Volume 5, Issue 1, Fall 2001

Change and Response on the Corporate Web Site
Greg Leichty & Stuart Esrock

With the introduction of a new communication technology, businesses and other social groups struggle with how to think about, organize and utilize the new medium (Davis, 1976). The transformation of the "wireless" from a two-way communication medium of radiotelegraphy to the one-way medium of commercial broadcast radio required more than 25 years of experimentation (Schofield, 1994). The introduction of the web page as a corporate communication tool is stimulating a similar process of social innovation and discussion. In the early phases of organizational web site development, many organizations simply desired to get an Internet presence (White & Raman, 1999). Many corporate communicators duplicated existing corporate communication activities by archiving their corporate press releases and employee newsletters online (Esrock & Leichty, 1998).

The current study extended earlier studies of corporate web site content and organization (Esrock & Leichty, 1998, 1999, 2000). It examined several aspects of how large companies were using their corporate signature sites. Corporate signature web sites bear the name of the parent company. As such, they represent and personify the corporation to a variety of publics. Because of this identity representation function, signature web sites usually have a multidimensional character that is not limited to single functions such as advertising or providing product information. The study investigated the degree to which Fortune 500 corporations changed their web site content and format over a one-year period. It also examined how corporations actively responded to public issue inquiries that were submitted through email links on the corporation's signature web site.

Literature Review

Organizations may differ in the kind of communication media that they consider their corporate web pages to be. An organization might treat an organizational website as a "static entity". This type of web page would be a one-way communication tool for conveying relatively fixed messages (e.g., company description, corporate history). Such a web site would function as an informational or promotional brochure. Alternately, corporate communicators might consider the web site to be a dynamic "broadcast" type of medium. This type of website would be a one-way communication tool, but its content would be continually be updated and refined. Web sites also may differ in the degree to which they incorporate interactive features in their design. In addition to its informational and persuasive functions, the web site would be considered to be a two-way communication tool. Customers and other publics could initiate and sustain interaction with the company.

The idea that organizational web sites can be interactive media has generated some attention among scholars. Compared to traditional mass communication media, the web site is seen to offer several advantages (e.g., continuous presence and the flexibility of asynchronous communication) (White & Raman, 1999). In addition, marketers conceive of the web page as an important component of "One to One" marketing that enables a mutually beneficial "learning relationship" between a customer and a company (e.g., Peppers & Rogers, 1997; Pepper, Rogers & Dorf, 1999). Other scholars have articulated web page design and support principles that organizations should follow so as to facilitate more equitable relationships with publics (Kent and Taylor, 1998).

In summary, web sites may differ in the degree of degree of relationship that they enable between the company and the web site visitor. At a minimum, a relationship implies repeated interactions over time with expectations from past interactions structuring the present interaction (Leichty, 1995). A "static" web site primarily supports one-time encounters by the site visitor. In contrast, a more dynamic web site encourages repeat visits because it provides fresh content on an ongoing basis. By providing features such as search engines, calculators, and carefully organized hyperlinks the organization enables the site visitor to customize content to his or her needs (Alexander & Tate, 1999; Flanders, 2000). A web site that facilitates two-way communication enables mutual feedback, response and adaptation. This may engender a higher state of relatedness than more one-way communication tools. A web site can incorporate interactive features such as bulletin boards (Esrock & Leichty, 1998), but the simplest way to add interactivity is to place an email link to corporate personnel on the site (Alexander & Tate, 1999).

The present investigation inquired about the degree of relationship that corporate web sites permit. However, there are several conceptual problems in differentiating between web sites in the degree of relationship that they facilitate with publics. The first problem is that some sites do not fit the simple triad discussed here (i.e., static sites, dynamic sites, and interactive sites). For instance, it is possible that some static sites also have interactive features. More importantly, it is possible that some corporations offer the appearance but not the substance of interactivity. One cannot rely on web site features or content alone to make this judgment. One must also test the actual responsiveness of the corporation to queries submitted by via its web site.

