Online File W8.1 Online Buyer Decision Making Process...

Chapter Eight: Marketing and Advertising in E-Commerce 8-1

Online File W8.1 Online Buyer Decision Making Process

Sources: Compiled from Silverman, B. G., M. Bachann, and K. Al-Akharas. “Implications of Buyer Decisions Theoryfor Design of E-Commerce Web Sites.” International Journal of Human-Computer Studies (November 2001, © 2006);Tutor2u.net. “Buyer Behavior Decision-Making Process.” 2011. tutor2u.net/business/marketing/buying_decision_process.asp (accessed April 2011); and author’s experience.

Consumer

Identify BuyingCriteria

Search forProducts, Vendors

CompareAlternatives

ProductInformation,Reviews, Ratings,SocialRecommendations

Search Types,Site Metaphor,Managerial—Provides Aids,(FAQs), SocialRecommendations

Scope ofComparison,Modes ofComparisons,Software Agents,Social Support

Price, FinancialTerms, Warranty,Shipping

Negotiations

Cross-Transaction Concerns

Personalization, Users’ Preferences, Customer Service

Buyer’sBehavior

Information,Support

Activities

EXHIBIT W8.1.1

8-2 Part 4: EC Support Services

Online File W8.2 EC Trust Model

Trust Certificates and SealsVendor Evaluation (BBBOnline)Product EvaluationConsumer EndorsementCo-BrandingAlliances/AffiliatesWebsite DesignReturn PolicyPrivacy StatementFree SamplesEducation Efforts by Vendor(The use of security, size, andfinancial resources)Simplicity of ShoppingRecommendation AgentAvatar

Demographics, Previous Experience,Personality, Cultural Difference

Trust in Internetas Shopping Channel

Trust inInternet Merchant

Trust in Business andRegulatory Environments

Trustworthiness of Internet Merchant

Trustworthiness ofShopping Channel

CompetencyBenevolenceIntegritySecurity ControlPrivacy Control

ReliabilityUnderstandabilitySecurity and PrivacyPayment and Logistics

Structural Assurance

Business CultureConsumer ProtectionEffective Law

EC Trust

EXHIBIT W8.2.1

Online File W8.3 Spyware

Advertisers use spyware to spy on your movements on the Internet and send you e-mail ads or pop-ups based on whatthey learn about you. Authors of shareware and freeware often use spyware in order to make money on their products,which are often offered free to users. Shareware authors may insert banner ads of their advertisers into their productsin exchange for a portion of the revenues generated by the banners. This way, end users do not have to pay for thesoftware, and the developers get paid for their efforts. End users who find the banners to be annoying can usually obtaina banner-free copy of the shareware by paying a regular licensing fee to the developer. Increasingly, however, spyware isbeing embedded in purchased software.

Spyware CategoriesSpyware threats come in different flavors. One variety of spyware is malware. This type of spyware can modify the systemsettings on the user’s computer (see Chapter 11) and perform other undesirable tasks. A hijacker is a piece of spyware thatredirects the user’s browser to a particular website. A dialer is spyware that dials a service, most likely porn sites, withoutthe user’s knowledge. In many cases, the user will be held responsible for the bill! A Trojan horse is spyware that isattached to a program and performs undesirable tasks on the user’s system. Collectware is spyware that collects informa-tion about the user. This information is then provided to a third party without the user’s knowledge.

(continued)


Online File W8.3 (continued)

Simply the Best (simplythebest.net/shareware/utilities/spyware_cleaners.html) identifies the following types of spyware:

◗ Adware networks. Ad-serving networks such as DoubleClick, Web3000, Radiate, SaveNow, and GAIN pay publishers ofgames, utilities, and music/video players per download to include their ad-serving programs.

◗ Stalking horses. A number of programs that collect information on the user and facilitate adware networks to function ondesktops. These programs are bundled in many popular programs and often are presented in installation disclosure screensas desirable add-ons to Trojan horse hosts. Examples include TopText, Cydoor, OnFlow, Medialoads, Delfin, WebHancer, andNew.net.

◗ Trojan horses. These are popular Internet downloads that usually come with the ad-serving network basic software and atleast one stalking horse. They are contained in Kazaa, Grokster, Morpheus, Limewire, AudioGalaxy, iMesh, and DivX.

◗ Backdoor Santas. These are stand-alone programs that incorporate similar approaches as those just discussed. However,they have no links to ad-serving networks, although their purpose is to collect information from users. They are locatedin Alexa, Hotbar, Comet Cursor, eWallet, CuteFTP, and BonziBuddy.

