The new marketing intermediaries--a multiple case study of three new e-business models.

ABSTRACT

The past few years have seen a dramatic rise in the use of affiliate relationships to drive Web-based sales. Through this model businesses actively market other business' products and services and in exchange receive a commission. However, there are various ways in which

this model can be applied on the Web. This paper examines and attempts to categorize these new marketing intermediaries through a series of case studies. This research identifies three specific new marketing intermediary models. The first model examines how marketing intermediaries make money by using pay-per-click (PPC) advertising to funnel traffic to their affiliate partners. The second model discusses the mini-site and micro-site approaches, in which a small highly targeted Web site is developed in order to rank well in the search engines and drive traffic to affiliates. The final model deals with a complete business-to-consumer Web site on which all the product links connect the user directly to the affiliate. The paper also examines some of the risks and drawbacks associated with the models discussed.

Keywords

1. INTRODUCTION

Affiliate marketing is a fairly straightforward concept. Using this model an online retailer agrees to pay a commission to those who send it traffic that leads to an action (usually a sale or filling out a form, etc). Thus the affiliate model is often referred to as cost-per-action (CPA). The affiliate marketing model is not new. It is basically the modern equivalent of paying a finders fee. In addition, well-known sites, such as Amazon.com, have been using affiliate marketing for years.

While exact figures are not available, there is evidence for tremendous growth in sites using and taking advantage of the affiliate model. According to Mooche (2005) a recent newsletter from Piper Jafrey indicated that the affiliate marketing model grew at a rate of 30% in 2005.

For online retailers the advantages are obvious. Alternative online marketing outlets, such as pay-per-click or banner ads, cost money whether a customer makes a purchase or not. With CPA advertising, online retailers only pay for a sale (or other qualified action).

The academic literature has been relatively sparse in examining the impact of CPA marketing. Most of the literature that does exist, such as Molander (2005) and Prapatla & Bhatnagar (2002), examines how merchants should choose affiliates. However, the process often works the other way--with affiliates choosing merchants. With the notable exception of Duffy (2005) the academic literature has been relatively sparse in examining the impact of CPA marketing from the affiliate standpoint.

The purpose of this paper is to examine the new ways in which affiliates are driving traffic to other sites. The paper is structured as follows. The next section provides a detailed overview of the affiliate business model. This is followed by an examination of the new marketing intermediary models. Three specific models, pay-per-click intermediary, micro and mini sites, and business-to-consumer (B2C) affiliate site are discussed. Case studies for each model are presented, as well as lessons learned. The last section of the paper contains conclusions and recommendations for implementing the models detailed herein.

2. AFFILIATE MODEL OVERVIEW

There are two main ways in which affiliate programs are run. The first is to use a third party provider (affiliate network), such as Commission Junction, LinkShare, or ShareASale. These sites serve as a trusted intermediary between the merchant and the affiliate. Merchants register with the affiliate network and provide links, banner ads, data feeds, and other advertising. Affiliates register with specific merchants on the network. The affiliate uses the advertising provided by the merchant, or its own advertising to send the merchant traffic. The affiliate network tracks traffic and sales, collects commissions from merchants, and pays affiliates.

Some merchants choose not to use an affiliate network and instead run their affiliate program in-house. Affiliates register directly with the merchant and must also trust that the merchant will correctly and honestly track traffic and sales.

Merchants who decide to establish an affiliate program must consider a number of factors. They must decide what type and amount of commission to pay affiliates. Commissions are usually paid on a sale, but some merchants will also pay for each lead. In addition, commission rates vary widely. For example, commissions for physical products typically range from 5-20%. However, commissions for information goods (e-books and software) are normally in the 40-60% range. Finally, most affiliate programs use cookies to determine if a customer is returning and to track the original source of the customer. Merchants must decide how long to honor these return cookies. The norm varies from 30-120 days. Obviously longer cookies are better for the affiliate.

3. NEW MARKETING INTERMEDIARIES

The new marketing intermediaries make use of merchant Web sites in order to make money. Their main task is to drive traffic to the merchant's site. It is up to the merchant to convert that traffic to sales. The merchant is also responsible for handling delivery, returns, and customer service inquiries.

This research has identified three main new marketing intermediary models. Each of these models is detailed below. In addition, case studies are used in order to understand how these models are applied and the advantages and drawbacks of each.

3.1 Model One--Pay-Per-Click Intermediaries

The following case studies examine the pay-per-click (PPC) intermediary model. In this model the affiliate uses a PPC campaign on various search engine sites, such as Google, Yahoo, and MSN, to drive traffic to the merchant. The key to this model is to pay less for the ad campaign than commissions earned. Therefore it is crucial for the affiliate to know the merchants earnings per hundred clicks (EPC). Fortunately, many third party affiliate networks provide this information. Let's examine how this works through two real cases.