Organizations may not support interactivity because of inadequate mechanisms or inadequate allocation of resources to monitor and respond to messages that are submitted via the organizational web site (Kent & Taylor, 1998). The amount of resources that need to be devoted to support interactivity can be substantial. For instance, Xerox employs four people to do nothing but respond to up to 1000 email inquiries a day (Sterne & Priore, 2000, p. 246). The logistics problems created by email are becoming more acute. The number of customer contacts via email was projected to increase by 250% between 1999 and 2001 (Greenbaum, 1999). A significant disparity between the offer of interactivity in web site design and actual responsiveness to submitted communications would indicate the presence of a cultural lag.

The theory of "cultural lag" can be applied to the introduction of new communication technologies. Ogburn (1964) wrote, "A cultural lag occurs when one of two parts of culture which are correlated changes before or in greater degree than the other part does, thereby causing less adjustment between the two parts than existed previously" (p. 86). The period in between the change of the first part of culture and the ultimate adjustment of the second part of the culture is referred to as the "cultural lag".

Many instances of cultural lag have occurred following the introduction of new inventions or technologies. Technological innovations bring about new possibilities, but they also create many unintended consequences. A "cultural lag" exists when the existing norms and practices of a culture create social and cultural dislocations because they are incompatible with the outcomes fostered by a new technology. A period of difficult "cultural learning" often ensues as the society or social group develops new practices and institutions to deal with the unintended consequences of the new technology. The "lag" between the introduction of a new technology and the development of cultural practices that are harmonious with the new technology can be substantial. Ogburn (1964) noted that automobiles radically increased in terms of the speeds they were capable of after 1910. However, it took more than 30 years of mayhem and increasing concerns about safety to bring about the construction of highways that were compatible with the increased speed and maneuverability of automobiles.

There is reason to believe that "cultural lag" is present in organizational practices surrounding organizational web sites. For instance, a recent FTC study found that only 20% of commercial web sites investigated had adequate standards for protecting the privacy of users, often in contradiction of their advertised privacy policies (Clausing, 2000). With regard to responsiveness to email inquiries, recent proprietary studies suggest that business response to email inquiries by customers is less frequent and slower than comparable inquiries via telephone. Up to 50% of customer inquiries may not be responded to within 48 hours (Porter, 2000; Ryle, 2000). Even the simplest inquiries sometimes generate a lagged response or no response at all. A study by Brightware involved sending an email with the following question to each Fortune 100 company: "What is your corporate address?" While three companies responded within five minutes, 33 of the companies did not respond at all (Sterne & Priore, 2000).

A variety of vendors sell software programs that promise companies powerful response systems that can track all of a company's interactions with a customer across multiple channels (Stearns, 1999; Sweat, 2000a). Early on-line customer service efforts sought to lower costs by substituting automated responses for communication with support personnel. But, more recent electronic support service products have begun to reintegrate support personnel back into customer service by blending telephone and chat into web sessions, combining voice and email queues (Berman, 1999; Sweat, 1999, 2000b). Some writers content that at least the possibility of interaction with a human being is needed to keep customers satisfied (Dawe, 2000). One expert in email management software estimated that less than 1% of companies have a state of the art system to process incoming email inquiries sent via web pages (Riga, 1999).

A company that fails to support the interactive features of its web site may suffer harm (Donath, 1998). The inclusion of feedback features on a web site creates expectations among audience members that their communication will be responded to quickly and relevantly (Sterne & Priore, 2000). E-commerce advisors suggest that on-line merchants give customers response options of one hour, two hours, six hours or 24 hours (Riedman & Cuneo, 1999). They write: "In the absence of an expectation . . . people expect an immediate response (p. 32)." Some previous research suggests that Internet users often have high expectations for response and interactivity and that many specifically frame their messages as if they were addressed to a very specific audience (Newhagen, Cordes, & Levy, 1995). Companies that provide communication links but who fail to respond to audience initiated communications as expected will likely arouse resentment for failing to honor the implied communication contract.

Direct response to inquiries may become a necessity as new institutional pressures develop. Proprietary web sites such as PassengerRights and Ecomplaints have emerged to process and forward customer complaints to congressional representatives and governmental regulatory agencies (Briley, 2000). In addition, the Federal Trade Commission has also monitored the responsiveness of companies to complaints initiated via email. E-Trade, a securities firm, was recently fined by federal regulators for its tardy response in responding to customer's email complaints (Zukerman, 2000).

