Chapter IV

Results

Survey

A Web site was established during the month of July, 1997 for the purpose of displaying an electronic survey form, and to subsequently capture participant results electronically. These results were stored in an MS-Access database for future query analysis. As previously mentioned, the survey was conducted as part of the Editor & Publisher annual survey of media companies.

From a list of 2,965 e-mail addresses obtained from the database of Editor & Publisher, 572 were invalid, leaving 2,393 potential respondents. The survey was conducted from 8/6/97 through 9/25/97 and generated 387 responses, or a 16.2% rate. This is far above the 10% suggested response rate of Leedy (1997). Of the 387 respondents, 243 (62.8%) newspapers, 72 (18.6%) magazines, 19 (4.9%) radio and 51 (13.2%) television stations responded. Two respondents fell into the "other" category as they did not indicate their type of business. It is not known how many of the 2,393 were newspapers, magazines, radio or television.

The survey ended in late September, 1997, and respondent data was analyzed at that time using a statistical software program (SPSS), as well as through various crosstab queries in MS-Access. The data was then analyzed for descriptive statistics, inferences, correlations, and trends. Special attention was given to the peculiarities among the different media companies to determine the existence, stages, and levels of competition. The critical issue is how the four different media groups are approaching and dealing with the web phenomenon.

The survey was divided into nine major sections. Within each area, a number of questions were asked in order to develop a profile of the Web site, and to obtain raw data for analysis. Questions from the nine areas are summarized as follows:

    1. General: The type of business, circulation level (if a newspaper), distribution of customer base (local, national etc.), and length of time on the Web was determined.
    2. Technology: The purpose of this section was to find out who operates their own server as opposed to outsourcing, who provides Web services to outside companies, the type of authoring tools used, and if any type of "push" technology is being used.
    3. Editorial/Content: A number of items were considered here, including referencing the Web site from the core product, ability to correct archival information, use of the Internet by the editorial staff, integration of the newsroom staff with the Web staff, types of content, scooping the core product, providing archival information, number of on-line articles available, and alliances with other companies.
    4. Advertising: Questions from this section provided and insight into which Web sites have paid advertising and/or sponsors, standardization of banner ads, average cost of a banner ad, outside measurement, page views per week, and whether or not they offer Web classified ads.
    5. Promotion: Important issues here were the types of promotion vehicles used by the four media groups, cross promotion, the size of the promotion budget, and Internet tactics used to bring in traffic.
    6. Staffing: Questions from this section asked how many staff members, full and part-time, are being used for each Web site. This information is subsequently used to determine a variety of things such as profitability, staff size by page views, staff size of Internet providers versus non-providers, and average staff sizes of the four media groups. Also asked was the anticipated change in staff size for 1998.
    7. Subscriptions/Transactions: The main focus of this section was to find who, if any, are charging for access to their Web site. For those that are charging, respondents were asked how much they charge and how (i.e. daily, monthly, yearly, per search per article) they charge. Additional questions included charging for archival information, charging in the future, and Web site registration.
    8. Sales and Profitability: Questions in this section included Web site expenses (less salaries), gross sales for banner and classified advertising, as well as subscription and transaction sales. An important question here was how do the four media groups rank their competition?
    9. Other: Although this section included a variety of questions, the most important was to determine who, if any, are capturing user’s names and addresses, and how they plan to use this information (promotions, selling lists to outside companies, etc.)

Data Analysis

As mentioned previously, the survey instrument was validated by ten representatives from the media industry. Because of their knowledge and experience in the industry, their input led to categorizing numerous questions. For example, news features referencing a Web site were categorized into daily, weekly, bi-weekly and monthly. Likewise, promotion budgets were categorized into less than $10,000, between $10,000 and $50,000, between $50,000 and $100,000, and over $100,000.

Survey data was analyzed by two software programs, SPSS and MS-Access. Based on the raw data, SPSS determined other categories in which to group by. For example, core product replication fell into four categories, 0-24%, 25-49%, 50-74% and 75-100%. Because of the high response in the lower (0-24%) and upper (75-100%) categories, these were further analyzed for banner standardization, banner pricing, promotional budgets, staffing, and spending. Other categorizing throughout the survey was done as a result of the input from the ten representatives mentioned above, or from the recommendation of the SPSS program.

The following results are industry segmented (Newspaper, Magazine, Radio, Television) responses. When appropriate, further interpretation has been introduced to explain the disparities. Particular attention has been given to cost cutting activities, investment, measurement, content and their relative impacts on revenue. Figure 10 graphically shows the number of respondents by media category. Of the 243 newspapers that responded, 171 were daily newspapers while the remaining 72 were weekly newspapers. Of the 171 daily newspapers, 84 (49.1%) have circulations under 50,000, 30 (17.5%)

  

Figure 10. Number of respondents by media category.

 

have circulation greater than 50,000 and less than 100,000, 47 (27.5%) have circulation between 100,000 and 500,000, while 10 (5.9%) have circulations of 500,000 or more (Figure 11).

Figure 11. Respondents by newspaper circulation.

 

Figure 12 shows the median length of time, in months, by media group, that their Web site has been on-line. Although magazine companies have had their web sites on-line longer, the variance between the longest and shortest duration is only five months.

Figure 12. Median of months that Web site has been in existence.

The average time on-line across all media groups is 19 months. A review of all media groups, reveals that 18.5% of the respondents have been on-line less than one year (Figure 13). More than 80% of the respondents having been on-line at least one year, matching the prediction established by this researcher, as well as researchers at Editor & Publisher magazine.

Figure 13. Percent of respondents - time on-line.

 

With the exception of magazines, it can be seen in Figure 14 that for their core product, media companies rely heavily on the local market, with national business making up roughly 18% of their overall business (Note: numbers do not always total to 100% due to rounding, as well as some invalid responses).

Figure 14. Percent distribution of core business.

These numbers change slightly for the Web product, with less emphasis on the local customer base, and more on national and international customer base (Figure 15).

Figure 15. Percent distribution of Web business.