◗ Cookies. Netscape Navigator and Internet Explorer still send out cookies even after cookies have been disabled in thebrowser settings. Any and all cookie files must be manually deleted on your system in order to eliminate being tracked bythird-party ad networks or spyware or adware providers.

For more information about spyware, see spywareguide.com and Chapters 9 and 14.

Online File W8.4 The List of Information Provided by Clickstream Data

B2C clickstream data can reveal information such as the following:

◗ What goods the customer has looked at◗ What goods the customer has purchased◗ What goods the customer examined but did not purchase◗ What items the customer bought in conjunction with other items◗ What items the customer looked at in conjunction with other items but did not purchase◗ Which ads and promotions were effective and which were not◗ Which ads generate a lot of attention but few sales◗ Whether certain products are too hard to find and/or too expensive◗ Whether there is a substitute product that the customer finds first◗ Whether there are too many products for the customer to wade through◗ Whether certain products are not being promoted◗ Whether the products have adequate descriptions


Online File W8.5 From Mass Advertising to Interactive Advertising

Mass Advertising Direct-Mail Advertising Interactive Advertising

Desired outcomes Volume sales Targeted reach, more sales,customer data

Volume sales, CRM, customerfeedback, acquire new customers,improve target marketing ROI

Consumer activities Passive Passive ActiveLeading products Food, personal care

products, beer, autos,cameras, computers,appliances

Credit cards, travel, autos,some appliances

Upscale apparel, banking, books,travel, insurance, computers,autos, jewelry, office supplies

Market strategy High-volume products Targeted goods to segments Targeted individual or groupsNerve centers(command centers)

Madison Avenue(advertisers)

Postal distribution centers,warehouses

Cyberspace, logistics companies

Preferred mediavehicle

Television,newspapers,magazines

Mailing lists Online services, e-commerce,banners

Preferred technology Storyboards, TV Databases Servers, on-screen navigators, the Web

Worst outcome Channel surfing Recycling bins Log off

REFERENCES FOR ONLINE FILE W8.5Hallerman, D. “The Death of Mass Marketing.”

iMediaConnection, June 16, 2006.InformationWeek. “From Mass Advertising to Interactive

Advertising.” InformationWeek, October 3, 1994, p. 26.


ONLINE FILE W8.6Application Case

1-800-FLOWERS.COM USES DATA MINING TO FOSTERCUSTOMER RELATIONSHIP MANAGEMENT1-800-Flowers.com (1800flowers.com) is a true Internetpioneer. It had an Internet presence in 1992 and full-fledgede-store capabilities in 1995. Online sales are a major market-ing channel (in addition to telephone and fax orders).Competition is very strong in this industry. The company’ssuccess was based on operational efficiency, convenience(24/7 accessibility), and reliability. However, all majorcompetitors provide the same features today. To maintain itscompetitive advantage, the company transformed itself intoa customer-intimate organization, caring for more than15 million customers. The challenge was to make 1-800-Flowers.com the only retailer that customers really trustwhen shopping for gifts online or by phone.

The company decided to cultivate brand loyalty throughcustomer relationships, which is based on intimate knowl-edge of customers. How is this accomplished? The companyuses SAS software that spans the entire decision-supportprocess for managing customer relationships: collecting dataat all customer contact points, the company turns that datainto knowledge for understanding and anticipating customerbehavior, meeting customer needs, building more profitablecustomer relationships, and gaining a holistic view of acustomer’s lifetime value.

Using SAS Enterprise Miner, 1-800-Flowers.com siftsthrough data to discover trends, explain outcomes, andpredict results so that the company can increase responserates and identify profitable customers. The rationale for thecustomer intimate effort is to build loyalty. In addition toselling and campaign management, the ultimate goal is tomake sure that when a customer wants to buy, he or shecontinues to buy from 1-800-Flowers.com and cannot becaptured by a competitor’s marketing. To build that kind ofloyalty, it is necessary to know your customers and build asolid relationship with each one of them.

Identifying Each CustomerAt 1-800-Flowers.com, the objective is not just about gettingcustomers to buy more. It is about making sure that when

they decide to purchase a gift online or by phone they donot think of going to the competition. Loyalty is earnedthrough the quality of the relationship offered. The difficultyis that not every customer wants the same relationship.Some want you to be more involved with them than others;some will give you different levels of permission on how tocontact them. At the end of the day, the data miningsoftware helps the company identify the many different typesof customers and how each would like to be treated.

