Personalization on the Internet
Marketing 6222
Spring 2003
Werawat Ponrattanapaiboon
Andi Sobek
Sutee Srichiraratana
Gerald Stanford
Ton Yoong-Hua
Lily Zhang
Introduction
“Personalization is the use of technology to integrate personal information into the presentation and content of any customer contact. Rather than creating one message for all audiences, personalization allows companies to interact more "personally" with their customers, clients, or partners by tailoring content, advertising and services to specific individuals or groups. Personalization brings together the advantages of mass marketing enabled by the Internet with the one-to-one approach of sales and service based on personal relationships.” (1) The methods of collection, key players, benefits to both the company and the customer, as well as a few key issues to be aware of will be discussed here.
Methods of
collection
How do they know it’s me? How did they know that’s what I like?
The answer is: they know because you’ve provided them with the information.
Every time customers use the internet to shop, browse, chat, etc. their
information is collected and used to the advantage of the companies. Companies
that own websites, their advertising partners, internet relationship management
companies, and marketing analysis companies all have created ways of collecting
information from consumers who use the internet. Over the evolution of internet
business, several ways to collect customer information emerged. These collection
methods include registrations, surveys, customized personal home pages, check
outs processes, “creeping personalization”, and emails.
Whether customers knowingly or unknowingly provide their
information, collection begins as soon as they enter a website. Initial
information collection usually comes in the form of registrations. For example,
websites such as Excite.com (2) will allow users to have free email
if they register with the website. Registration will give the company the
customer’s name, address, some demographic information (i.e. age, sex), and not
to mention his/her email address.
Some companies get customers to register their information when they are
signing up to win a prize, or if they need to get more information, or if they
want something for free. Coupons.com (3), for example, has customers
register if they want to print free coupons that can be used in brick and
mortar stores.
Surveys and questionnaires are another way to gather
information from internet users. Websites may ask customers to help them answer
a few questions in order for the website to provide better products and
information. Websites, such as WebMd.com (4), for example, regularly
conducts a health quiz sponsored by pharmaceutical companies, to provide users
with a short analysis of their health status. If the user wants more
information, then they will also have to enter their email address in order to
receive more information. When customers are answering questions, they may not
feel as if they are giving up personal information. Customers may feel as if
they were helping to create a better website or as if they were getting
something for free. The companies and their sponsoring partners, on the other
hand, are using customers’ questionnaire answers and their requests for more
information as usable information to infer customers’ preference.
A less intrusive path to collecting information is offering
customers the opportunity to customize their own home page. Websites such as
MSN.com (5) have a “My MSN” page where customers can select their
favorite contents, local weather report, personal finance, and even the colors
and layout of the page. From this approach, companies and their partners can
see exactly what a customer’s interests are without having to ask for the
information.
When a customer makes a purchase through a website, they must
pay through a checkout processes. The checkout requires the customer’s name,
address, email address, credit card information, and shipping preferences. Most
customers willingly give the information as part of the checkout procedure.
What the customer does not realize is that their information is being collected
for future usage. When the customer
makes another purchase at the same website, all of his/her information is already
on file and they will not have to reenter the information. This is a very
convenient way for returning customers, and an easy way for to collect
customer’s information for the company.
“Creeping personalization” (6) on the surface
seems to be a less intrusive approach than offering customization of a home
page, registrations, or surveys. With this approach, companies track customer
usage patterns to gather information about them without the customer’s
knowledge. Whenever a customer clicks on a link, clicks an ad, or visits a
website, that information is collected. Collaborative-filtering software
resides on a web server and is used to track a user’s movement between pages.
Different pages have different merchandise. Each page or link will let the software
know what you are interested in. It compares the types of items you view and
navigate to and then is able to suggest other products that similar people have
bought. Companies then use the collected information to infer
what products, subjects, financial agencies, food, sports, pets, and so on that
the customer or similar customers may be interested in. Customers are never
asked to do anything and are not offered anything in return. This is how
websites such as Amazon.com (7) can suggest what books a customer
may be interested in reading without the customer having to ask for
advice.
When companies have a customer’s email address, they and
their advertising partners can send information on discounts, events, contests,
etc. to the customer. Most people are bombarded with these types of emails and
consider them as spam. However, if a customer reads the email and actually
clicks on the links or advertisement that is included in the email, then
further customer information can be collected and the company then can send
even more emails to the customer.