As seen in Figures 14 and 15, the composition of newspapers, television, and radio customers is similar in both the core and on-line products. Magazine companies show an inverse relationship with local and national customers. Magazines indicated that both their core and on-line customers are much more national than local. The newspaper industry appears to know who their customers are, while magazine (17%), television (22%) and radio (43%) indicated they don’t know.

Technology

The use of various Web technologies among the media groups is wide spread. Figure 16 clearly shows the mix, with Frontpage-97 being used most frequently, especially by the television industry.

Figure 16. Percent usage of various Web technologies.

Not surprisingly, television leads the way in the use of video, audio, and animations on their Web sites. While all media groups make use of animations, newspapers, magazines and radio are far behind in the use of video (Figure 17).  Elderkin (1996) pointed out that newspapers would have to restrict themselves on how

 

Figure 17. Percent usage of video, audio, and animations.

far they carry multimedia, because of multimedia's complexity and the expense to produce it. Elderkin felt that newspapers would have to find the point where they can use multimedia to maximize their audience while maximizing their advertising potential.

With newspapers, magazines, and radio falling relative to one another (Figure 17), it may be that these industries recognize Elderkin’s findings, and, realize that it will be difficult to compete with television in this area.About one-half of the Web sites surveyed archive their own HTML page structures for each edition, and roughly 70% have established HTML standards or other structural guidelines (Figures 18 and 19). These two questions were posed in the survey for informative purposes and are merely presented here for that purpose.

 

Figure 18. Percent of media companies who archive HTML page structures.

Figure 19. Percent of media companies that have established HTML standards or other structural guidelines.

As discussed previously in the research section of this dissertation, "push" technology is an excellent way for companies to get their product to the end consumer. In fact, pundits recognize this technology as being the future for all media, although presently, few media companies are employing this technology. Figure 20 depicts the lack of use of push technology across the four media groups.

Figure 20. Percent of companies using push technology to deliver Web information.

Of those that do use the technology, 4.9% indicated they use PointCast as a delivery medium, while 9.8% indicated they use some other form of push technology.

In general, media companies rely on outside sources to operate their Web site (Figure 21). Although newspapers have the highest incidence at 38.5%, the media industry in general has not invested in their own software, hardware, or staff to maintain their site. As a result, it can be assumed that to date, media companies are not sure where this new medium (the Web) is going, and they are keeping investments at a minimum.

Figure 21. Percent of media companies that operate their own server.

Of those Media Companies that do not operate their own server, Figure 22 depicts where their sites operate. As shown, more than 50% use sources outside of the media industry.

Figure 22. Percent of media companies that operate their Web sites with other non-media companies.

Although media companies generally do not operate their own server, many provide their customer’s Internet services such as Web site design, construction and consulting to their customers. Figure 22 indicates that newspapers lead the way in providing Internet services to other companies. In fact, 51.6% indicate that they engage in this cost defraying practice to some degree. The differences between those newspapers that offer Internet services to other companies is quite simply a function of the "richness" of their own site. These companies have a greater amount of authoring tools, animation, audio, video and content mix.

Figure 23. Percent of media companies that provide Internet services for other companies.

Newspapers in aggregate show a mix in the various technologies they use to produce their web sites. For example, the percentage of newspapers with these particular technologies is shown in Figure 24 (respondents were allowed to choose more than one).

 

Figure 24. Percent of various technologies used by newspaper Web sites.

As shown in Figure 24, there is no monopoly held by any particular software package for newspapers. The most common response to the type of software used was in the "other’ category (not shown). Looking closely at the "other" category, it can be seen that there are no consistent similar responses. In short, newspapers use a wide variety of Internet software to product their site.

The disparity between those newspapers that offer Internet services and "all other newspapers" lies in the "higher end" technological capabilities. Specifically, video, audio, and animations are more often found within newspapers that offer Internet services (Figure 25).

Figure 25. Internet service providers offer more technological capabilities (percent usage).

Beyond the technological differences, there exists a greater incidence of content mix. For instance, newspapers that offer Internet services are more likely to have community information, entertainment, non-news, real estate, special events, and tourism. Newspapers in the aggregate have content as indicated in Figure 26.

Figure 26. Percent of newspapers with various content.

Newspapers that provide Internet services are more likely to have a higher frequency of multiple content categories than not. The percentage of newspapers that provide Internet services, and whether they have a specific content are shown in Figure 27.

Figure 27. Percent of specific content of newspapers that provide Internet services.

Results of the survey (not graphed) found that in 1997, newspapers in an aggregate with zero banner revenue were 23.2% of the total. Fifty Six percent of this 23.2% belong to newspapers that provide Internet services. This is a higher proportion than the overall 51% that provide Internet services, but not by much. Looking at the banner advertising revenue in the "less than $50K" range, it can be seen that 57.1% of all newspapers experienced this level. Newspapers that provide Internet services were more likely to fall in the "less than $50K" range than those which did not. The next sales increment ($50-$100K) which is 7% of the total, shows a true departure between these two groups of newspapers. In fact, newspapers that are providers of Internet service to other companies are four times more likely to be in the $50-$100K group than those who are not.

Examining the investment patterns of newspapers, 58.7% of all newspapers spend less than $50K yearly on equipment and other investments. The composition of that 58.7% also reflects that proportion of newspapers that are "providers." To be exact, 46% of the newspapers that invest less than $50K, are Internet providers. Taking a further look into the next investment increment (between $50K to $100K) there is a departure in spending patterns between the two newspaper groups. The total percentage of newspapers that spend between $50-$100K is 13%. Within that percentage, 61% of it belongs to newspapers that provide Internet services.

Outside Internet services conclusions

Newspapers that provide Internet services to other companies appear to be a part of sites that have a greater depth within their own site, both technologically and content. Whatever revenues are generated by these ancillary services cannot be determined at this time. It can be said that although newspapers that provide these services do have a higher incidence of greater revenue (as defined by banner advertising), they account for only a fraction of the total newspapers. Whatever the incremental increases in revenue, be it in banner advertising, transactions, or classified, the investment that has been made by those participating newspapers is nominal. No marked increase in investment spending can be linked to the activity of providing Internet services.