The company plans its ad campaigns based on theresults of the data mining done on a one-to-one basis. Forexample, some customers like to be notified about sales,others do not. Some prefer notification via e-mail. The datafor the analysis is derived from the data center, which isbacked-up by AT&T technology.

Many factors have contributed to the company’s recentrevenue growth—customer relationship management amongthem. The data mining analysis provides rapid access to bettercustomer information and reduces the amount of time thecompany needs to spend on the phone with customers, whichmakes better use of everybody’s time. The net result is thatcustomer retention has increased 15 percent over 2 years.

Questions1. Why is being number one in operation efficiency not

enough to keep 1-800-Flowers.com at the top of itsindustry?

2. Identify all market research activities and methods.

3. How was the transformation to a customer-intimatebusiness accomplished?

4. What is the role of data mining and personalization incustomer relationship management in the case?

5. How is the one-to-one relationship achieved in thecase?

REFERENCES FOR ONLINE FILE W8.61800flowers.com (accessed March 2011).Anonymous. “Ultimate Relationship in Bloom. Customer

Success: 1-800-Flowers.” Sas.com, 2006. sas.com/success/1800flowers.html (accessed March 2011).

Reda, S. “1-800-Flowers.com and AT&T CultivateRelationship Rooted in Common Business Objectives.”Stores, October 2006.


Online File W8.7 Advantages and Limitations of Internet Advertising

Medium Advantages LimitationsTV ◗ Intrusive impact—high attention getter

◗ Ability to demonstrate product and tofeature “slice of life” situations

◗ Very “merchandisable” with mediabuyers

◗ Fragmented ratings, rising costs,“clutter”

◗ Heavy “downscale” audience skew

◗ Time is sold in multiprogram packages;networks often require major up-frontcommitments; both limit theadvertiser’s flexibility

Radio ◗ Highly selective by station format

◗ Allows advertisers to choose the timeof day or the day of the week to exploittiming factors

◗ Copy can rely on the listener’s mood orimagination

◗ Audience surveys are limited in scope,do not provide socioeconomicdemographics

◗ Difficult to buy with so many stationsto consider

◗ Testing of copy is difficult becausethere are few statistical guidelines

Magazines ◗ Offer unique opportunities to segmentmarkets, both demographically andpsychographically

◗ Ads can be studied and reviewed atleisure

◗ High impact can be attained with goodgraphics and literate, informative copy

◗ Reader controls ad exposure, can ignorecampaign

◗ Difficult to exploit “timing” aspects

Newspapers ◗ High single-day reach opportunity

◗ Reader often shops for specificinformation when ready to buy

◗ Portable format

◗ Lack of creative opportunities for“emotional” selling campaigns

◗ High cost for large-size ads

◗ Lack of demographic selectivity; despiteincreased zoning, many markets haveonly one paper

◗ Low-quality reproduction, lack of color

Internet ◗ Internet advertisements are available24 hours a day, 365 days a year

◗ Costs are the same regardless ofaudience location

◗ Accessed primarily because of interestin the content, so market segmentationopportunity is large

◗ Opportunity to create one-to-one directmarketing relationship with consumer

◗ Multimedia will increasingly create moreattractive and compelling ads

◗ No clear standard or language ofmeasurement

◗ Immature measurement tools andmetrics

◗ Although the variety of ad contentformat and style that the Internetallows can be considered a positive insome respects, it also makes apples-to-apples comparisons difficult for mediabuyers

◗ Difficult to measure size of market,therefore it is difficult to estimaterating, share, or reach and frequency

◗ Audience is still small


Online File W8.8 How to Attract Web Surfers

Advertisers use dozens of innovative techniques to lure consumers into viewing online ads. The following list is only asample of the many interesting ideas companies have used to attract Web surfers. For more on promotions, visitpromomagazine.com.

◗ Retailers can provide online shoppers with special offers while they are purchasing or “checking out.” If a shopper’sprofile or shopping history is known, the ads can be targeted.

◗ Netstakes (webstakes.com) runs sweepstakes that require no skill. Users register only once and win prizes at random.Sponsors pay Netstakes to send them traffic. Netstakes also runs online ads, both on the Web and through e-mail lists thatpeople subscribe to.

◗ Cydoor (cydoor.com) places ads, news, and other items on software applications. Consumers who download the softwarereceive a reward each time they use the software (and presumably read the ads).

◗ CBS MarketWatch (marketwatch.com) uses animated beer bottles and interactive charts to attract viewers to its free finan-cial site.