What happens to the information after it has been collected?--- There are several software that manipulate information to develop customer personalization. One main type of software is the Rule based software. This software divides customers into different segments based on their profiles or demographics, such as age, income, profession, and marital status. With this information they can offer products that those who visit their site are most likely to buy, based on the rules they have developed. The way the rules work is like computer code.
IF USER IS ‘MALE’
DISPLAY MALE PAGE
ELSE
DISPLAY FEMALE PAGE
A customer probably would not notice whether a site used Rule Based software or not and therefore would not know that it was offering them something different from everyone else.
There are several big players of the personalization software industry and they will be further discussed at the latter part of the paper.
A new area of research in the personalization arena is Spatio -Temporal page personalization. The idea behind this is that the section of a page a user visits most is kept track of and the page is rearranged so that the users favorite section is in the best viewing position. This would work well for sites that have many different sections to their web site. Large sites like Yahoo, AOL, and others where a user could easily get lost. A hybrid system called CRADLE, which is a combination of Rule Based and CASE Based Reasoning is being developed that improves the personalization process. First the Case Base Reasoning part determines the content by finding the appropriate profile and from the profile finds the best set of attributes that correlate to the particular profile. Its use is appropriate for selecting the best web community for a user based on their profile. The user’s sex, age, address, and other appropriate criteria would help the system determine the best community for the user.
So far, the present methods of collecting information have been
effective and efficient ways for most companies. However, as customers become
tired of registrations, surveys, emails and privacy becomes more of an issue,
companies will have to come up with more creative ways of getting information.
Benefits
of Personalization
Personalization will continue to work if the basic idea of value exchange is followed. This means we will get something of value from our customers and users if we provide them with something of equivalent value.
Benefits to Company
An organization will receive several benefits from providing personalization on their site, including customer loyalty, competitive advantage, lower marketing costs, the ability to identify the most profitable relationships, additional revenue from premium services, and the ability to adapt and improve their site, products, and services.
Loyalty
The benefits of loyalty flow straight to the bottom line. When a customer takes the time to invest in telling your company more about his or her wants, needs, and purchase decisions, he or she now has ownership in the relationship with the company. The cost to the customer to switch to the competitor continues to increase as he or she does more business with you.
The main benefits of customer loyalty are as follows:
Profitability: The revenue growth and cost savings associated with loyal customers lead to increase in profitability over the long term.
Referrals: Loyal customers can be your best and most cost-efficient advertisers. It has been shown that customer referrals produce a better quality of customer versus customers that come to your website from advertising or other forms of lead-generating communications. These referrals can also translate into profitable and loyal customers much faster.
Second chances: Loyal customers will give you a second chance if you mess up once in a while.
Competitive Advantage
Web technology itself will not give you a competitive advantage. Making it convenient for your online customers or users to gain a substantial benefit from what your company offers will be a key to your competitive advantage. Web site personalization will help you serve each customer to his or her individual satisfaction, over and over again. When online customers take the time to tell you what they want and need, they now have an investment in your service. As long as personalized service provides your customer with value, they will do business with you more often, and will think long and hard about switching to competitors especially if it means having to go through this investment process all over again with your competitor. If you are in a highly competitive market, where services and products are perceived as equal, then individualized service will help you compete effectively.
Lower Marketing Cost
Web site personalization will require sites to collect and store user data and interact with other back-end databases and systems. There will be an additional technology and marketing investment to design a personalized site; however, web sites that are database driven will reduce the time and costs associated with the continuous process of updating and revising.
It has been said that it costs five times more to acquire a new customer than it does to get business from existing customer. When you start the process of encouraging customer retention by using web site personalization, you can move down the path to profitability much faster.
Identifying the Most Profitable Relationships
Not all customers provide the same value to the organization. Web and information technologies can be combined to serve each customer according to his or her individuality. These technologies can also be used to evaluate each customer’s value. This allows you to segment your customers into categories. For example, most valuable customers are those who have the highest lifetime value (LTV). The objective with this customer group is to center your marketing activities on customer retention. The way to identify the most profitable relationships with customers or users is through interactive dialog and interaction tracking. The web has the unique ability to track customers. Your site can also be tied to databases and other systems such as sales, purchasing, shipping, and support. Customer data is required to determine which customers are your most profitable ones.
Additional Revenue from Premium Personalization
Web site personalization technology can be used to tie into premium services such as frequent buyer and other membership clubs. Premium services are those types of services for which you can charge the customer to participate. If there is enough value, you can enhance loyalty among members. These services also help you segment your customers and identify and reward highly loyal customers.