Editorial/Content

Many companies in the four media groups reference their Web site within news features of their core product. The results of this are shown in Figure 28.

Figure 28. News features in/on the core product that reference the Web site (percent of respondents).

The frequency in which these references are made is shown in Figure 29. Television and radio make use of their core product, regularly, to reference their Web site. For newspapers, the daily percentage is somewhat skewed by the 72 weekly newspapers who participated in the survey. Removing these 72 newspapers from the aggregate, the daily percentage for newspapers rises to 56.8%, still clearly behind television. Magazines show a heavy (63.9%) use of monthly references, however it should be noted that most magazines are produced monthly. As a result, Figure 29 is presented for informational purposes and does not correlate to any other variables. The point to be made in Figure 28 and Figure 29, is that while newspapers reference their Web sites (78.7%), they lag behind magazines and television.

Figure 29. Frequency that the news features in the core product reference the Web site (percent).

The majority of media companies do not have a procedure in which corrections are edited in their archive. Once a Web article goes to archive, needed corrections are not done. For the end-user, this represents a flaw that potentially makes these Web sites unreliable (Figure 30).

Figure 30. Percent of media that correct their archived text.

Media companies now use the Internet quite often to gather information. In fact, an average, 93.3% of the media groups use the Internet in some fashion (Figure 31).

Figure 31. Editorial staff use the Internet to gather information (percent).

The Internet is frequently used to research stories and obtain feedback from Web users. A seen in Figure 32, less than half of these media groups use the Internet to track their competitors. For example, 94% of newspapers use the Internet to research stories, 72% use the Internet to obtain feedback from Web users, and 43% use the Internet to track competitors.

Figure 32. How the Editorial staffs indicate they use the Internet (percent).

Generally the Web newsroom staff is integrated with the newsroom of the core product. Those that function as a separate entity account for only 24%, with the exception of radio which claims better than 42% separation. As seen in Figure 33, these media groups do not appear willing to invest heavily in separate editorial staff and tend to use their existing staff to produce their Web product.

 

Figure 33. Integration of Newsroom Staff (percent of responses).

Most media have developed editorial content for their Web sites, however newspapers trail all other media (Figure 34). It is important to note this because content is often why customers come to a particular site.

Figure 34. Percent that have developed editorial content for Web site.

Figure 35 shows the areas in which this editorial content is generally developed. Community information, special events, and entertainment rank among the highest in all media categories.

Figure 35. Types of editorial content developed for Web site (percent).

The content mix between the four industries reflects that newspapers offer more tourism information than the others (Figure 35). A higher percentage of radio and television offer community information, special events, and entertainment. Newspapers only surpass magazines in these two areas. Radio and television indicated they have a greater incidence of entertainment and non-news events than newspapers.

Scooping the core product means putting the information on the Web site before it is available in the core product. Cochran (1995) believed that newspapers need to "scoop" themselves regularly, and until they do, there will not be enough of a reason for readers to turn to the on-line product. Newspapers and magazines "scoop" themselves about 50% of the time, while television and radio lag far behind (Figure 36). This lag may be due to the nature of the medium, because television and radio have the capability to deliver their information almost instantaneously with their core product.

Figure 36. Percent who "scoop" the core product.

Few sites originate content specifically for the Web. Figure 37 shows a 10% median across all four media groups combined, with a low of 5% for newspapers and a high of 20% for television. These low numbers indicate that media companies are not willing to invest time and effort into the production of Web information.

 

Figure 37. Percent of content originated for Web.

Respondents across all four media groups indicated that some portion of their core product is replicated for their Web product (Figure 38). In fact, newspapers and magazines replicate 50% of their core product to the Web. Although Kline (1996) felt that hundreds of newspapers were making a mistake by focusing on putting their content on-line (shovelware), many are doing just that.

Figure 38. Percent of core product replication.

Focusing on newspapers only, and to obtain a more precise distribution for core product replication, responses were broken down into four even categories of 25% each. These categories are shown in Figure 39. Survey results found 28.4% of the respondents indicated that 0-24% of their core product is replicated on-line, while 15.2% indicated that 25-49% is replicated on-line, 16.9% indicated that 50-74% is of their core product is replicated, and 39.5% said that 75-100% of their core product is replicated on-line (Figure 39).

Figure 39. Percent of newspapers core product that is replicated on-line.

Although newspapers indicated that 50% (mean) of the core product is replicated on-line, the most common increment is between 75-100%, with 39.5% (Figure 39). Looking further into the two increments of 0-24% and 75-100%, the following can be seen:

1) The lower increment (0-24%) and higher increment (75-100%) are similar to each other in regards to circulation size, core customer distribution, and web user distribution. It does not appear, however, that these three characteristics play any part in whether a newspaper replicated their core product on-line.

2) Those newspapers that also have content from outside sources did not indicate any strong correlation. Both upper and lower increments reflect a percentage of outside sources to all other newspapers.

3) Those newspapers that indicated they have a standardized banner size also indicated that they replicate more of their core product content on-line. (Figure 40).

Figure 40. Content replicators have more standardization of banners (percent).

4) Consistent with standardized banner sizes is the charge for a month of advertising. Figure 40 indicates that those newspapers who replicate a higher percentage of their core product on-line also charge a higher percentage for their banner ads than others. (lower $463, all $1,332, and upper $2,343).

Figure 41. Banner pricing for the lower and upper groups (in dollars).

5) The number of page views per week seem to increase as the percentage of replication increases. It can be seen that newspapers with lower content replication generally receive 5,000 page views weekly, all newspapers 10,000 weekly and those with greater than 75% replication receive 23,500 page views weekly (Figure 42).

Figure 42. The number of page views per week (for content replicators).

6) Those that have a higher replication are more likely to provide advertisers with some measurement data. The lower content replication has 36% providing measurement, all other 51%, and the upper content replication show 61% providing data (Figure 43).

Figure 43. Content replicators provide more advertising measurement (percent of providers).