◗ Sometimes a catchy name draws Web surfers. For example, an old-economy seller of hard-to-find lightbulbs changed itsname to topbulb.com and created an online catalog, called the Bulbguy, through which it sells lightbulbs online at adiscount. The website is advertised both online and offline, and business is booming!

◗ Promotionworld.com is a magazine-format site dedicated to website promotions. Users can find rich resources and promo-tions on how to increase Web traffic.

◗ To promote its sport utility vehicle, the 4Runner, Toyota wanted to reach as many Internet users as possible. The companydisplayed Toyota banners on the search engine AltaVista (altavista.com). Whenever someone used AltaVista to search foranything related to automotives, they would see the Toyota banner. Also, Kelly Blue Book’s new-car pricing catalog(kbb.com) had links to Toyota’s car. In the first 2 months of the campaign, over 10,000 potential car buyers clicked on thebanner ads looking for more detailed information about the Toyota 4Runner.

◗ Web surfers can play games, win prizes, and see “e-tractions” at uproar.com. Special promotion campaigns are also featured.◗ To promote its job-recruiting visits on U.S. college campuses, IBM created over 75,000 college-specific banners such as, “There

is life after Boston College: click to see why.” The students clicked on the banners at a very high rate (5 to 30 percent). As aresult of this success rate, IBM restructured its traditional media plans using the “Club Cyberblue” scheme.

◗ Each year, almost 500,000 brides-to-be use theknot.com to plan their wedding. A “Knot Box” with insert folders is sent tousers by regular mail. Each insert is linked to a corresponding page at theknot.com. Advertisers underwrite the mailcampaign. The website provides brides with information and help in planning the wedding and selecting vendors. Orderscan be placed by phone or online (not all products can be ordered online). weddings411.com is a similar service, operat-ing primarily online.

Bargain hunters can find lots of bargains on the Internet. Special sales, auctions, and downloading of coupons arefrequently combined with ads. Of special interest are sites such as coolsavings.com, hotcoupons.com, supercoups.com,clickrewards.com, and mypoints.com. A popular lottery site is worldwidelotto.com. In addition to lotteries and coupons, freesamples are of interest to many consumers, and “try-before-you buy” gives consumers confidence in what they are buying.Freesamples.com began to offer free samples in June 2000.


Online File W8.9 E-Mail Advertising Methods

E-Mail Promotions. E-Greetings Network (egreetings.com) produces digital postcards and animations for its customers,who are both individuals and corporations. For a modest membership fee ($13.95 annually), members have access to over5,000 e-greeting cards, plus designs for flyers, fax covers, and envelopes. Through its free membership trial and itsmembers list, E-Greetings Network has compiled a database of millions of recipients. E-Greetings Network’s goals for its promotion campaign include bringing value to its customers, driving traffic and transactions at the customer’s site, stimu-lating involvement with the site, expanding customer relationships, offering added means of sponsorship, and supportingbrand affinity. E-Greetings’ main relationship-building tool is its newsletter “What’s Up @ E-greetings!” Key factors in itssuccess are the fact that the mailing list is totally voluntarily (opt-in); newsletters are distributed on a regular, biweeklyschedule; the content is relevant; and it handles unsubscribe difficulties and customer service in a timely manner.

Discussion Lists. Internet Security Systems (ISS), with $1 billion in sales per year, provides software that detectsweaknesses in systems and gives detailed corrective measures for fixing security holes. Its success began when its founder,Chris Klaus, posted a notice about his security software on a newsgroup. He then offered a shareware version of theprogram to the newsgroup members and received 200 e-mail responses the following day. His company’s discussion listprogram includes approximately 80 specialized e-mail lists reaching over 100,000 people through discussions, partner lists,customer lists, and product announcement lists (Kinnard 2002). Sponsorship of discussion groups, communities, andnewsletters is becoming quite popular on the Web (see sponsorship.com).

E-Mail List Management. L-Soft’s Listserv (lsoft.com), the leader in software for e-mail list management and marketing, isknown for its electronic newsletters, discussion groups, and direct e-mail messaging. According to Kinnard (2002), thecompany understands that 9 out of 10 customer interactions are not transactions, so it offers database integration, mailmerges, and customizable Web interfaces that allow companies to send pertinent information, such as product details oradvertising, to specific customers. Listserv delivers 30 million messages each weekday and 1 million messages per hourfrom a single server. For e-mail challenges, see Berkowitz (2004).

