model
Yet another method is to use inference methodologies by plotting data mining techniques on the web data provided by the website. The model works by tracking various data types, and tries making inferences.The best example is of users visiting chat rooms or the FAQs page, just before making a purchase. So based on this data in his or her personalized page, the FAQ and chat links are prominent. The theory can backfire as it simply states that the user is likely to buy something after taking a look at the FAQs page of that product.
How To Plan Your Personalizing Strategy?
Like any good software development program, personalizing your website requires a strong vision, solid planning and a method of judging success. Be sure to resolve these seven critical issues:
Make a Sound Plan
What behavior are you hoping to enable? Driving sales, generating traffic or creating a knowledge base? Before considering products and implementation strategies, create a well-developed plan for what you're trying to accomplish. Figure out how you'll measure success and judge software options against these plans.
Set Ground Rules
Define your business rules; whatever your product, there is a logic to how it is marketed and sold. Sample business rules include making certain items available only to subscribers, giving some items priority shipping and suggesting item y if item x is out of stock.
Organize Your Databases
Is your data ready for use by a consumer-accessible website? Is your data organized with the proper intelligence so that products like skis, for example, are linked to bindings? In order to create this kind of data connection, you'll need to work with your marketing team to determine the relationships between products and information. Only when these issues are settled will you be able to offer visitors custom product-sets on the fly.
Get the Business Logic Right
Don't underestimate the effort needed to create the necessary logic for your personalization effort, as well as the time and learning curve required to tag the content. Make sure you have sufficient technical resources to complete the personalization effort.
Step-by-Step Method is the Best
Don't try to add customization to a large site all at once. Begin with the areas most likely to drive sales, traffic and customer loyalty, and then branch out. After determining the level of customization you'd like to enable, figure out what it will cost to build it (including hardware and support).Throughout the planning process, don't forget to engage your users for feedback in whatever way possible-random surveys with prize incentives, focus groups and interviews with potential users at targeted trade shows. Ask them whether they need this functionality and, if so, which features in particular?
If you think about all these issues before dropping the big bucks, you'll be one up on many companies.
Assessing Your Personalization Effort
What will it take to personalize your website, or some part of it? The answer depends on the expansiveness of your vision.Some software applications, like Net Perception's Grouplens, are designed to let you import existing data to seed your customization engine, and this can take anything from one week to three weeks for implementation. But if you are setting up a rules-based system, the process could take much longer. Defining the business rules and the decision tree's schema can be time-consuming if your company hasn't already collected this data. And you can count on running into some hairy snags-such as unexpected new branches in the decision tree and a company-wide absence of written documentation of its business rules.
The best personalization projects go far beyond check boxes into anticipating customers' needs. When Cybermeals first started developing its back-end engines, the technical staff ran into a major problem it didn't anticipate: meal-ordering entails a lot more information than is cont-ained on a menu.
Waiters typically nudge patrons as to the kind of dressing they'd like, or if they want sour cream or butter on their baked potato.
A relational database worked well for basic restaurant information (hours of operation, address, phone number), but a more object-oriented approach was needed to support the varied data involved in actually ordering food.
Measuring Personalization
Many web builders who have created personalized sites say the technology still faces plenty of growing pains. In some cases, users must enter too much data or spend too much time configuring the site (or worse, rating content) before they gain any benefit.
Often, the payoff to the user for all that effort remains rather unclear. However, there are some simple ways of measuring the personalization and the simplest method of all will be to measure the boost in sales and traffic.
However, there are other ways to find out. This is by finding out the amount of time users spend on the website in correlation to personalization efforts. The thumb rule is that if a user spends more time on a website, then more likely is he or she to buy something from that page.
Conclusion
In the business on the net, surfers will definitely feel different when a personal touch is provided to the mouse clicks. A web site which can provide that extra personal touch can definitely score better in the rat race on the net.
Unfortunately, the fact that print medium is non-interactive makes it very ineffective when it comes to personalization factors. However, with the Internet boom, personalized services takes yet another dimension.
If you have ever been a registered user or consistent visitor to either Excite.com or Firefly or Amazon, you would notice changes in the look of the website after every visit. Excite.com studies a visitor using a concept called discovery learning.
In essence, what the website tries to do is to define what the surfer needs by studying the visitor's previous experiences on the web.Every month, a number of websites are jumping onto the personalization bandwagon. As a developer or an entrepreneur, it makes a lot of sense for you to personalize your website. There are a couple of constants.
First, successful per-sonalization almost al-ways requires you to pay close attention to user behaviour. And you must also consider privacy issues. In India of course, privacy issues are not that important since the legal system has a lot of loopholes and internet related legal infrastructure is still not in place.
However, if the site is global, user trust is critical because personalization generally entails the collection and use of personal data.
To help you determine if personalization is right for your site, we'll demystify the jargon, work through the basic issues, sample successfully personalized sites and tell you what software is available to help you make your site more responsive to individual users.
What is Personalization?
Website personalization is still so new that there's little agreement on terminology. Personalization (sometimes called customization) generally refers to making a website more responsive to the unique and individual needs of each user.
This can be accomplished in a number of ways, some of which require the user's active involvement (typically through filling out a form or following a decision-tree set of questions).Other approaches operate behind the scenes, without relying on user input-by using cookies, for example, or by looking at an IP address and serving up content based on the user's browser. (In some cases, web servers can even figure out geog-raphic locales by resolving IP addresses into domain names, which must be registered to a person or a company and a corresponding address. However, the address is typically that of the ISP, and that may or may not be enough to specify location.)