7) Looking closely at the promotion budgets, there are higher amounts for those newspapers with high content replication, however the difference between the lower and upper range is minimal and show similar patterns, as can be seen in Figure 44 with the 4 major promotion budget increments:

Figure 44. Promotion budget of content replicators (percent of those who replicate).

8) Although not substantially different than all newspapers, it can be shown that staffing for the higher replication incidence is higher than the low replication. When there is mid to high replication, there will be more staffing, but the difference is seen only in the editorial department (Figure 45).

Figure 45. Staffing levels of content replicators (number of people).

9) Spending for equipment and services (Figure 46) is also relatively higher for those who replicate content, however there is not a big difference between the three categories.

Figure 46. Investment spending of content replicators (percent).

From the nine points above, it is notable that the move to replicate the traditional product to the on-line product generally creates a correlation in which the Web site receives twice as many page views.

Circulation sizes indicate that content replication is not limited to large, or small sized newspapers. Revenue streams do not show any marked improvement from those newspapers with little to no on-line replication. Tom Dahlin, Director of Forecasting for Editor & Publisher stated, "Like other newspaper web sites with a high percentage of outside sources of content, the users recognize the overall attractiveness of the web site, but quality has not yet translated into advertising dollars" (Dahlin, 1998).

Few replicate the Web product to the core product. As shown in Figure 44, radio replicates the most with 12.5%. This seems to indicate that content is first developed for the core product and subsequently used by the Web product. This could also be an indication that content produced by the Web product is not taken seriously enough by the core product editors.

Figure 47. Percent of content from Web product used in core product.

To some degree, most newspapers use outside sources to produce content for their Web site. In fact, the median percentage of outside content is 5%. Although, this percentage of content is low, 12% of all newspapers indicated that at least 30% or more of their content did come from outside sources. The newspapers that have this "30% or more content" were specifically looked at to see if they behave differently, and what lessons might be learned (Figure 48).

Figure 48. Percent that use outside content.

The areas that were examined for the "30% content" are measurement, promotion, page views, staffing, investment spending, and sales revenue.

1) Measurement - newspapers that provide measurement data to advertisers is 51.8% (Figure 62). Newspapers that have "30% content" and provide data to advertisers is 70.8%

2) Weekly page views – on average newspapers have 10,000 per week (median) as shown in Figure 49.

Figure 49. Page views per week.

Interestingly, newspapers with "30% content" have more than double the weekly page views at 22,500 median.

3) Promotion dollars spent by the "30% content" group parallels the incidence of outside content behavior as shown in Figure 50.

Figure 50. Newspapers with high percent of outside content spend more on promotion.

4) Staffing - newspapers with a high percent of outside content have more full-time staffing (Figure 51). In fact, newspapers overall, employ one advertising, two editorial, and one technical person. The "30% content" content group staffing numbers are much higher, employing on average, two advertising, three and one-half editorial, and 3 technical personnel (all median).

Figure 51. Newspaper staff sizes.

5) Investment spending - newspapers that have a higher outside content also spend more money on investments for equipment and software (Figure 52).

Figure 52. Newspapers with high percent of outside content spend more on investment.

The number of newspapers with "30% content" expect that same increment to rise from 25% to 33.3%. In essence, those newspapers that are more involved with outside sources for content are planning to spend more than all other newspapers.

6) Sales revenue - banner advertising between all newspapers and those newspapers that fall into the "30% content" group, show differences as well (Figure 53).

 

Figure 53. Newspapers with high outside content experience higher banner revenue.

With respect to classified advertising, however, those newspapers that fall into the "30% content" group show no marked increase of revenue over all other newspapers.

It can be concluded that those newspapers indicating they have at least a 30% level of their content from outside sources have more staffing, more investment in equipment and software, more promotion dollars, more page views per week, more measurement, and they experience more banner revenue than their counterparts. Although their pricing of banners is twice as high as the typical newspaper, their page views warrant higher pricing.

An analysis of the survey data found that newspapers with at least 30% content from outside sources claim to have more tourism, community info, special events, entertainment, non-news, and real-estate than other newspapers. However, none exceed more than 10% of the norm. There is also a higher incidence of original editorial content with these newspapers (80.6% versus 68.4%).

Although this group (30% content) of newspapers is in-line with banner pricing and site development, the revenues have not followed. It may be that the approach to web site development and investment must be consistent with high revenue generators. In this particular group the one item that deviated considerably, not only with other newspapers but especially with the high revenue generators, is site measurement (Web sites that provide advertising measurement to their advertisers). Although this group experiences a greater level of measurement to advertisers than the norm (70.8% to 51%), it has considerably less for "outside" measurement (9% to the industry’s 16%). This is exceptionally low if compared to the high revenue generating newspapers which provide 46.7% of their measurement from outside sources.

Turning to archival information, all four media groups provide archives to some degree, with the magazine industry on top at 72.5% (Figure 54).

Figure 54. Providing archival information (percent).

The amount of time articles stay in the Web edition before going to archival or deletion is shown in Figure 55. The only significance here is that some sites charge for articles once they are stored as archival information, so the sooner an article is moved to archive, the sooner they can start charging for it.

Figure 55. Time articles stay in Web edition before going to archival or deletion (percent of respondents).

Along with cross promotion, alliances with other media companies are another avenue to reduce promotion spending. Most top revenue generating newspapers are in fact in alliances. However, the percentage of newspapers (in the aggregate) that have alliances has actually decreased from 1996 to 1997 (source Editor & Publisher, 1997). Within the radio industry, which has a tradition of "trade" practices, commensurate alliance levels as newspapers is seen (Figure 56).

Figure 56. In alliances with other communications companies (percent).

One of the underlying themes in numerous articles and at industry conferences during 1997 has been that newspapers need to partner with other industries in order to build a strong alliance in the community, and ultimately make a profit. Figure 56 indicates that media companies are not partnering with other communication companies.

Advertising

With respect to advertising, about 75% of the newspapers and magazines reported that they have paid advertising on their site. On the other hand, television and radio were around the 50% range as shown in Figure 57.

Figure 57. Web sites with paid advertising (percent).

A number of sites also have sponsors as depicted in Figure 58. Although there is a larger percentage of Web sites with no sponsors, a surprise finding was in the high percentage of sponsors, which averaged 42.8% overall.