Adaptation and Continual Improvement of Product and
Services
Using web site personalization, you can tailor enhancements according to different segments of their user population, down to a single customer. Never before has it been easier to get customers and users to provide feedback. Technology has decreased the cost and minimized the time it takes to receive and respond to feedback.
Benefits to Customer
Web site personalization provides your online customers with benefits that include choice, significant time savings, and personalized service.
Choice
Web site personalization allows people to build profiles based on their own choices. The registration process itself gives people a sense that they are in control of choosing their preference. However, it is important to allow your customers the choice to opt-in and opt-out of the personalization process. Nothing frustrates an online user more than being forced to provide information to a web site in order to participate in personalization. If you provide a paid service, then customers know they must give a certain amount of personal information. You will want to determine what information is necessary for your marketing objectives and what is optional.
Time Savings
Even though people are spending increasingly more time on the internet, they still want the internet to save them time over alternative ways of gathering information, interacting with others that are geographically distant and buying products and services. Since the web is available 24 hours a day, 7 days a week, customers have an expectation of being able to interact and transact with a website at any time. There is a delicate balance between the online audience’s need for immediate gratification and the web marketer’s need to keep visitors on their site for several minutes in order to have a chance to build a relationship. Web site personalization provides the user with immediate gratification. This personalization model appears to be at odds with marketers’ needs to have users spend more time on their site. The key is to provide an interactive, personal experience that gives the users value. If this is done, you will no longer have to worry about the time people spend on your site; rather, you will be able to focus on the number of times particular users visit the site over time.
Personalized Service
It seems like a luxury to go into a retail store and have one of the salespeople remember you, your name, your last purchase, and your favorite products. In many instances, this type of personalized service was typical of how customers interacted with merchants. The merchants would remember their customers and the aggregation of their past purchases. In the past, this type of personalized service was not too difficult for merchants since there were fewer product choices. Now, this level of personalized service comes primarily from upscale merchants or extremely technologically advanced companies. The increased use of computer technology in the sales and marketing process, along with the ever growing access to the web, will make what was once considered a luxury the norm. For example, when you visit Amazon.com, you will automatically receive product recommendations, which are personalized book recommendations ready for your review, based on the past purchases of other Amazon.com customers with similar purchase patterns.
Big Players
Web personalization can be considered One-to-One Marketing. It benefits both the company and customers. Personalization permits a user to have a web site that tailors web content to their preferences. With web personalization, companies are able to generate a long lasting relationship with their customers. This technology allows a company to gather data from the pages the user visits and convert it into meaningful information. With this meaningful information, companies will be able to identify their areas of strength and weakness (what works and what doesn’t). Millions of dollars have been invested in this technology in order to better study consumer behavior.
Several big companies such as IBM, Net Perceptions, and Macromedia are playing a very important role in developing and enhancing this technology. Here are two examples of the big players and their technology:
Net Perceptions
Net Perceptions is an industry-leading provider of predictive analytic solution for the multi channel retail, manufacturing and distribution industries (8). Its software product “NetP 7’ utilizes the sophisticated data analytical tools to convert data (e.g. product, customer, transaction data) into usable marketing tools. Not only does NetP 7 enable businesses to make wise product recommendation to each customer, but also allows them to match the best customer for the product. NetP 7 focuses on One-to-One marketing. It helps product penetration, increased revenue, improved cross-sell success rate, and improved customer royalty.
One of the key advantages Net Perceptions has over its competitor is that NetP 7 does not require a major over-haul to an existing system. Since NetP 7 can easily integrate with the most existing CRM and e-commerce platforms, it typically requires 4-8 weeks of setup time. NetP 7 real-time intelligence is tailored to every customer interaction –face-to-face, phone, web, and direct mail. NetP 7 is using the latest Java-based client server application. The state of the art “Intelligence Manager” allows marketers to setup rules, filters, and campaigns on a browser-based application. In addition, NetP 7 also allows easy customization and upgrade.
Following is the graphic illustration of NetP 7 (Java-based) technology (9).
IBM
WebSphere Personalization for Multiplatform is one of IBM’s most famous personalization tools. IBM WebSphere provides the capabilities to build and deliver personalized intranet, extranet, or Web site pages (10). WebSphere can be run on HP-UX, AIX, Linux, Solaris, Windows 2000, and Windows NT operating systems.
