Response assessment in the Web site promotional mix

Domenico Tassone, February 1997

The Lilypad White Paper is the first in a series of short White Papers addressing the Lilypad software product and its use. Forthcoming topics will address technical issues and include case studies for the purposes of demonstrating how evaluating the effectiveness of Web site promotions adds value.

The purpose of this first White Paper is to provide Web site managers, and those directly involved with Web site promotion, with a fundamental approach to measure the performance of Web marketing efforts. Lilypad will provide more effective Web site promotion through better quality information. This higher quality of information will assist Web site promotional decision making.

Response is not the only means by which to evaluate Web advertising - it's just the easiest. An effective and long range Web marketing effort should include both response and "awareness." This combination can positively impact their short-term and long-term marketing objectives. Similar to evaluating the intangible value of awareness in the traditional advertising industry, measuring the effectiveness of Web advertising is a much more complex undertaking than simply measuring response. Just because it is easier to count responses does not mean it is more important; this is the first fallacy of Web site promotion.

The importance of awareness in any brand communications program is likely to remain significant despite opposing views. Its intangible value often confuses and confounds many well meaning marketers that view everything in terms of ROI. Often this short-sighted approach ignores the fundamental value of advertising. Nonetheless, the computer-based Web inherently allows for census level data gathering and dynamic reporting of banner ad response, the exciting realm of Lilypad.


"Web marketing" is a misleading term, similar to "television marketing." The very notion of "Web marketing" inherently creates confusion in the minds of academics, practitioners and the media that concern themselves with the subject. Hundreds of new media consultants, and others particularly in the "below the line" marketing communications business have huddled around the Web and its promise of "convergence." These new consultants almost always slam traditional advertising media and agencies in the process. Despite having the most power (commissionable media budgets and firm size) relative to direct marketing, public relations and sales promotion firms, traditional media advertising agencies have been the slowest to get on the interactive bandwagon.

The last three years have revealed the growth potential of the Web and its harnessing by businesses for numerous marketing purposes. The Web is not just a medium, it's also a self-contained marketplace. It perpetually changes and is slowly becoming more representative of a diverse global market. Savvy Web marketers whose product or services appeal to the demographics of the Web have done quite well. Other businesses who are simply enamored of the technology, and sales pitches. have found quick success elusive.

Some activities conducted by Web marketers include extensive online market research, online sales transactions, and communicating with their target audiences in a whole new way. Since the nature of the Web is computer-based, many processes which traditionally required human assistance are now automated. Unfortunately, this phenomenon alone has sprouted a cottage industry of internet marketers who ignore sound marketing and advertising principles. To use a traditional analogy, they often "let the tail wag the dog."

Simply because a new medium is widely and easily available should not be the key issue that drives marketing decision-making. All too often, business executives forget the fundamentals of marketing and communication when they listen to the technical gurus who chant in unison, "Advertising is dead!" The first observation of Web marketing recognizes that although the technology may change, the needs of the marketer will not.

There are a number of attributes for success on the Internet. The most fundamental marketing processes such as research, segmenting, targeting and careful integration within the marketing communications mix are still pertinent for success. For many marketers it will be a while before the mass market, or any particular niche is reachable online. In the meantime, many firms are learning right now what works and what doesn't, hoping to exploit their experience in the near future when the mass market arrives.

Since 1994, the Web has expanded considerably from a subset of the consumer mass market - higher income, well educated, white males in technical professions. But it's still interesting to note that marketers providing offerings to this target market are quite successful with products and services which meet the needs of this group.

There are a handful of new products now available to help the less technically-oriented consumers to get online for minimal cost. All that's required are a TV and a telephone line. Traditionally, new technology takes a long time for the general population to assimilate it into their lifestyles. The acceptance of the Web has been unprecedented, however, critical mass has yet to be achieved.

Although consumer electronic commerce is held to be the "holy grail" of the Internet by marketing professionals, this perspective is far too myopic. There are lingering consumer concerns about credit card security. This adds resistance to successfully facilitating online purchases. The Web will be a dismal failure and a money pit for a good while, if one evaluates effectiveness strictly from a consumer sales standpoint However, business to business marketers are successfully using the Web as their own customizable Wide Area Network (WAN) and Electronic Data Interchange (EDI). Business customers and vendors can easily transact business online right now, reducing time and administrative paperwork substantially.


There are any number of measurements of the activity of a Web server. Any of these measurements are of concern to the Web site marketer. But the unit of interest that is of primary concern to the Web site promoter is by far the undifferentiated "visit." All other measures are of secondary importance; a fabulously designed Web site with an online catalog is of little value if no one visits the site. Only the visit can help the marketer quickly understand how well their promotional efforts have fared, which is often practiced with trial and error experimentation.