 

Figure 58. Web sites with sponsors (percent).

Banner standardization has been an approach to homogenize a product throughout the industry. Although there are still many different sizes that newspapers use, there is a conscientious move to become more uniform. In 1996, 44% of all newspapers indicated that they had a standard banner size, according to Editor & Publisher 1996 statistics. It was found in this study that the percentage jumped to 63.1% (Figure 59).

Figure 59. Percent that indicated they have a standardized banner size.

A move to define a standard unit of measurement (in this case a standard banner size) is a move to offer advertisers a more price competitive product. This offer enables advertisers to shop for the best banner buy, while at the same time enables newspapers an opportunity to cut costs through efficiency. The key issue is to convince advertisers on the performance of the web site and the profile of the typical user.

Even though respondents say they have standard banner sizes, they do not. Although there was a high indication of standardized banner sizes by the survey respondents (as shown in Figure 59), the most common size was 468x60 pixels, which is used by only 15.5% of the respondents. Other common sizes were not easily distinguishable. This is an important area for the industry to improve on, and one that should not be overlooked in future surveys.

Newspapers were asked what they charged for their typical banners, the results reflected the greatest difference of any variable in the entire study. This is largely due to the underlying factors that impact banner rates. For example, the responses ranged from $5 to $30,000 monthly. The median price was $125 and the mean was $1,563.

Looking closer at this pricing, the range of banner prices was delineated into four ranges. They are each evaluated for correlations with each other and the causation which drives the banner revenue. Consider the following four ranges:

$0-$48 - the bottom 24% price range (Lower)

                $49-$120 - the bottom 25%-49% price range (Mid-Lower)

                $121-$440 - the middle 50% - 74% price range (Mid-Upper)

                $441 and above - the upper 75%-100% price range (Upper)

Considering the price of banners to be a function of page views per week, a positive correlation should be seen between these two variables. In other words, as page views per week increase, the price of banners should increase as well.

The page views across all newspapers has a median value of 10,000 weekly; charging $125 (median value). This represents a price of approximately $0.01 per page view. The page views across the Lower range has a median value of 4,800 page views per week. Those charging in the Lower range (0-$48) are receiving $0.00-$0.01 per page view. The page views across the Mid-Lower range has a median value of 5,730 page views per week. Those charging in the Mid-Lower range ($49-$120) are receiving $0.008-$0.02 per page view. The page views across the Mid-Upper range has a median value of 24,500 page views per week. Those charging in the Mid-Upper range ($121-$440) are receiving $0.0049-$0.017 per page view. The page views across the Upper range has a median value of 127,500 page views per week. Those charging in the Upper range ($441-up) are receiving $0.003 and up per page view (Figure 60).

Figure 60. Page views of the four banner ad pricing ranges.

From these results it can be seen that as page views increase, the overall price for banners increases as well. However, the marginal increase in page views does not proportionately increase with unit prices. Instead, there exists a decreasing marginal price system. Newspapers that have a higher number of page views also have a lower value for the advertising banners. For example, in the $49-$120 category, the average price per page view is 0.01 cents compared to 0.003 cents per page view in the >$440 category.

This behavior can be explained by two factors. First, only 16% of all newspapers provide advertisers with outside measurement data. This data enables advertisers the freedom to examine the best alternatives among newspaper web sites. In the absence of this measurement data, advertisers must rely on the newspaper’s own internal measurement data, which presents a conflict of interest, or advertisers must make some non-quantifiable determination with some web site’s presence. Another explanation for decreasing unit prices is the cost structure in which web sites operate. Unlike most operations, the cost of additional cyberspace output is nebulous at best. Once a site is put on-line, there are virtually no variable costs, so there are no additional costs if the site receives 5,000 page views or 50,000 page views.

Looking at the total revenues from each of the four pricing categories, Figure 61 indicates that revenue is skewed more towards the upper pricing range. As can be seen,

Figure 61. Pricing and revenue streams (percent).

of the newspapers in the Upper range, 15% indicated revenues in the $50-$100K range, 23.4% indicated revenues in the $100-$249K range, and 0.5 indicated revenues in the $250-$500K range. Combining these three ranges ($50-$499K), it can be seen that 37% of the respondents indicated revenues above $50,000, far above the other three pricing ranges.

By aggregating the total revenues for each category, it can be seen more clearly that there is a polar distribution of the total banner revenue. Both the lower range and upper range of prices reap 75% of the total revenue, as shown in Figure 62. Clearly, the majority of revenue is captured by the upper end of the pricing category with 54%.

Figure 62. Pricing levels and percent of total revenue.

Even with the lowest price per unit (.003 cents per page view), the upper range of pricing still receives the majority of revenue. It is important to note that revenue in the upper category is represented by 20% of the total number of newspaper web sites, while the lower category (price per unit of 0.01 cents per page view) is represented by only 16% of the industry. The remaining 64% of other newspapers charge between .0049 to .02 cents (per page view), crossing between the lower levels and higher levels. When the pricing is double that of the total industry (.02 cents compared to .01 cent), the mid pricing is partially preventing them from capturing their share due to pricing which is too high.

Another reason why this mid range is not capturing a share, lies in the issue of measurement. Of these respective price categories, measurement for advertisers plays differing roles. As seen in Figure 63, the lower price category has a lower incidence of measurement compared to the upper levels. This lends itself to conclude that at lower price levels, measurement is less critical in determining overall banner revenues. However at higher price levels, measurement practices are more than twice as common (than the lower price level) and should be deemed an important factor in banner revenues. This measurement would seem to justify their higher banner rates that are proportionately higher with respect to the users. (Twice as many users equal twice the banner rates).

Figure 63. Percent of newspapers that use measurement by price category.

Looking strictly at the price of a banner ad per month, by industry, Figure 64 shows that the median for all media, newspapers and television is $100, while magazines average five times that amount at $500 per month. Note that these rates were for their particular "standardized" banner, which varied greatly.

Figure 64. Price of banner ad running 1 month.