For users, the desired result is relevant content appearing more prominently on the site, while less relevant content may be buried in menus.
In addition to basic website personalization techniques such as rules-based filtering, new app-roaches like collaborative filtering and learning-agent, technology can further refine the process of giving each individual user exactly what he or she wants.
Rules-based Filtering
Rules-based filtering poses users a series of questions (or asks them to meet a set of criteria), and then delivers content appropriate to their responses. The criteria can be anything from making sure a user lives in a state served by the company sponsoring the site, to making sure the site offers products that meet specific needs. Examples of rules-based filtering questions include: What is your zip code? Are you male or female? Do you use a Mac or a PC? Are you a graduate?
This approach differs from basic personalization by providing users with specific content based on their responses. Users are led down a path by answering a set of yes/no or multiple-choice questions of increasing specificity and relevance.
Collaborative Filtering
Collaborative filtering (also called group filtering), meanwhile, is designed to serve relevant material to users by combining their own personal preferences with the preferences of like-minded users. An example can be a website which sells film music. Imagine User A likes O P Nayyar and Lata Mangeshkar, and User B likes S D Burman and Hemant Kumar.
The site suggests the names of all CDs purchased or referred by User B to User A, since their tastes are similar. However, to do so the site uses filtering techniques using questionnaires or web-based rating tools, which users use.
No single personalization technology or model works best: It depends on what you are trying to accomplish. Collaborative filtering and rules-based filtering are not mutually exclusive. In fact, it's often desirable to offer users the opportunity to input information, while at the same time leading them where you want them to go.
Most collaborative filtering technologies let web builders weigh certain criteria more highly than others, adding some rules-based rigidity to help control where users actually go. The bottom line is to enable one-to-one marketing-uniquely targeting specific offers and products to specific customers.
Inference Model
Yet another method is to use inference methodologies by plotting data mining techniques on the web data provided by the website. The model works by tracking various data types, and tries making inferences.The best example is of users visiting chat rooms or the FAQs page, just before making a purchase. So based on this data in his or her personalized page, the FAQ and chat links are prominent. The theory can backfire as it simply states that the user is likely to buy something after taking a look at the FAQs page of that product.
How To Plan Your Personalizing Strategy?
Like any good software development program, personalizing your website requires a strong vision, solid planning and a method of judging success. Be sure to resolve these seven critical issues:
Make a Sound Plan
What behavior are you hoping to enable? Driving sales, generating traffic or creating a knowledge base? Before considering products and implementation strategies, create a well-developed plan for what you're trying to accomplish. Figure out how you'll measure success and judge software options against these plans.
Set Ground Rules
Define your business rules; whatever your product, there is a logic to how it is marketed and sold. Sample business rules include making certain items available only to subscribers, giving some items priority shipping and suggesting item y if item x is out of stock.
Organize Your Databases
Is your data ready for use by a consumer-accessible website? Is your data organized with the proper intelligence so that products like skis, for example, are linked to bindings? In order to create this kind of data connection, you'll need to work with your marketing team to determine the relationships between products and information. Only when these issues are settled will you be able to offer visitors custom product-sets on the fly.
Get the Business Logic Right
Don't underestimate the effort needed to create the necessary logic for your personalization effort, as well as the time and learning curve required to tag the content. Make sure you have sufficient technical resources to complete the personalization effort.
Step-by-Step Method is the Best
Don't try to add customization to a large site all at once. Begin with the areas most likely to drive sales, traffic and customer loyalty, and then branch out. After determining the level of customization you'd like to enable, figure out what it will cost to build it (including hardware and support).Throughout the planning process, don't forget to engage your users for feedback in whatever way possible-random surveys with prize incentives, focus groups and interviews with potential users at targeted trade shows. Ask them whether they need this functionality and, if so, which features in particular?
If you think about all these issues before dropping the big bucks, you'll be one up on many companies.
Assessing Your Personalization Effort
What will it take to personalize your website, or some part of it? The answer depends on the expansiveness of your vision.Some software applications, like Net Perception's Grouplens, are designed to let you import existing data to seed your customization engine, and this can take anything from one week to three weeks for implementation. But if you are setting up a rules-based system, the process could take much longer. Defining the business rules and the decision tree's schema can be time-consuming if your company hasn't already collected this data. And you can count on running into some hairy snags-such as unexpected new branches in the decision tree and a company-wide absence of written documentation of its business rules.
The best personalization projects go far beyond check boxes into anticipating customers' needs. When Cybermeals first started developing its back-end engines, the technical staff ran into a major problem it didn't anticipate: meal-ordering entails a lot more information than is cont-ained on a menu.
Waiters typically nudge patrons as to the kind of dressing they'd like, or if they want sour cream or butter on their baked potato.
A relational database worked well for basic restaurant information (hours of operation, address, phone number), but a more object-oriented approach was needed to support the varied data involved in actually ordering food.
Measuring Personalization
Many web builders who have created personalized sites say the technology still faces plenty of growing pains. In some cases, users must enter too much data or spend too much time configuring the site (or worse, rating content) before they gain any benefit.
Often, the payoff to the user for all that effort remains rather unclear. However, there are some simple ways of measuring the personalization and the simplest method of all will be to measure the boost in sales and traffic.
However, there are other ways to find out. This is by finding out the amount of time users spend on the website in correlation to personalization efforts. The thumb rule is that if a user spends more time on a website, then more likely is he or she to buy something from that page.
Conclusion
In the business on the net, surfers will definitely feel different when a personal touch is provided to the mouse clicks. A web site which can provide that extra personal touch can definitely score better in the rat race on the net.