REFERENCES FOR ONLINE FILE W8.9Berkowitz, D. “Responsys Responds to E-Mail Challenges.”

eMarketer, April 9, 2004.Kinnard, S. Marketing with E-Mail, 3rd ed. Gulf Breeze,

FL: Maximum Press, 2002.


Online File W8.10 Software Agents in Marketing and Advertising Applications

As the volume of customers, products, vendors, and information increases, it becomes uneconomical, or even impossible, forcustomers to consider all relevant information and to manually match their interests with available products and services.The practical solution to handling such information overload is to use software (intelligent) agents. In Chapter 3, wedemonstrated how intelligent agents help online shoppers find and compare products, resulting in significant time savings.

In this section, we will concentrate on how software agents can assist customers in the online purchasing decision-making process as well as in advertisement. Depending on their level of intelligence, agents can do many things.

A Framework for Classifying EC AgentsExhibit 8.2 (p. 399) detailed the customer’s purchase decision-making process. A logical way to classify EC agents is byrelating them to this decision-making process (in a slightly expanded form), as shown in Exhibit 8.10 (p. 442). In thedecision-making model in Exhibit 8.2 (p. 399), the second step was information search. Because of the vast quantity ofinformation that software (intelligent) agents can sift through, the step has been split here into two types of agents:those that first answer the question, “What to buy?” and those that answer the next question, “From whom?” Let’s seehow agents support each of the phases of the decision-making process.

Agents That Support Need Identification (What to Buy)Agents can help buyers recognize their need for products or services by providing product information and stimuli. Forexample, Expedia expedia.com notifies customers about low airfares to a customer’s desired destination whenever theybecome available.

Several commercial agents can facilitate need recognition directly or indirectly. For example, FindGift (findgift.com)asks customers questions about the person for whom they are buying a gift, and helps the buyer hunt down the perfect gift.

Agents That Support Product Brokering (From Whom to Buy)Once a need is established, customers search for a product (or service) that will satisfy the need. Several agents are avail-able to assist customers with this task. The comparison agents cited in Chapter 3 belong in this category. An example ofhow these agents are used in advertising is provided in Case W8.10.1.

Some agents can match people that have similar interest profiles. Even more ambitious agents try to predict whichbrands of computers, cars, and other goods will appeal to customers based on market segmentation preferences in avariety of different product categories, such as wine, music, or breakfast cereal. (See the discussion in Section 8.4 oncollaborative filtering.)


CASE W8.10.1EC Application

FUJITSU USES AGENTS FOR TARGETED ADVERTISING IN JAPANFujitsu (fujitsu.com) is a Japanese-based global provider ofInternet-focused information technology solutions. Since theend of 1996, Fujitsu has been using an agent-based technol-ogy called the Interactive Marketing Interface (iMi). The system allows advertisers to interact directly withspecific segments of the consumer market through the useof software agents, while ensuring that consumers remainanonymous to advertisers. Consumers submit a personalprofile to iMi, indicating such characteristics as productcategories of interests, hobbies, travel habits, and themaximum number of e-mail messages per week that theyare willing to receive. In turn, customers receive productannouncements, advertisements, and marketing surveys by e-mail from advertisers based on their personal profileinformation. By answering the marketing surveys or

acknowledging receipt of advertisements, consumers earn iMi points, redeemable for gift certificates and phone cards.Many other companies in Japan (e.g., nifty.com andlifemedia.co.jp) also use this technology.

Fujitsu’s recent interactive marketing is called DigitalMedia Solutions, for information campaigns for retailers.

Sources: Compiled from fujitsu.com (accessed January 2011) andFujitsu (2005).

Questions1. Why would customers agree to have a personal profile

built on them?

2. What is the role of the software agent in this case?



Agents That Support Merchant Brokering and ComparisonsOnce a consumer has a specific product in mind, he or she needs to find a place to buy it. BargainFinder (from Accenture)was the pioneering agent in this category. When used for online CD shopping, for example, this agent queried the price ofa specific CD from a number of online vendors and returned a list of prices. However, this system encountered problemsbecause vendors who did not want to compete on price managed to block out the agent’s requests. (Today’s version is atcdrom-guide.com, “Bargain finder.”) The blocking problem has been solved by agents such as Inktomi Shopping Agent, MySimon (mysimon.com), and Junglee (of amazon.com). These agents originate the requests from whatever computer theuser is accessing at the time. This way, vendors have no way of determining whether the request comes direct from a realcustomer or from the comparison agent.

Fraud is of major concern to buyers because buyers cannot see the products or the sellers (see Chapter 9 and 14).Several vendors offer agent-based fraud detection systems. One such system is Risk Suite (fairisaac.com). It is based on patternrecognition driven by neural computing. Other products from Fairisaac are FICO Risk Score and VeriComp Fraud Managers.