Overall, approximately 50% of the media indicated that they have standardized ad sizes. While they were not asked to report these sizes, it may well be that they vary greatly, much like the findings with banner ads. Figure 65 shows the responses by media group. Note that newspapers and magazines are comparable, while television and radio are far behind. The significance is in the increase of standardization from 1996 (15%) to 1997 (54%) for newspapers, based on 1996 Editor & Publisher statistics.

Figure 65. Percent that indicated they have standardized ad sizes.

Few media companies are participating in network advertising. Newspapers lead in participation (Figure 66), however there is little difference between the four media groups. Seybold (1995) pointed out that the potential of this kind of participation is in sharing resources and ultimately reducing costs. It appears that, in general, media companies are unwilling to participate, or, they do not have the technical savvy and/or resources to do so.

Figure 66. Percent participation in network advertising.

One of the areas of concern that was discussed earlier, is in providing advertising measurement data to advertisers. While more than 50% of all media (Figure 67) indicated they provide advertising measurement data, much of this comes from their own internal sources (Figure 68).

Figure 67. Percent that provide advertising measurement data.

Those that provide outside measurement average only about 15% of the aggregate (Figure 68). This is an important fact to note as those media that provide outside measurement generally attract more viewers and higher banner rates.

Figure 68. Percent that provide outside advertising measurment.

Television and radio offer little in the way of Web classifieds. Figure 69 indicates that a large percentage (71.3%) of newspapers offer Web classifieds, with an impressive 41.2% (of the 71.3%) being keyword searchable.

Figure 69. Percent that offer Web classifieds.

Focusing on newspapers, 55% indicated that their Web classified advertising is sold in conjunction with their core product. Only 14% indicated that their classified ads are sold as a separate entity from the core product. Of the 55% above, 75% of their classified advertising on the Web site is also in their core product.

Web classified ads seem to be an important function of many sites. Figure 70 shows the average page views generated for those who have classifieds on-line, and those who do not.

Figure 70. Page views of Web sites offering classified ads.

Dynamic advertising that is query specific gives the user the ability to search classified ads for very specific things. For example, cars may be searched for a particular price range, make model, miles, color etc. The flexibility is endless and tends to attract more users. Figure 71 indicates that roughly 20% of the media companies surveyed provide this service. This may be an area in which expansion should be considered.

Figure 71. Percent that provide dynamic query advertising.

Figure 72 shows that sites providing dynamic query experience much greater page view counts than those who do not.

Figure 72. Page views of those companies who do and do not provide dynamic query advertising.

 

Coupons are found in newspapers daily, and many newspapers are purchased by consumers solely for the coupons. Web coupons have been a topic of discussion within the industry for some time. Figure 73 shows that the industry, as a whole, has not taken advantage of this technology. This could be due to a variety of reasons including lack of standards, non interest, development costs, or quite possibly, sites do not want to offer coupons because the results (low consumer response) to advertisers may be low, therefore creating a negative effect.

Figure 73. Percent that offer Web coupons.

 

Promotion

Media companies use a variety of methods to promote their Web sites as shown in Figure 74. It can be seen that promotion tends to stay within the industry. That is, newspapers promote in other newspapers, magazines in other magazines, television in other television, and radio in other radio stations. Direct mail has a strong use as well, with a 25% across the board usage.

 

Figure 74. Promotion vehicles used by Web sites.

Figure 74 also reflects the lack of partnering within the four media groups. As discussed previously, few Web sites are in alliance with other media companies. This can present a number of problems, the most obvious being the higher cost to promote a Web site. By taking advantage of partnering, cross promotion can take place with little or no cost to the partners. On the other hand, media companies are cross promoting with other Web sites. The

Figure 75. Cross promotion with other Web sites (percent).

cross promotion of products with other web sites is an opportunity to defray the costs of promotion. Newspapers have a lower incidence rate than magazines, television and radio (Figure 75). With newspaper promotion budgets decreasing, the lack of promotion and alliances seems counterintuitive, or the possibilities for this activity are minimal.

Promotion budgets increase the ability of a company to generate greater interest in their Web site. The trend however, has been a reduction in the overall on-line budget. In 1996, Editor & Publisher found that newspapers were budgeting an average (mean score) of $44,000 per year for Web site promotion. This survey asked the same question as the 1996 Editor & Publisher survey, and found 1997 promotion budgets drastically reduced to an average of $19,950. Considering most Web site promotion budgets are under $10,000 (Figure 76), media companies need to work together, creating alliances and promoting their efforts.

Figure 76. Size of promotion budget (percent of respondents).

Internet marketing tactics used to bring traffic to the Web site are shown in Figure 77. Multiple responses were offered among the media groups, with listing URL on search engines, updating content, and building community interest ranking among the highest. Coming in last was push technology.

 

Figure 77. Internet marketing tactics used.

Staffing

As discussed previously, the editorial Web staff generally functions as a part of the core product editorial staff. Only 23% indicated their editorial Web staff functions as a separate entity. Looking now at the other Web staff (advertising and technology), Figure 78 indicates that about 41.7% of the media Web sites have separate advertising and technology staffs.

Figure 78. Advertising and technology Web staff as a separate entity from the core product (percent).

Overall, the number of employees working on the typical media Web site are shown in Figure 79 (median). This table represents the overall aggregate of the industry. As a result it does not show specific areas of interest, such as staffing by the amount of page views a site receives each week. Figure 80 shows these figures in four different page view increments, less than 50,000 page views, 50-250,000 page views, 250-1 million page views, and greater than 1 million page views. Note that part time employees were counted as working four hours per day, or .5 persons per week.

Figure 79. Staffing of advertising, editorial, and technology.

Figure 80. Staff sizes by amount of page views per week.

It can also be seen (Figure 81), that for those who provide Internet services there is a higher incidence in staffing.

 

Figure 81. Average staffing size of those who provide and do not provide Internet services.

For those weekly newspapers interested specifically in the staffing of weekly newspapers, Figure 82 shows the breakdown of the six major categories. Overall, weekly newspapers average 1.9 full time employees and 1.9 part time working on their Web site.