There are a few differing definitions for the term "visit." For our purposes, CASIE defines the visit as "the number of occasions on which a user looked up X site or Y domain during Z time." Similar to a gross exposure, the visit is usually defined as a Web user requesting a Web page. As such, the gross exposure measurement is a good indicator of response to Web site promotional efforts. These visits can be caused by different means - commercial Web hyperlinks, editorial or goodwill hyperlinks, Newsgroups, bookmarks, or email.

Once an end-user visits a Web site, this activity is recorded by the destination Web server as a file request. This provides the Web server a good deal of information about when the request occurred and to which Web server the file is was routed to. Unfortunately, most tracking software gets lost in the easily accessible log data, in hopes that by mining this information great wisdom can be found. The second observation of Web site promotion is to limit the value placed on server log files.

For the Web marketer's purpose, a visit indicates that a Web user has visited their Web site for some reason. Why and how this occurred is of extreme interest to observant Web marketers, simply because it helps benchmark their Web site promotional efforts. It answers the oft posed question, "How did they get here?" What they do after they arrive is of limited value, and is best left to Webmasters and content developers to interpret.

Since many Web visits are not directly caused by Web site promotion and are therefore "free" or "goodwill" they are of nominal value to most Web marketers. When an advertiser is trying to enhance traffic levels for some marketing objective and has committed resources to doing so, results for paid efforts are far more important - decisions need to be justified. Determining and analyzing visits is a means of determining what has worked and what hasn't. Adjustments and refinements to their approach will likely need to be made.


For the purposes of Web marketers there are three kinds of Web sites: marketer, commercial content-provider and all others. The marketer's Web site usually has a purpose - electronic commerce, corporate branding, prospect generation, public relations, and all too often plain corporate hubris. Commercial content-providers are Web media suppliers, akin to television networks, cable programmers, radio stations or magazine publishers. For the most part commercial content-providers, or Web media suppliers, are selling banner advertising and promotional opportunities. The others are generally personal, government, non-business related or academic Web sites which have little or no bearing on the success of most marketer's Web sites.

A common-sense objective which Web marketer would have such as maximizing traffic, might also carry some traditional media purpose like brand awareness as its objectives. This can create objective incongruities as we'll find later on. Yet, banner ads typically lead to a Web marketer's Web site; further subdivided into yours, your competitors' and other non-competitive marketer Web sites.

Labor intensive methods for getting traffic to the marketer's Web site include a myriad of approaches: search engines and directories, link exchanges, online editorial, Newsgroups, Email lists and others. As might be expected, the most dependable source of consistent levels of traffic are through relatively small graphics called banner ads. This new form of advertising can be found in the content-provider's Web Sites, charged with the big job of getting your attention, stimulating your interest, enhancing your desire and putting forth a call to action.

Many Web marketers are unsure which commercial content-provider's Web sites to sponsor, i.e. which Web media to place their online media purchases with. Although they seek to enhance traffic levels to their Web sites, often only by trial and error do they soon learn which Web sites deliver the best response to their particular solicitations. Building a stronger presence in these high clickthrough Web sites helps maximize a response-oriented promotional budget.

However, getting to this point requires an in-depth and ongoing analysis of the reports provided for the most part by the Web advertising media themselves. In traditional media advertising, this "conflict or interest" is mediated by the existence of independent third-party auditors and research firms like the Audit Bureau of Circulations and Nielsen. Similarly in the online media, these firms are providing assurance that the Web media which they are auditing are providing accurate data concerning awareness and even clickthroughs. However, often a "comparing apples to apples" approach is impossible.

Simply put, the standards for reporting this information just do not exist yet. To aggravate this situation from the Web marketer's perspective, some Web media suppliers offer clickthrough information while others do not. Some offer packages based on time, while others by CPM. Some Web media are audited and others are not. There is no standard yet among the auditors themselves. Some Web media sell by banners delivered, clickthroughs or both. This can easily overwhelm Web marketers attempting to execute controlled experiments for maximum response.


Many Web marketers are finding that simply putting up a vending machine in cyberspace does not an effective Web marketing plan make. A marketer's Web site is akin to collateral or a catalog in the traditional media world. Without appropriate Web site promotion, a marketer's Web site remains unaccessed somewhere in a digital file cabinet. Because there are so many Web sites online today, it is foolhardy to expect the Web public to inherently be interested in any one marketer's offerings, and beat a path to their respective door. For many Web marketers, the moment of realization is the short time period after the Web site is launched and nothing happens - just a deafening silence.