Here is the graphic illustration of WebSphere Personalization:
There are basically 5 main components for WebSphere Personalization:
1) Personalization Workspace
2) WebSphere Site Analyzer
3)
WebSphere Personalization(11)
· Rules Engine: Executes the business rules that determine the content to be displayed to each visitor.
· Resource Engine: Enables business owners to call upon content and profile information from various sources in order to optimize their personalization strategy.
· Recommendation Engine: Uses collaborative filtering and item affinity analysis that offers content and product recommendations to customers. This in turn permits cross-selling and up-selling.
4)
WebSphere Application Server or
5) Profiling and Content
IBM has won numerous awards for its WebSphere product. A few are listed here (12):
- InfoWorld Reader’ Choice Awards 2002 Best Application Server: WebSphere Application Server 4.0. Best Portal Server: WebSphere Portal Server 1.2. Jul 2002
- eWeek Excellent Awards WebSphere Application Server. Feb 2002
- Network Computing Well-Connected Award Business Application – Web Application Servers: WebSphere Application Server3.5.1. Jun 2002
The reason WebSphere has won so many awards and has become so successful is because of its technology (as described above) and its brand name. Its reputation, service, support, and business computing experience are well-respected in the industry. IBM has years of experience in the business and because of its solid infrastructure, it is very difficult for competitors to match the integrity of IBM.
Challenges
Collection and Storage
The challenges that companies are facing now about information collection and storage result from the very simple characteristics of collecting information on-line - the overwhelming quantity. By saying quantity here, we are referring to the depth and breadth of the collected data. Information collected across different fields about the same customer is the depth, and information collected about different customers in the same field is the breadth. In order to make good use of this customer information, companies need to have their own strategy. Ethics, organizational structure, organizational culture, and even budgets are all the concerns of a company. It could be a painstaking process for a company to figure out what it should do to acquire the advantage it needs without being in conflict with its own interests. The following paragraphs present some basic concerns a company should take into consideration regarding these challenges (13).
Ownership of data
Web publishers have long used cookies or electronic tracking tags to monitor visitor habits and responses to web advertisements. Third-party ad networks and agencies do the same to create comprehensive pictures of consumer behaviors across multiple Web sites. The collection and usage of all this data leads to one question: Who owns this information?
Companies collect all this information because of their attempt to serve their customers better by providing them with more personalized service. Customers may or may not be aware of this while they are clicking through pages. Can companies claim ownership of the collected information? This question has been long argued since the day customer data was collected by electronic gadgets, but as of now there is still no consensus. Regarding this issue, most organizations have an implicit privacy policy in place, which customers can access at any time. The privacy policy usually states that customer information will be kept secured within the organization and will not be exposed to any third party without authorization. Some companies even react more proactively by asking customers for the right to use their information. No matter what, the respect for customers’ privacy is the key to avoid possible outbreaks. Eventually, it seems the question of who owns the information will no longer be the critical issue but “who can use the information” that companies should carefully deal with.
Security of data
Another challenge companies face when using personalization methods is the security of the data. Valuable data could be stolen or lost. When companies slowly move to IP storage, they have to figure out how to protect IP storage from internal or external hackers. In order to ensure the safety and confidentiality of customer records, companies must establish an information security system to protect against any threats or hazards to the security or integrity of hits information, and to protect against unauthorized access to or use of these records or information that would result in substantial harm or inconvenience to a customer. Of course, not all the companies can afford to establish a completely secure data system. Most companies use outsourcing to find technical help when setting up basic securities. Given the fact that even the most secured system in human’s mind could also be hacked by a human, a more passive way in preserving data is also widely used--disaster recovery. Disaster recovery is now driving many storage projects. Companies simply replicate data at two locations in case one system goes down. Even though it is not a good practice in terms of protecting customers’ rights, it is so in terms of protecting a company’s interests.
Management software
Of course, as we know, it is not difficult for companies to
acquire the means to store their massive amounts of data. The challenge here is
not about capacity, it is about information management.
Traditional ways of connecting servers and storage no longer
cut it because they are too slow. To talk about data management, we must first
know what SAN is. One definition found was published by Clint Boulton in the
article Storage Networking The Next Big Thing? printed
A SAN is a high-speed network, consisting of servers and
storage. Its purpose is to handle large amounts of data traffic between servers
and storage devices, without reducing the bandwidth of LAN/WAN…A SAN uses
storage protocols (SCSI), giving it the ability to transmit larger pieces of
data with reduced overhead and higher bandwidth… Operating behind the server,
SANs are often connected by a super-fast transmission technology known as the
Fibre Channel. A SAN consists of a communication infrastructure,
which provides physical connections, and a management layer, which organizes
the connections, storage elements and computer systems. (14)
This definition gives us a rough picture of the technology
that currently prevails for data storage.