Comparison AgentsPart of the merchant-brokering process is determining price and other purchase criteria. Large numbers of agents enableconsumers to perform all kinds of comparisons, as was shown in Chapter 3. Here are some additional examples:

◗ Allbookstores.com and bestbookbuys.com are two of several agents that help consumers find the lowest prices of booksavailable online.

◗ Bottomdollar.com, compare.net, shopper.cnet.com, and roboshopper.com are examples of agents (out of several dozen) thatsuggest brands and compare prices once consumers specify what they want to buy.

◗ Pricescan.com guides consumers to the best prices on thousands of computer hardware and software products.◗ Buyerzone.com is a B2B portal at which businesses can find the best prices on many products and services.

Agents That Support Buyer–Seller NegotiationThe traditional concept of “market” implies negotiation, mostly about price. Whereas many large retail stores engage in fixed-price selling, many small retail stores and many markets use negotiation extensively. In several cultures (e.g., Chinese),negotiation is very common. In many B2B transactions, negotiation is common, too. The benefit of dynamically negotiating aprice is that the pricing decision is shifted from the seller to the marketplace. In a fixed price situation, if the seller fixes aprice that is too high, sales volume will suffer. If the price is set too low, profits will be lower.

Negotiations, however, are time consuming and often disliked by individual customers who cannot negotiate properlybecause they lack information about the marketplace and prices or because they do not know how to negotiate. Manyvendors do not like to negotiate either. Therefore, electronic support of negotiation can be extremely useful.

Agents can negotiate in pairs, or one agent can negotiate for a buyer with several sellers’ agents. In the latter case,the contact is done with each seller’s agent individually, and the buyer’s agent can conduct comparisons. Also, customerscan negotiate with sellers’ agents. One system automates the bargaining on a seller’s side. The system can bargain withcustomers based on their bargaining behavior. For example, if the customer starts very low, the system helps the sellerknow how to respond. For details, see Chapter 4 (for B2B).

Agents That Support Purchase and DeliveryAgents are used extensively during the actual purchase, often arranging payment and delivery. For example, if a customermakes a mistake when completing an electronic order form, an agent will point it out immediately. When a customer buysstocks, for example, the pricing agent will tell the customer when a stock they want to buy on margin is not marginable orwhen the customer does not have sufficient funds. Similarly, delivery options are posted by agents at amazon.com andother e-tailers, and the total cost of the transaction is calculated in real time.

Agents That Support After-Sale Service and EvaluationAgents also can be used to facilitate after-sale service. For example, the automatic e-mail answering agents described inChapter 11 usually are effective in answering customer queries. A non-Internet-based agent can monitor automobile usageand notify owners when it is time to take their car in for periodic maintenance. Agents that facilitate feedback fromcustomers also are useful.

(continued)



Character-Based Animated Interactive AgentsSeveral agents enhance customer service by interacting with customers via animated characters. Similar agents are used to facili-tate advertising. As presented in Chapter 2, animated characters are software agents with personalities. They are versatile andemploy friendly front ends to communicate with users. They are not necessarily intelligent. These animated agents also are calledavatars. Avatars are animated computer representations of humanlike movements and behaviors in acomputer-generated 3-D world. Advanced avatars can “speak” and exhibit behaviors such as gesturesand facial expressions. They can be fully automated to act like robots. The purpose of avatars is tointroduce believable emotions so that the agents gain credibility with users.

Example: Brand Names at a Korean Virtual Mall. Avatars are big business in South Korea.Internet users express themselves by putting clothes, shoes, and accessories on their avatars.The clothes are really pixels on the computer screen designed for avatars that represent the users and are moved arounda virtual chat room. Clothing the avatars in attire bought in virtual malls is part of the fun. Sayclub, operated by NeoWiz,was the first to introduce avatar services there in 2000. The company had more than 15 million members in 2002,who spent a total of $1.6 million a month on their avatars, dressing them in the over 30,000 outfits from the virtualshopping mall (sayclub.com).

“It is an unusual strategy, but avatars can be very effective marketing tools,” says Chung Jae Hyung, chief executiveofficer at DKIMS Communications, an online marketing agency. “They are so popular with young people these days thatthey can get you a lot of exposure very quickly. An avatar is a given; just like everyone has a cell phone, everyone has anavatar,” says Chung.