As in traditional media, clutter pervades the Web. A sound Web site promotional strategy is needed to ensure that the Web marketer's site does not fall on deaf ears. Like traditional marketing, getting your message to the right people at the right time is as much an art as it is a science. However, there are some fundamentals which can be applied to Web site promotion. Web sites can be promoted off line by supporting the Web site with Uniform Resource Locators (URL), in other advertising media, letterhead, events, press releases, print collateral and the like. To some Web marketers these efforts are one and the same. That is, they fail to make the distinction between their corporate Web site and a product Web site, the second fallacy of Web marketing.

In the online world the basics and goals are the same, only the technical execution differs. Where response is a key component of benchmarking the effectiveness of a particular effort, many promotions can be closely tracked. Often Web marketers experiment to find the right combination of creative, media placement, database parameters and the like to achieve optimal results.

Let's first examine more closely just how Web users come to visit any Web site.


Empirical observations suggest that many of the same Web users regularly visit content which they enjoy or find informative. Such traffic into a Web site is caused in several different ways, but can generally be broken into two psycho-behaviorial groups - casual and intentional browsing. In terms of casual browsing, this is accomplished several ways including clicking on Web site hyperlinks, Newsgroup embedded URLs and email embedded URLs. Note that not all Newsgroup or email software packages allow for hyperlinking, although this appears to be a growing trend. Intentional visits can be subdivided into visits caused by bookmarks, direct accesses and local HTML files, all of which suggests premeditated cognition towards reaching the desired destination.

Depending on the Web site and the content offered therein, there are a few other means by which Web users might encounter an advertiser's Web site. First, a number of unpaid "goodwill" hyperlinks will inevitably aggregate over time, in somewhat of a grass-roots manner. As Web users learn of your Web site, those with a particular affinity for it might place hyperlinks on their own Web sites as well as bookmark it in their browser. Web sites are also subject to visits by automated search engine spiders, which routinely scour the Web indexing all the Web pages in their path, meticulously checking each link. Sophisticated Web marketers can harness these phenomena.

By and large, traffic caused by hyperlinks from other Web sites appears to be the lion's share of site activity, especially for Web marketers. The theory that banner advertising works for the purposes of direct-response Web promotion (thereby potentially reinforcing its effectiveness) actually depend on the destination site.

Conversely, content-providers seem to show a higher relative propensity to attract premeditated visits. This supports the assertion that content is by itself a major attraction, and creates a "pull" opportunity to informative and entertaining Web sites. Experienced Web users can easily bookmark or save their favorite Web site destinations, thereby avoiding the tedious manual entry of what are often lengthy file paths and strange file names. Bookmarks eliminate the possibility of entering the wrong Uniform Resource Locator (URL.) The bookmark can only be created after a user has already visited a site and decides if they liked the content, or might return in the future for any reason.

Another method of eliminating the time consuming step of entering URLs is evidenced by the use of local HTML files. At its most basic level a local HTML file is usually just a Web page consisting of someone's bookmarks. Having a local file facilitates entry of URLs especially for shared computers with browsers such as schools or libraries. Shared computers have the potential to create an enormously unmanageable list of bookmarks. As a result, many computer lab administrators require that users save their files on the local hard drive, or their own personal floppy disk. In other cases, Web users may learn of a Web site from another medium entirely. With the popularity of the Internet one can now find URLs everywhere -television, print advertising, radio, billboards, napkins, the sides of vehicles and even popcorn containers at the movies. Amazingly enough, even telephone conversations can provide URLs.


The most prevalent form of Web promotion, and the most successful, is the banner ad. Although small, the banner ad is ubiquitous as an easily commodifiable form of exposure which can be bought and sold with great ease. In execution however, the banner ad is far more complex than many believe. Often trumpeted, the direct-marketer's approach typically criticizes the value of brand advertising for the more tangible promise of response simultaneously deprecating traditional media and awareness. Far from perfect and capable of much more than they are given credit for, banner ads will continue to evolve as the dominant unit of Web advertising media for some time. For our purposes, we will focus strictly on the response to banner advertising.

Banner advertising has many critics that correctly cite its size, purpose and pricing as invalid and too expensive. Conceptually, somewhere between a print ad and a billboard (some with limited animation), banner ads have had mixed success. Often within the same banner ad there are conflicting goals. The reality is that banner ads in their current form, can do one of two things very well - communicate a brand message or get the viewer to click on it. Banner advertising which achieves great clickthroughs usually defeats any brand objectives, possibly even causing damage to the brand's image. Furthermore, these are mutually exclusive propositions, i.e. you can do one or the other but not both very well. The third observation of Web marketing is that a banner ad is either for response, or for awareness - but never both.