SAN enables business flexibility. It makes it possible for users to
adjust to unknown requirement changes and allows the alteration of capacity
online. Users could manage about four times more storage than with
direct-attached storage.
Nowadays, most of the large companies have at least one
Fiber Channel SAN, and it is said that the growth would more than double over
years; however, the use of SAN has its pitfalls. Storage management tools in
general have been weak, and SAN is no exception. SAN has been critiqued for its
lack of coherent and user-friendly software. Such pitfalls have been proven to be the
cause of inefficiency. This last point has been made abundantly clear by recent
industry studies that show the average utilization rate of storage resources still
lagging below 50%. It took a while
before it was realized the problem is not how much capacity you have, but how
to manage the data you do have.
Monitoring Cost and Risk
While storage demands are aggressively increasing, operating budgets aren’t. There is no such thing as cheap storage, but there are ways to cut costs. However, to save money in the long run, money has to be spent up front on storage networking run. The major disadvantages of SAN have been both cost and complexity. The expense is coming down, but for a lot of firms, it is still beyond what they can afford (or what they are willing to afford). A company has to consider not only their needs, but also, their budgets and risks. An organization’s culture is also a factor that may influence the decision. Since there is still much room for improvement for SAN, should a company be conservative and wait until SAN is completely developed, or should it adopt SAN in an early stage so as to take the advantage as an early mover? Deciding on the right time for a company to adopt SAN is a major challenge for many companies.
Data Analysis
For companies attempting to acquire an advantage by rendering
their customer data in use, they need to find effective ways to first
standardize critical information, and then offer a way for employees to access
it from all enterprise systems. Some of the data companies collect are of low
quality and possible haven't been used for a long time. To bring together all
this data is a big task. Nobody would disagree that the most difficult part of
data analysis is integration; it requires pulling together different data
stores with data of varying ages on different databases. The quality of data
(clean data), therefore, is critical to the efficiency of data analysis.
The process to retrieve public records data and perform data analysis is
very labor intensive; it is not a one-day assignment. Sometimes companies just
can’t act quickly enough on the data they collect. People schedule several
meetings to analyze the data; however, by the time the analysis is over, the
window of opportunity has been missed.
"It's massive," says Derek Strauss,
CEO of Bethesda, Md.-based BI (business-intelligence) applications hoster
Assurenet, of the problems of data quality in CRM applications. "You have
to have accurate information, and most of the front-end systems which deal with
customers do not have accurate information about the customers. There's
disjointed [data]; there's a lot of blanks in some of the critical
fields." (15)
Imagine how a seemingly minute mistake such as being one digit
off on a customer's street address can make data sets a plague. This kind of
problem can lead to a company sending out multiple mailings to the same
customer, because the address is listed in two different ways. Faulty and inconsistent data sets would
prevent enterprises from having a clear and unified profile of each customer.
Conclusion
Personalization and
customization are both useful tools for internet based companies, and can lead
to many long term benefits to the company and to a lesser extent, the customer.
There are many methods of collecting information needed for personalization,
and many more are sure to be developed. Companies who address he issues of
collecting and analyzing the information received into useful data into their
business models, will continue to reap both the long term and short term
benefits personalization offers.
References:
3. http://my.webmd.com/content/article/50/40552?
4. www.msn.com
5. “Web Personalization Does It Your Way,” Online Magazine, March 1999, http://www.findarticles.com/cf_0/m1388/2_23/53923511/p1/article.jhtml?term=personalization
7. http://www.personalization.com/about.php
8. http://www.netperceptions.com/tech.php
9. http://www-3/ibm.com/software/webservers/personalization/about.html
10. http://www-3.ibm.com/software/webservers/personalization/
11. http://www-3.ibm.com/software/webservers/wsawards.html
12. http://www.internetnews.com/xSP/article.php/736171
13. http://www.computerworld.com/news/1999/story/0,11280,43546,00.html
14. http://plc.fis.utoronto.ca/it/personal.asp
15. http://www.crmxchange.com/whitepapers/answerthink-business-case-personalization.html