As competition grew from the top portals, such as Yahoo!, Sayclub responded by offering more items, including hairdyes, accessories, and brands. “We needed something to differentiate ourselves and improve our brand image,” says ChangHyun Guk, manager of the business planning team at Sayclub. “The best way to do that was to bring real-life brands to ourvirtual mall.” Therefore, to improve its own brand image, Sayclub sought to offer well-known consumer brands as productsone could buy for one’s avatar from the Sayclub site. In addition to improving the Sayclub brand, these products wouldgenerate money from additional sales.

However, convincing top brands to go virtual was not easy. For example, because Mattel, the maker of Barbie, did not haveany idea what the avatar market was all about, it needed to be educated before it would sign a licensing agreement allowingoutfits from the Barbie Fashion Avenue line of doll dresses to be “avatarized” for a percentage of total sales. Numerous Barbieoutfits have gone on sale at prices ranging from $4 to $5.35. That may not sound like much, but as of June 2002, avatar outfitsmade up almost 15 percent of Barbie’s licensing business in Korea. By exposing Sayclub’s users (mostly people in their teens and20s) to Barbie paraphernalia, Mattel has been able to extend her popularity beyond children the ages of 8 or 9. Jisun Lee, a 23-year-old student, had not owned anything “Barbie” in over a decade but spent more than $85 in 2002 dressing her avatarin Barbie outfits.

During the Korean World Cup games in June 2002, Sayclub formed a partnership with Nike Korea and introducedavatars based on real images of Korean soccer players. They provided various soccer-related avatar items, includinguniforms and Nike products. Sayclub granted uniform numbers of national soccer players to all users who bought Nikesoccer items, including soccer player avatars or national team uniforms. They also gave away gifts to users with uniformnumbers (selected by lottery) if the players their avatars represented scored during the games.

Sources: Compiled from sayclub.com (accessed January 2011) and neowiz.com (accessed January 2011).

ChatterbotsA special category of animated characters is characters that can chat, known as chatterbots.A chatterbot is a program that attempts to simulate a conversation, with the aim of at leasttemporarily fooling a customer into thinking they are conversing with a human. The conceptstarted with Eliza, created by Joseph Weizenbaum at MIT in 1957. In his program, users conversedwith a psychoanalyst. Today’s version is very powerful (try simonlaven.com/ eliza.htm; the programcan be downloaded for free). The major differences are that today the programs are on the Web andthey include a static or moving character. The technology is based on natural language program-ming (NLP), an applied artificial intelligence program that can recognize typed or spoken key wordsand short sentences.

avatarsAnimated computercharacters that exhibithumanlike movementsand behaviors.

chatterbotA program that attemptsto simulate a conversa-tion, with the aim of atleast temporarily foolingcustomers into thinkingthey are conversing witha human.



A major use of character-based interactive agents is in customer service and CRM. The following sites offer demos and the opportunity to converse with virtual representatives:

◗ Artificial-life.com. This site offers CRM and other agents. The site can be accessed by cell phones as well as traditional Webconnections. This company offers e-learning applications, too.

◗ Verity.com. This site offers “Virtual Response,” which acts as a “v-rep” (virtual representative) that can answer customerquestions and provide potential solutions.

◗ Zabaware.com. This site provides desktop assistance for answering customers’ queries. For an inventory of chatterbots andother resources, visit Simon Laven’s site (simonlaven.com).

Chatterbots can do many things to enhance customer service, such as greeting consumers when they enter a site orgiving the consumer a guided tour of the site. For example, consider the following chatterbot agents: “Ed, Harmony, andNina” are virtual guides that help visitors who wish to learn more about products and tools available at extempo.com(which specializes in avatars). “Arthur bot” helps people build their own chatterbots at ai-buddy.com. For additional infor-mation on interactive characters, see Agent Interactive (2006), microsoft.com/msagent, and artificial-life.com. Rehm andAndre (2005) studied communication behaviors of humans and whether these behaviors differed when conversing with anagent as opposed to talking with other humans.

Other EC AgentsOther agents support consumer behavior, customer service, and advertising activities. For example, Hiring Gatewayvivente.com/resumix/website is an application from Yahoo! Resumix, which wanders the Web looking for Web pages con-taining résumé information. If it identifies a page as being a résumé, it tries to extract pertinent information from thepage, such as the person’s e-mail address, phone number, skill description, and location. The resulting database is usedto connect job seekers with recruiters. For current lists of various EC agents, see botspot.com and agents.umbc.edu(see the “Agents 101” tutorial).

Questions1. List the major types of software (intelligent) agents used in customer-related and advertising applications.