When the direct-response promotional approach is utilized, getting to the clickthrough is paramount. Free to achieve just that, banner ads can truly bring home the bacon. Performance can be tracked and analyzed to determine what happened. Adjustments to banner creative and Web media placement can be made to attain optimal results. For the purposes of attracting users to a Web site for a specific promotional purpose, e.g. giveaways, banner ads can be quite useful in the Web site promotional mix.

For new products, product repositionings and other instances where a brand awareness approach is warranted, banner ads can be optimally utilized. Unlike the response approach however, measuring the success of a brand awareness effort on the Web is just as complicated as doing so with traditional advertising media. Both, however, still require special research. Awareness data beyond gross impressions delivered cannot be captured on the Web server - it's in the mind of the end-user.

As stated earlier, the first fallacy of Web marketing posits that the ease at which Web browser behavior can be tracked is not necessarily proportional to the importance of the information to the Web marketer.


There appears to be five distinct approaches to Web site promotion, particularly in terms of banner advertising. The diagram depicts several models which represent what happens after Web users encounter Web marketer's advertising message and then sequentially flow through to key interactions with the marketer online and off. As suggested, no single model will optimize marketer awareness and response objectives.

Note, that for many Web marketer's electronic commerce is simply not an option or does not make any sense. For example, automobile manufacturers and quick-service restaurants aren't likely to sell cars or sandwiches online. However, as a medium most marketer's recognize that the Web can certainly facilitate the realization of many marketing communications objectives just short of electronic commerce.

The key criteria in determining which model is being used include: purpose of banner ad, time lapsed from first exposure to banner ad and next Web site visit, and the fulfillment system. By design, these models do not address the potential of dynamic Web-enabled database marketing.


Figure A depicts the Direct-Response Model, which is most closely related to direct-response advertising. It is based on the unlikely supposition that the product or service offered in the banner ad will immediately lead to an immediate online sale. For most Web advertisers this approach ignores reality, it is unlikely that a Web user that casually encounters a marketer's banner ad will be ready, willing or able to buy immediately after being exposed to an offer for the first time. The technical ease or capability, to track a prospect through to the purchase decision has created what might be called the fourth fallacy of Web site promotional expectations.


In Figure B we find the Online Awareness Model, an approach which has a clear and simple goal - awareness. The purpose is to seed commercial Web media to create awareness of a specific product or service, primarily for branding purposes. Similar to traditional advertising, it is upon these banner ads that many Web users will first form opinions and attitudes about the marketer's offerings. Online fulfillment is expected at some later date.


Figure C illustrates the Promotion-Response Model, which acknowledges that electronic commerce may be an option for the first time visitor, albeit unlikely. The target audience is induced to click on the banner ad to reach the destination Web site for some immediate gain or gratification, e.g. a free chance, a coupon, a contest, or even a game. The prospect may either continue online to execute a purchase or visit a local participating retail location for fulfillment.


The 3-Step Offline Awareness Model demonstrated in Figure D is one of the most popular with traditional major brand advertisers presently experimenting with Web media. Often a traditional media agency is involved which suggests that response to the banner advertising, or the destination Web site is not a great concern. Although partially responsible for the growth of commercial Web media who are happy to sell them Web media placements, to the Web marketer this approach does not maximize the potential of the medium and is therefore of very limited value.


Finally, Figure E represents the 2-Step Offline Awareness Model which is noticeably quite rare today. Early on, many Web marketers executed banner ad buys with no destination Web site to speak of. Instead, these Web advertisers had a microsite developed by and located on the Web server of the Web media supplier who they purchased sponsorships from. Offline fulfillment was usually the only option available.


In conclusion, Web marketing has come along way from its restricted beginnings and primitive origins. Remarkably, the banner ad is still with us and still evolving.

There are many questions which are currently unanswered about the effectiveness of Web site promotions. Given the models suggested above, a combination of awareness and response oriented promotions appear to be the keys to successful Web site promotion - particularly in terms of banner advertising.

The purpose of this White Paper (and future ones as well) is to stimulate further discussion of Web marketing among academics, practitioners and the trade media. All comments, concerns and questions are welcomed.



© 1995-97 Streams. "Lilypad" and the Lilypad logo, "Streams" and the Streams logo are our trademarks. Honest. All rights reserved.
phone 773.342.7747