Figure 82. Average staff size of weekly newspapers.

One of the big questions for 1998 is how will staffing change. In essence, do Web publishers expect an increase, decrease, or do they think that staffing sizes will remain the same in 1998. Although 45.9% of the respondents indicated there would be no change in staffing for 1998, 54.1% indicated there would be an increase in staffing. From the 54.1% that indicated an increase, respondents from newspapers and magazines anticipate almost a 50% increase in staff, while television and radio are preparing to greatly expand, somewhere in the range of 80%. Specifically, these changes are shown in Figure 83.

Figure 83. Percent increase in staff size expected in 1998.

In concluding this section on staffing, less than 40% of all media indicated that they have a full time editorial director. Figure 84 shows newspapers with the highest incidence, and radio trailing far behind. The indication here is that Web sites are not willing to invest heavily in staffing. This trend was also seen in Figure 33.

Figure 84. Web sites with full time editorial director (percent).

Subscriptions /Transactions

Few companies charge for access to their Web site. Figure 85 indicates that neither the television, nor the radio industries are currently charging. Those who do charge make up only about 10% of the aggregate. McKinnon (1996) stated that on the Internet, people generally expect most everything to be free. When they cannot find free news at one site, with so many others to choose from, consumers will simply go to another site that does offer the news free.

Figure 85. Percent of media that charge for access to their Web site.

Generally, media companies that are charging subscriptions, are able to do so because they offer a niche product, such as the Wall Street Journal. Of those who do charge for Web site access (Figure 85), Figure 86 shows what these sites charge.

 

Figure 86. Monthly charge for Web site access (percent of those who charge).

Though most newspapers do not currently charge a subscription fee for their on-line product, Peterson (1996) stated that eventually, newspaper-sponsored information on the Internet will be available only to paying subscribers using access passwords. He believed that once there is a sufficient mass of users, and they get in the habit of getting their information on-line, publishers will then be able to charge. Figure 87 shows that newspapers and magazines plan to charge for access to their product in the future. That may prove difficult if television and radio can produce a similar product. For the 40% or so who plan to charge, there are 60% who do not plan to charge. This 60% who do not charge will be competition to the 40% that intend to start charging. This will be hard to overcome as competition not charging for their product will become another obstacle to those who want to charge.

Figure 87. Web sites that intend to charge for their product in the future (percent).

When it comes to archival information, few media companies charge for access to their archive. Of the 224 companies that indicated they offer archival information, only three charge per minute, at an average rate of 93 cents per minute, and 12 sites charge an average of $1.10 per article. One site indicated they charge a $50 per year access fee ($12.50 quarterly). With 93.3% of all media archives available for free, it may be difficult for these media companies to begin charging for archival information in the future.

In giving access to their Web site and archive, media companies should require registration to enable them to learn more about their consumers. Clearly, this is not happening, as shown in Figure 88.

Figure 88. Web sites not requiring registration (percent).

With registration, Web sites would have a better understanding of their customer base, the demographics of these customers, and would be able to give their advertisers better measurement data. Another advantage of requiring registration is in the direct marketing of these consumers. This seems to be an area that media companies need to work on.

Another area of concern is dealing with secure electronic financial transactions. Most sites (Figure 89) do not currently provide this type of security. This limits the ability of the site

Figure 89. Web sites that provide secure electronic transactions (percent).

to market products, charge for subscriptions, or collect for classified advertising. However, media companies recognize this potential problem. Figure 90 shows that more than 60% of the media companies intend to offer secure financial transactions in the future.

Figure 90. Plan to offer secure financial transactions in the future (percent).

 

Sales and profitability

Pogash (1996) stated that millions of dollars have been spent, and, millions more will be spent to start Web site operations and to continue their existence. This survey examined the years 1995, 1996, 1997, and 1998 for spending and potential spending of the four media groups. The following figures show this spending, which is for equipment and services, and does not include salaries.

Figure 91. Newspaper investment (in dollars).

Figure 92. Magazine investment (in dollars).

 

Figure 93. Television investment (in dollars).

Figure 94. Radio investment (in dollars).

As shown in Figure 95, the expenditures for 1997 have very similar patterns between newspaper, magazine, and the television industries. Also, as previously seen, newspapers are decreasing their expected expenditures relative to their historical rate. This may open the

Figure 95. 1997 expenditures for on-line products by industry (percent of respondents).

door for the magazine and television industries to develop a Web site capable of surpassing the newspaper’s Web site, both content wise and technologically.

 

Figure 96. 1998 projected expenditures for on-line products by industry (percent of respondents).

For 1998, media companies plan to decrease spending in the less than $50K range, while increasing spending in the $50K-$99K range. It can be seen (Figure 96) that television and radio continue to outspend newspapers in the less than $99K ranges.

The big question for the last few years, has been whether or not media companies are making money with their on-line ventures. Pundits acknowledge that this is a new industry and losses can be expected. This survey found that to be generally true, and went deeper into the question by looking at profits and losses associated with page views per week and number of employees working on the Web product. The following five figures show these profits and losses for the four media groups for 1997.

Figure 97 shows 14 newspapers (5.7%) with losses greater than 1 million dollars. These newspapers averaged 26 employees and a little over 1 million page views per week. Two magazines (2.6%) also fell into this category, loosing more than 1.5 million dollars. For these magazines, page views per week are considerably lower than newspapers, however the number of employees used to produce the product is almost the same.

Business type Page views Employees Income Losses # in category
Newspaper

1,233,801

26

553,571

-1,385,714

14

Magazine

215,000

22

350,000

-1,575,000

2

Figure 97. 1997 media losses greater than $1,000,000.

Figure 98 shows 20 newspapers (8.2%), 3 magazines (4.2%), and 2 television stations (3.9%) lost between $500,000 and $1,000,000 during 1997. Interesting to note is the number of employees versus page views and losses. In comparing Figure 97 and Figure 98, it can be seen that the number of employees is relative to page views and losses. As employees decrease, so do page views and losses.