3.1.1 Case One--Bean Bag Games

The first case involves the sale of outdoor beanbag games. The merchant was registered on one of the major affiliate networks. The affiliate's ad campaign began in April 2006. At that time the merchant had an EPC of about $30. That means an affiliate could expect to earn $30 for every 100 visitors sent to the site. The EPC is a factor of the overall conversion rate, the price of the products, and the 15% commission earned on each sale.

The affiliate began by researching relevant keyword terms on Google's AdWords PPC network. The affiliate determined that ads with good visibility would cost $0.05 per click. Thus the affiliate would expect to pay $5 for one hundred clicks, but expect to earn $30 for a profit of $25 for every one hundred visitors.

Over the course of two months the affiliate sent 529 visitors to the merchant. The 529 visitors resulted in 12 sales for earnings of $165.29, for an EPC of $31.19. On the cost side the affiliate spent $20.09 for PPC ads. So over the two months period the affiliate earned a profit of $144.30. Obviously, this is not a lot, but many affiliates who use this model do so for hundreds or even thousands of merchants.

3.1.2 Case Two--Herbal Remedy

The second case of PPC intermediation involves an herbal remedy. This merchant had an EPC of about $63, more than double the previous case. Again a Google PPC campaign was used. The affiliate determined that $0.10 per click was required in order to achieve good ad visibility. Over the course of a month the affiliate sent 53 visitors to the merchant. Based on the EPC of $63 we would expect the affiliate to earn about $33 over this time period. However, the affiliate only had one sale and only earned $6.74. In addition, the affiliate paid $6.89 for the PPC campaign (a cost of $0.13 per click). Obviously, this campaign was not profitable and was eventually dropped.

3.1.3 Pay-Per-Click Lessons Learned

There are a number of important lessons that can be learned from these cases. First, while EPC is a useful measure, actual results may vary widely. The main problem is that EPC is an aggregate measure and affiliates do not know how other affiliates are driving traffic to the merchant. Traffic from different sources may have very different conversion rates. In addition, consumers respond differently to various ad campaigns. Therefore, experimentation is important when using this model. Affiliates should set a budget for each merchant and try various ads on multiple search engines to determine which, if any, work best. Campaigns that are not profitable should be dropped. Increasing the budget, adding new key words, and using multiple search engines can expand profitable campaigns.

Pay-per-click intermediaries must have a good handle on their costs. All of the major search engine PPC systems allow users to set daily and monthly budgets. In addition, affiliates must be knowledgeable about how bidding works on each system. For example, Google Adwords allows users to bid on a per click basis or the user can decide to allow the system to automatically bid in order to ensure the ad appears in a certain place. Both Google and MSN allow users to set certain times and days of the week to run ads. Understanding exactly how each system works allows the affiliate to gain the maximum exposure at minimum cost.

Finally, it should be noted that many merchants place restrictions on PPC ads from their affiliates. For example, some merchants will not allow affiliates to bid on their trademarked name. Other merchants do not allow any PPC ad campaigns at all. PPC rules are usually stated clearly, so affiliates who want to use this model should only register with those merchants who allow PPC.

3.2 Model Two--Mini-Sites and Micro-Sites

The second new marketing intermediary model involves the affiliate building a small Web site that contains links to merchant sites. These sites contain anywhere from one to a few dozen pages. They are typically targeted toward a small niche market. Affiliates using this model usually use search engine optimization (SEO) as their main marketing mechanism. SEO works well, since the sites are aimed at a niche market with little competition. As the cases below indicate, we have identified two main types of sites in this category--the mini-site and the micro-site.

3.2.1 Case Three--Hoodia Mini-Site

The mini-site typically contains a few to a few dozen pages. These pages typically provide detailed information about the product or service offered. Hoodia Gordonii is an appetite suppressor and is widely marketed as a weight loss pill. A large number of merchants with affiliate programs offer Hoodia via their Web site. The Hoodia micro-site in question consists of about a dozen pages. The main page provides some basic information about Hoodia and before and after pictures of those who have lost weight using the product. There are also pages for testimonials, frequently asked questions, history of the product, dosage, and independent research. Throughout the site are links to affiliated merchants selling Hoodia. In addition, this affiliate included Google AdSense on the site to increase revenues. Google AdSense is Google's affiliate program. It enables Web site owners to show context appropriate Google ads on their site. If a user clicks on an ad, the site owner shares the revenue generated with Google.