Existing research into email and on-line interaction has value has several limitations. Proprietary firms have done most of the research on business responsiveness to email. For instance, Jupiter Communications does a quarterly study of "125 top web sites" broken down into the five categories of content, consumer brands, travel, retail and financial services. Unfortunately, relatively few of the methodological details are made available in the press releases and the news stories that result from them. These studies also collapse conceptually distinct categories. For instance, in the Jupiter Communications study reported that 46% of the sites took five or more days to respond to a customer request, never responded, or didn't post an email address on the site in the first place (Porter, 2000, Riedman & Cuneo, 1999). However, it is advisable to differentiate between sites that do not offer interactivity and sites that offer interactivity that they don't support. Only the latter sites involve unfulfilled promises. Proprietary research on email response rates is also confined to inquiries from customers. Customers may be the most important audience for web sites in general, but corporate sites are also designed to meet the information needs of other audiences such as investors, suppliers, the press, and even activist groups (Esrock & Leichty, 1998, 1999, 2000).

Research Questions & Hypotheses

The first phase of the study examined the degree to which the front pages of corporate web sites had remained static for a period of one calendar year. In a previous study in early 1999, the researchers coded the content and hyperlink structure of nearly 90 of the Fortune 500 corporate front pages (Esrock & Leichty, 2000). The researchers printed off copies of each corporate front page. In the present study they revisited the same corporate web pages and coded the degree of change in the content and structure of the corporate front pages. The first descriptive research question for the study was:

RQ1: How static or dynamic are the front pages of corporate web sites over a 12 month period?

The second phase of the study examined how "interactive" Fortune 500 companies were in their response to queries of issues of public interest. The second research question dealt with the presence of interactive features on the corporate web site. The third research question dealt with the actual responsiveness of a company to a query.

RQ2: What percentage of Fortune 500 companies provide an email link on their corporate web site to field general inquiries?
RQ3: What percentage of the Fortune 500 companies providing an email link on their corporate web sites respond to inquiries within one week (5 business days)? For those companies that do respond, what is the typical response time?

The corporate responses to email inquiries were also coded for whether or not they provided a relevant response and/or provided an opportunity for additional contact with organizational personnel. Many inquiries can be handled effectively by simply providing information. However, higher levels of interactivity are indicated when a response identifies an organizational representative that one can continue to communicate with. For instance, a response with a name, job title and department affiliation represents a higher level of "interaction" than an automated computer response. As such, it represents a higher commitment of organizational resources to service and respond to email inquiries. Low interactive contacts would involve automated responses that acknowledged receipt of the message. Intermediate levels of response would involve direct replies that did not identify the correspondent. High-level interactive responses would identify the departmental affiliation of the message response, or give the name and/or affiliation of the organizational representative.

RQ4: What degree of interactivity is represented in corporate responses to e-mail?

We were also interested in several factors that might affect the rate and promptness of response to public issue inquiries. For our first hypothesis we reasoned that companies would be more likely to have established protocols for handling issues related to web site policies and management than they would be to field general inquiries about corporate policies. We hypothesized that companies are most likely to develop response capacities for the most frequent categories of response. In this case we expected that this would involve inquiries about the web site and corporate policies related to the web site (privacy concerns etc.). Inquiries relating to a company's web site may not need to be referred out of the department that supports the web site. In contrast, an inquiry about a company's policies on workforce diversity should tap the degree to which the company has developed a comprehensive information structure to route and respond to email inquiries. Even if the overall response rate was the same for both categories, we anticipated that an inquiry about corporate employment policies would take longer to process because of the forwarding and referral process.

H1: Corporate responses to email inquiries will be more frequent and more prompt when they involve an inquiry about an issue related to the Internet or web site (i.e., company policy on use of cookies) than for a general issue (i.e., the company's policy on workforce diversity).