2. What role do software agents play in need identification?

3. How do software agents support product brokering and merchant brokering?

4. What is an avatar? Why is it used on websites?

REFERENCES FOR ONLINE FILE W8.10Agent Interactive. “Interactive Agents, 2006.” agentinter-

active.se/en/interactive_agents.php (no longer aval-able online).

neowiz.com (accessed July 2011).Fujistu. “Commercial and Loyalty: Creating Customer

Responsiveness.” 2005. fujitsu-services.ae/common/downloads/uae/pdf/industr ies/commercial_loyalty.pdf (accessed March 2011).

Rehm, M., and E. André. “From Chatterbots toNatural Interaction—Face-to-Face Communication withEmbodied Conversational Agents.” IEICE Transactionson Information and SystemsE88D, no. 11 (2005).

sayclub.com (accessed July 2011).


REFERENCES FOR ONLINE FILE W8.11Reed, J. Getting Up to Speed with Online Marketing.

Edinburgh Gate, Harlow, UK: Pearson EducationalLtd., 2011.

Web Ad.vantage. “Online Marketing Strategy.” 2011.webadvantage.net/internet-marketing-services/inter-net-marketing-consulting/online-marketing-strategy(accessed March 2011).

Online File W8.11 The Life Cycle of an Online Advertising Plan

Step 1: Set the Advertising GoalsThe first step is to determine the goals of your online advertising. The plan for increasing brand awareness may be very differ-ent from improving brand image or identifying target customers. Different goals can be achieved through different media anddifferent advertising methods. For introducing a new product, brand awareness may be achieved by banner advertising on apopular portal, but identifying customers can be done by targeted advertising, such as keyword advertising on search engines.

Step 2: Identify Your Target CustomersDifferent online media tend to access different types of customers. For example, comparison of the demographics ofFacebook and Twitter users in 2010 is provided at digitalbuzzblog.com. In order for a marketing plan to be effective, there-fore, it is important to determine who your target customers are before you select the media and advertising methods.In general, market segmentation can be done based on age, income, location, and the social status of the user.

Step 3: Choose the Right Media and Online ToolsThe purpose of this step is to find the more appropriate media and advertising methods that can reach the target customerand deliver the message effectively. This includes putting three types of tools together to build a portfolio (Reed 2011).

1. Content tools. What kind of content you intend to deliver and in what formats? Possible tools include banners, blogs,photo sharing, podcasting, and online videos.

2. Outreach tools. What media (channels) are used to distribute the content? The content may be put on a portalwebsite or using social media such as Facebook and LinkedIn. Media selection must consider the nature of media usersand their costs. The key is how to put together a set of media that can cover our targeted customers within thebudget constraint.

3. Listening tools. What you use to gain market intelligence and customer feedback. Big brands may build social com-munities to receive feedback and influence the conversation and respond properly. Smaller companies may use blogsor their own websites to listen.

Call the media representative and ask for the price information and put the price together with the combination ofdifferent online advertising tools, write a summary report and call an internal meeting with the people who are involved inthe campaign. Consider whether the media is cost-effective for the company. After the meeting, go back to the media repsagain to see if there is space for an alternative media package deal with better price fit for the company. Finalize the pricewith the media reps, and sign the contract with them; make sure to secure the buy (Web Ad.vantage 2011).

Step 4: Develop the Action PlanThe plan of action is an elaboration of the preliminary plan being developed in step 3. It consists of a detailed descriptionof the procedures and means to implement the actions we want to take. For example, the description of an action shouldinclude where the advertisements will be placed, the dates and frequency of the advertising campaigns, a set of proceduresto evaluate their effectiveness. The actions we plan to take must be clearly formulated, measurable, and the results mustbe monitored and evaluated. The implementation plan consists of how to secure the budget and get the action planapproved. At this stage, it is critical to gain the support in the organization.

Step 5: Build Performance Monitoring Evaluation PlanA performance evaluation plan is also critical because we can only achieve what we can measure. This plan may includethe selection of performance criteria, evaluation tools, evaluation period, and personnel who will conduct the evaluation.

One managerial issue is how to measure the effectiveness of online advertising.The resulting advertising plan will include clear and measurable goals, defined target customers, selected media and

advertising tools, plan of action, budgets, and plan for implementation and performance evaluation.

Step 6: Measure Actual Performance and Act Accordingly

Online File W8.1 Online Buyer Decision Making Process...

Documents

Transcript of Online File W8.1 Online Buyer Decision Making Process...