The affiliate partnered with a merchant that offers a huge 40% commission on Hoodia sales. The merchant runs the affiliate program in-house. The average bottle of Hoodia sells for about $50 (for about $20 per bottle to the affiliate). The site uses search engine optimization (SEO) as its main marketing mechanism. It has achieved high search rankings on Yahoo and MSN. The cost of the site is only $3.95 per month for hosting. During a recent month the site sent 384 visitors to the merchant's site, which resulted in 3 orders for a total commission of $73.94 (one order was for multiple bottles). While these results represent a healthy return on investment, the conversion rate of only 0.07% seems low. This has led to concern on the part of the affiliate that the merchant is not crediting his account for all purchases.

3.2.2 Case Four--Key Word Software--Micro-Site

Where the mini-site usually contains a number of pages, the micro-site is essentially one long page. This page is used to pro-sell the product--convince the customer to purchase the product. The micro-site is an extremely popular method for selling information goods, such as e-books and software. Most micro-sites have a similar structure. They usually begin with the main marketing claims in big bold type. Somewhere further down the page the elements include a checklist of major features, customer testimonials, and the offer. The site usually offers a discount over the normal price of the product and a number of freebies if you order soon. Of course there are links to a payment page.

The micro-site examined for this case sells a software product that allows the user to research key words for search engine optimization and pay-per-click advertising. The software sells for $167 and the affiliate earns a 60% commission, or $100.20 per sale. The merchant uses ClickBank as the third party affiliate network. ClickBank specializes in information products.

The micro-site was structured with the main marketing message in bold at the top of the page. The site included a somewhat unique feature--video clips of the main features of the software. In addition, while the software sells for $167 on ClickBank, the site states that the actual price is $297. This allows the site to claim that the customer can purchase the product at a substantial discount.

The hosting cost for the site is $12.95 per month. The affiliate used a combination of PPC ads and search engine optimization. The PPC campaign cost about $2 per click to achieve good ad visibility. Over the course of a recent month the PPC ads were viewed by 78 people and clicked on by 4 of them (for a click through rate of 5%). The cost for the PPC ads was $6.48. The site also received 22 visitors from other sources. During this month the site did not generate any sales--for a loss of $19.43. Despite the loss the affiliate decided to continue with the campaign based on the low cost and potential upside.

3.2.3 Mini-Site/Micro-Site Lessons Learned

One of the primary lessons learned from these cases is that affiliates must align their marketing approach with their merchants' products. While micro-sites work well for information goods, they are not as useful for physical products.

Case three points out the problems associated with affiliate programs that are run in-house. The affiliate is concerned that the merchant has not credited some of his sales. Unfortunately, there is no way for the affiliate to track sales on a merchant's site. Therefore, it is important that affiliates choose wisely when picking in-house programs. In addition, affiliates using in-house programs might turn to friends or relatives to secretly shop the merchant to see if the sales are properly credited.

3.3 Case Study Three--B2C Affiliate Site

While the mini or micro site has a very narrow focus, the B2C affiliate site typically has a broad focus. The idea is to offer thousands or even tens of thousands of products on a single site. The B2C affiliate usually combines products from various merchants in order to achieve this goal. The B2C sites looks like a typical online storefront, with a menu of categories, product pictures, prices, descriptions, etc. However, when a user clicks on a product, instead of providing more information or adding the product to a shopping card, the user is taken to the product's page on the merchant's Web site.

3.3.1 Case Five--Vitamin and Nutritional Supplement Site The case for this category is a vitamin and nutritional supplement site. The site used two main merchants for the majority of the products offered. Both merchants offered about 6,000 products on each of their sites. The affiliate used a data feed from each merchant to quickly populate a database for the new site. Therefore, the affiliate's site was able to offer over 10,000 unique products (there were a lot of product overlaps between the two merchants).

In addition, the affiliate registered with a seven other sites that offered proprietary supplements. For these merchants the affiliate primarily used banner ads on the site's homepage to market the merchants' products. The affiliate also decided to add Google AdSense to the left side menu, so it would appear on all pages on the site.

The affiliate used search engine optimization (SEO) as the main marketing mechanism for this site. After three months the site was ranked on the first page for the term "discount nutritional supplements" on both Yahoo and MSN. With over 10,000 products the site contained a huge number of pages, which helped with the SEO effort. While the site could use PPC ads, the affiliate determined that many of the most widely searched terms would cost too much per click for good ad visibility. For example, good visibility (position 4-6) for the term "vitamins" would cost at least $1 per click. With an average EPC of about $20, paying $1 per click would result in an average loss of about $80 per one hundred customers.