Finally, we believed that the size of the corporation would likely be related to the overall rate and promptness of the response. At least one past study has shown that large corporations are more likely to incorporate information about corporate responses to issues of public concern on their web sites (Esrock & Leichty, 1998). By extension, they should also be more likely to respond to inquiries about public issues than smaller corporations that receive a much smaller volume of issues related inquiries. In short, we believed that larger companies would be more likely to develop the infrastructure to field and respond to concerns about public issues related to the corporation. Thus they are more likely to meet the demand by creating communication systems and resources to handle such inquiries.

H2: Larger companies will respond more frequently to inquiries about public issues than will smaller companies. Moreover, the mean time of response should be faster in large companies than in smaller ones.

Operationally, this means that company size, as indicated by fortune 500 rank, would be negatively correlated with a tendency to respond and with mean response time.

Methods

A stratified random sample of web pages from Fortune 500 corporations was included in the initial portion of the study. This sample had been used in a previous analysis of front-page content and structure of the web pages of 100 corporations (Esrock & Leichty, 1999). Ninety of those corporations had web sites in 1999 providing the baseline for longitudinal comparisons in the present study. A total of 82 web sites were coded in the March 2000 coding. Eight companies no longer had web sites, had been merged (e.g., Exxon and Mobile) or had gone out of business.

In the earlier study we printed copies of each corporate home page in order to facilitate tracking changes in content and design. In the current study, we revisited these web sites one year later and compared the current home page with that from the 1999 sample. We coded for any noticeable changes in artwork, pictorial elements, text, hyperlinks, and the underlying web site template. In the latter category, in order to be coded as having undergone a change in template, the front page required some degree of basic redesign. We also coded each site for the presence and positioning of email links. Two coders coded 20 sites to check for interrater reliability. The percentages of exact agreement for study variables ranged between 83% and 100% (Mean=94%).

Next, a supplemental systematic random sample of 50 companies was drawn and added to the initial sample for the email response phase of the study (Total sample=132 companies). When a corporate web site was located, it was then scrutinized to determine if there was an email hyperlink on the site. Web sites that did not have any email links (N=21 or 16%) were eliminated from the second phase of the study. Inquiries were sent to a total of 111 companies. Illustrative information on the supplemental sample appears in Table 1.

The companies were randomly assigned into one of two groups differentiated by the issue inquiry that the company received. Each company received an inquiry from a commercial domain email account. The first group of companies received an inquiry as to whether their web site used cookies to collect information on site visitors.1 The second group of companies received an inquiry about diversity policies. 2

Several types of information were recorded about the responses that were received from the companies. The number of days between sending the inquiry and an initial response to the inquiry was recorded. In addition, we noted whether the response came in the form of an automated response, a direct reply with an answer, a direct reply with a query, or some other combination. We also recorded whether or not the responses gave the names and departmental affiliation of the respondent or of an organizational contact person.

Results

RQ1 asked about the degree of change in corporate front pages over a one-year period. We found considerable evolution and change in front page content and structure. Visual elements had changed on 93% of the sites; 95% altered the hyperlinks (i.e., added, deleted or changed hyperlinks); and 96% had textual changes. We also found that 68% of the corporate web sites had changed their front-page template within the 12-month interval.

We also constructed a weighted index of the individual change items. We awarded a score of one if the corporate web site had changed an individual element, and a score of 3 if the site had changed its basic template. Relatively minor changes could receive a coding under the visual, hyperlink and textual changes, but a template change involved a fundamental reorientation or design of the page as a whole. A maximum score of 6 indicated a change on each element as well as a change of the template of the web site. Fully 68% of the sites in this sample achieved the maximum score of 6. An additional 21% obtained a score of 3, which indicated a change scores on each of the individual indices, but not of the site template. Finally, 11% of the sites had changes in one or two of the elements (graphics, text or links), but none of the sites remained the same on all of these features over the year period.

RQ2 asked what percentage of the companies in the sample provided an email link on their corporate web site. For this research question we found that a large majority had email links to field web user inquiries or comments (i.e., 84% of the sites had an email link somewhere on the site of which 75% were on the front page).