The cost associated with this site is $6.95 per month for hosting and $5 per month for membership in a link exchange network, for a total of $11.95 per month. During a recent one-month period the site earned commissions from its main merchants of $86.11. Commissions from other merchants were $6.72 and Google AdSense revenue was $1.87 for total revenues of $94.70 and a profit of $82.75.

A further analysis revealed that one of the two main merchants had not generated any commissions. This was unexpected as the non-performing merchant had an EPC twice that of the other merchant. A deeper investigation showed that the performing merchant offered more popular items, such a weight loss products. In addition, the well performing merchant made excellent use of points-of-action (PoA) messages, such as a "Shop With Confidence" box that appeared on every product page. The box reiterated the merchant's money back guarantee and site security policies.

3.3.2 B2C Lessons Learned

The basis of the B2C affiliate site is offering as many products as possible. Thousands or even tens of thousands of products are the norm. Therefore, it is imperative that affiliates who use this model choose merchants that provide a data feed of their products. In addition, affiliates should be aware that different affiliate networks may charge a fee for access to product data feeds.

Affiliates should also be aware that not all data feeds are created alike. The best data feeds provide various categories for each product. Some data feeds do not provide any categorization, which makes finding specific products very difficult.

4. CONCLUSIONS

Web-based affiliate programs have given rise to new marketing intermediary business models. This paper provides an initial analysis of these new models. It identifies three main types of new marketing intermediary and some of the critical success factors for each.

While the cases presented herein show relatively little profit, we must remember that many affiliates use automated tools to quickly implement their ad campaigns or build Web sites. They then repeat this process hundreds or even thousands of times per month. For example, case one--beanbag games--made about $70 per month. However, we must consider that this campaign only took about 10 minutes to setup. So the astute affiliate can easily implement dozens of this type of campaign a day.

One of the main lessons learned from all of the cases is that affiliates must look beyond the standard metrics when choosing a merchant program. Many new affiliates focus solely on the commission rate. All affiliate networks and in-house programs provide commission rate information, so it is very easy to compare programs using this metric. However, there are many instances where a merchant with a 5% commission outperforms one with a 20% commission. Even more detailed metrics, such as EPC, are not always an accurate predictor of success.

Affiliates must align their choice of merchants with their preferred new marketing intermediary model (see Table 1). An affiliate who primarily uses mini-sites would want to choose merchants who sell information products. Affiliates must seriously consider the merchant's ability to convert traffic to sales. Reviewing merchants' Web sites like a potential customer may reveal certain problems. Affiliates should ask themselves if they would buy from the merchant.

Affiliates need to decide whether they will register with in-house programs or exclusively use third party affiliate networks. Given the large number of merchants available on third party networks and the risks involved with in-house programs, there seems little advantage to affiliates using in-house programs. However, in-house programs may provide higher commissions. An empirical examination of in-house versus third party affiliate programs is a ripe area for future research.

All of the models discussed require experimentation and monitoring. Affiliates should set an initial budget and profit goal for each campaign or Web site. If profits fall significantly below the goal, a new method may be used or the merchant dropped entirely.

REFERENCES:

Duffy, D.L. (2005). "Affiliate marketing and its impact on e-commerce" The Journal of Consumer Marketing, Vol.22, No. 2/3, pg. 161.

Molander, J.G. (2005). "The Evolution of Affiliate Marketing" Target Marketing; Oct 2005; 28, 10; pg. 115.

Mooche, F. (2005). "Online affiliate marketing to grow 30%" http://www.axcessnews.com/technology_070505.shtml.

Papatla, P. & Bhatnagar, A. (2002). "Choosing the right mix of on-line affiliates: How do you select the best?" Journal of Advertising; Fall 2002; 31, 3; pg. 69.

Ross A. Malaga, Montclair State University, Montclair, NJ, USA

Dr. Ross A. Malaga received his Ph.D. in 1998 from George Mason University. He is currently an Associate Professor of Management and Information Systems in the School of Business at Montclair State University and serves on the editorial review boards of the Information Resources Management Journal and the Journal of Electronic Commerce in Organizations.

Table 1--New Marketing Intermediary Summary

New Marketing   Best Used For     Main Marketing
Intermediary                      Approach
Model

PPC             Information and   PPC
                Physical
                Products

Micro Site      Physical          SEO
                Products

Mini Site       Information       SEO and PPC
                Products

B2C Site        Physical          SEO and PPC
                Products

New Marketing   Critical Success Factors
Intermediary
Model

PPC             Ensuring PPC costs are
                considerably less than true EPC

Micro Site      Provide good content to support
                SEO effort and re-sell product

Mini Site       Strong marketing content to pre-
                sell the product

B2C Site        Large number of products--
                supports SEO effort and provides
                a one stop shop