RQ3 inquired as to the percentage of companies that would respond to an inquiry and how long such responses would typically take. Of the 111 inquiries that were sent, 52 replies were received (47% response rate), and 51 replies were received within five business days or one week (46% response rate). For promptness of response, we found that two-thirds of the companies that responded did so with in 24 hours. In addition, 83% had responded within 48 hours and 90% had responded within 3 days. The mean response time was 1.76 days (Std. dev.=1.50).

RQ4 inquired about the degree of interactivity offered in the corporate response to the email inquiries. Of the responses that were received, eight (15%) were either acknowledgement letters or form letters. At an intermediate level of interactivity, three cases or 6% included a two-step process that first acknowledged the receipt of the message and promised a later more direct response to the inquiry. The remaining cases showed a more direct manner of response. Forty-eight percent of the responses included the requested information, and an additional 31% of the responses included a direct reply with a query to open dialogue and a promise of follow-up actions or contact (e.g., a promise to send further information).

The degree of interactivity in the corporation's response was also gauged by the degree of personalization in email correspondence. Of the 52 responses, 16 (31%) gave the full name of the responding person and an additional 16 (31%) gave both the person's name, and the person's job title and/or department. The remaining 20 responses were less personal or were incomplete in identifying the respondent's identity. Four responses (8%) were automated responses only. One automated response explained why the corporation was unable to respond to the request: "As you can imagine, we receive numerous requests from all over the globe and unfortunately we cannot comply with all of them." Additionally, nine responses (17%) were clearly adapted to the inquiry, but did not identify a responding person, or give a person to contact within the organization. Of the seven remaining responses, three gave the first name of the responding person, and four gave the departmental affiliation of the responding person.

H1 predicted that the cookies query would generate higher and faster rates of response than would the diversity inquiry. There was partial support for this hypothesis. The response rate was higher for the cookies inquiry (55%) than for the diversity inquiry (38%) (chi-square = 3.3, df = 1, p < .04). However, a t-test showed that there was no difference in the actual promptness of responses to the two queries (t=.03, df=50, p<.98). The mean response times for the two types of inquiry were virtually identical (i.e., diversity M = 1.76 days, cookies M = 1.77 days). A follow-up analysis revealed that there was a significant difference in the type of replies to the different queries (chi-square = 14.57, df=4, p < .01). Responses to the cookies inquiry were primarily in the form of a direct reply with an answer (cookies = 68%, diversity = 19%), whereas responses to the diversity inquiry were more likely to come in the form of a direct reply with additional queries (diversity = 24%, cookies = 7%), or other more elaborated forms of response (diversity = 33%, cookies = 7%).

H2 predicted that larger companies would be more likely to respond to inquiries about public issues than would smaller companies. This hypothesis was partially confirmed. As hypothesized, company size company size, as indicated by Fortune 500 rank, was negatively correlated with likelihood of response (rho = -.28, p < .003). A follow-up analysis showed a 57% response rate from companies ranked in the top half of the Fortune 500 compared to a 36% response rate in the bottom half of the Fortune 500. However, the second half of hypothesis 2 was not confirmed. As expected, corporation size was negatively related to the number of days it took to receive a response, but this correlation was not statistically significant (rho = -.14, p=.34). In summary, larger corporations were more likely to respond, but they did not respond more quickly.

Discussion

Very few corporate web sites were "static". We coded for both minute changes in front page structure as well as more substantial template changes. Very few corporate pages retained the same visuals, text, and hyperlinks. More than 95% of the sites showed changes on each of these elements and more than 90% of the sites registered change on all of these elements. It should be acknowledged we coded for any substantive change that took place on the corporate home page in the intervening year (e.g. any new links or graphics). However, nearly 70% of the web sites had basic format or template changes in the year period, so the degree of site change on average was substantial. Hence, it appears that most corporate sites are maintained in ways that support repeated visits by site users.

The more contradictory and more interesting findings of the study came from the data about web site interactivity. A vast majority of corporate web sites had email links (85%). While some sites were not well adapted for general inquiries because they were addressed to the webmaster or customer service, most sites (76%) prominently displayed an email link on the site's front page. In this respect, a vast majority of corporate sites appear to support interactive communication. However, the email response test revealed a gap between the promise of interactivity and actual corporate responsiveness.

Consistent with the existing proprietary research on response to customer inquiries (e.g., Riedman & Cuneo, 1999), less than half of the corporations (46%) that received an inquiry acknowledged or responded to the inquiry within one week. Although, corporate response to issue inquiries is a little lower than responses to customer inquiries in past studies (Porter, 2000; Ryle, 2000) the difference was relatively small. In short, a majority of the web sites that highlighted interactive links did not actually respond.

It is important to note that the large minority of corporations that did respond to the inquiries, did so quite promptly. Nearly two-thirds of those replying answered within 24 hours and more than 90% responded within 3 days. It appears that corporations either provide interactive support for their web sites inquiries, or they do not. There were very few cases of delayed response.

The results for RQ 4 also indicate that the modal response to an issues inquiry was a response that addressed the substance of the submitted query. Eighty-five percent of the response messages either directly answered the inquiry or promised to do so in a subsequent communication. In addition, than 60% of the responses included the full name of the respondent or contact person, and many of these also included the correspondent's departmental affiliation or job title. Such information gives the inquirer a means to continue the conversation or interaction with an identifiable organizational representative. The offering of such contact information can be read as an additional indicator of support to interactivity.

The results for H1 and H2 also provide insight as to the structural factors that influence how the interactive features of web technologies are handled. Larger companies appear to be more likely to have established procedures for responding to issues inquiries. One can surmise that larger companies are more likely to receive inquiries and thereby reach the critical mass that is needed to justify and create procedures and systems for handling email traffic. Moreover, the results for H1 partially support our reasoning that local inquiries (e.g., inquiries that don't need to be referred outside of the department responsible for web development and support) are more likely to be responded to than inquiries about wider corporate practices (e.g., workforce diversity).

Overall, we interpret the study findings to indicate that companies may experience a "cultural lag" between having a corporate web site with interactive communication features and actually supporting and using those capabilities. Whereas existing research suggests that computer users expect email to have greater interactivity than "snail mail" (e.g., Newhagen, Cordes, & Levy, 1995), they actually may get less response and interactivity. We base our cultural lag interpretation on several facts. For hypothesis 1, we found better response rates in organizations that one would expect to be more institutionalized (larger companies). For hypothesis 2 we found that response rates were higher when the inquiry could be localized in the web support department, thereby reducing the need for coordination across department within the organization. Finally, bimodal distribution of our response data also favors a cultural lag explanation. Corporations tended to give no response at all, or they tended to give responses that were prompt, relevant and often personal. Very few responses were delayed for more than three days. Like a light switch, corporate response in this sample was either on or off.

The current study had several limitations. First, the coding system may have overplayed the degree of dynamism in corporate web sites. The coding system primarily assessed the presence of change, rather than the degree of change. Even small alterations in hyperlinks, text, or graphics were coded as indicating change. In addition, a post-hoc analysis of query type by response showed that the diversity query generated more elaborate responses than the cookies query. Perhaps companies were less likely to respond to the diversity query because they considered it more difficult to answer. There also may be limits on how far the study findings generalize. The results may not generalize beyond large corporations. The pattern of response for smaller companies may be different. Additional research is needed to determine whether or not the descriptive results for this study are sample sensitive, time sensitive and query sensitive.

Conclusion

In summary, this study revealed that most corporations make serious efforts to keep web site content current and up to date. However, it appears that many corporations embrace the interactive features of web sites in word but not in deed. A large majority of signature corporate web sites prominently displayed email links, but fewer than half of these companies responded to public relations related inquiries submitted to their site.

One implication of the findings is that companies need to carefully consider the communication implications of web site design. The results of the current study suggest that a cultural lag exists between putting email links on site and developing the policies and infrastructure to support interactivity. A seemingly simple decision to put an email link on a site has potential ramifications for potential relationships with publics. Web sites with interactive links require additional support and infrastructure. Companies should not provide interactive links in their web sites unless they are willing to support them like they do other modes of communication (i.e., complaint letters). If a company truly wants interactive relating, it needs to make sure that its deeds match its promises.

Works Cited, Endnotes, & Tables

Back to Top
Home | Current Issue | Archives | Editorial Information | Search | Interact