Business type

Page views

Employees

Income

Losses

# in category

Newspaper

676,468

14

402,500

-675,000

20

Magazine

323,333

14

350,000

-666,666

3

Television

10,050

11

37,500

-562,500

2

Figure 98. 1997 media losses between $500,000 and $1,000,000.

Most media companies fall into the next category, with losses between zero and $500,000. Page views and losses continue to decrease with the reduction of employees.

Business type

Page views

Employees

Income

Losses

# in category

Newspaper

49,643

3

76,171

-152,148

128

Magazine

52,153

3

73,214

-130,357

28

Radio

56,744

2

30,555

-100,000

9

Television

58,475

3

32,608

-126,086

23

Figure 99. 1997 media losses between $0.00 and $500,000.

Turning to profits, it can be seen that some media companies are in fact generating profits. Figure 100 shows 35 newspapers (14.3%) with profits averaging almost $30,000. These 35 newspapers did this with less than one (0.7) employee producing the on-line edition. Interestingly, page views for these 35 newspapers average 21,545.

Business type

Page views

Employees

Income

Profits

# in category

Newspaper

21,545

0.7

92,142

29,285

35

Magazine

9,810

1

150,000

70,000

5

Radio

1,200

0

25,000

25,000

1

Television

62,583

1.75

212,500

70,833

6

Figure 100. 1997 media profits between $0.00 and $500,000.

Of the 35 newspapers in the $0.00 - $500,000 profit category, 18, or about half are weekly newspapers, 14 are dailies with circulation under 60,000. The remaining three are dailies with circulation levels between 100,000 and 160,000. The interesting point here is that weeklies, and small to medium sized newspapers are the ones making the money (Figure 101).

Figure 101. Number and type of newspapers that indicated a 1997 profit.

Finally, seven companies were found to be making more than $500,000 in profits during 1997. A reverse trend can now be seen (Figure 102), as employees, page views and profits are increasing. Of the six newspapers shown, two indicated circulation sizes of over 500,000, two were in the 100-150,000 circulation range. The other two did not indicate a circulation size.

Business type

Page views

Employees

Income

Profits

# in category

Newspaper

459,580

5

1,412,500

895,833

6

Magazine

300,000

9

1,125,000

500,000

1

Figure 102. 1997 media profits greater than $500,000.

Looking closer at income, four major categories were explored as part of this survey. They consisted of banner advertising, Classified advertising, Subscriptions, and Transactions. Figure 103 shows percentages of 1997 classified advertising revenue by industry.

Figure 103. 1997 Classified advertising revenue by industry (percent of respondents).

Newspapers capture proportionately more classified advertising than the other three industries. Newspapers have a lower percentage of zero revenue earners and higher percentage of <$50K earners. The three other industries (magazine, television, and radio) show the majority receiving no revenue.

Although newspapers are experiencing impressive growth rates in banner advertising, magazines and television are moving upwards in the same range. Radio is decidedly lagging behind with lower revenue as shown in Figure 104.

Figure 104. 1997 banner advertising by industry (percent of respondents).

Having determined the net result on overall banner revenue, this analysis now looks at the levels of banner revenue to determine if there is any differences or correlations between their behaviors and directions. It has been determined that newspapers charging in the upper range ($440 per month and up) of banner pricing capture 51% of the total revenue. The next determination is to see if the highest revenue earning newspapers consistently operate in accordance to the conjectures previously outlined. Specifically, if their actions towards measurement, alliances, promotion, and content, respond to the demands of a competitive product.

The data collected for banner advertising was reported in seven increments (None, <$50K, $50K-$99K, $100-$249K, $250-$499K, $500-$999K, and >$1Million). For graphing purposes, the seven increments were collpsed into the following four increments:

1) None 2) Less than $50,000 3) $50,000-$250,000 4) Over $250,000

Figure 105 shows that for those newspapers receiving higher levels of revenue, it is more likely that they provide measurement for advertisers. There is a strong positive correlation between revenue levels and incidence of measurement. In fact, 93.3% of the newspapers that are making $250,000+ provide measurement. This is a significant increase from the industry wide practice of only 51.2%.

Figure 105. Revenue levels and percent that provide measurement.

The percentage of newspapers that have alliances with outside media companies is only 21.3%. Based on Editor & Publisher statistics from 1996, this is actually down from 29% that participated in alliances just last year. Using the same category delineation, Figure 106 shows that the high revenue generating newspapers are more often in alliances. In fact, 53.3% of them are in some type of alliances, whether it be in editorial, technological, marketing, advertising or other (Figure 106).

Figure 106. Revenue generators and percentage of those with alliances.

It was also discovered that as revenue streams increase, the number of weekly page views increases as well (Figure 107). It cannot be concluded that there is a causal relationship, however, a strong positive relationship exists between the two variables.

Figure 107. Page views and revenue streams.

It can also be concluded that high revenue generating newspaper sites are more substantive in editorial content. Newspapers in aggregate indicated that 68.4% of them have original editorial content, compared to 100% of the high revenue generators (over $250K annually). Further, it is important to point out the greater emphasis in the content mix, as seen in Figure 108.

Figure 108. All newspapers versus. high revenue generators and the content they provide (percent).

High revenue generators consistently have a higher incidence of the mentioned original content categories. Newspapers in the aggregate indicate that 5% of their content is from outside sources, compared to 10% of the high revenue generators. This is especially interesting because the high revenue generators have much more staffing. In fact, they have on average ten full-time staff members for editorial, relative to the aggregate newspaper’s two staffers. This heavy investment of personnel is also extended into other areas such as promotion and measurement.

As discussed earlier (Figure 67), 51.2% of all newspapers provide advertisers with measurement data. Of this amount, only 16% went to outside independent sources for this data (Figure 68). However, among the high revenue generators, 46.7% went to outside sources for measurement. This is particularly important due to the obvious conflict of interest involved and a price level which is 10 times as high as the lower levels.

With average promotion budgets around $10K-$12K, the high revenue generators are budgeting more often in the $50K-$100K annually. They also cross promote with other companies more often than the newspapers in aggregate (60% to 54%), and their emphasis in promotion, using other promotion vehicles (besides their core product), are more common. Figure 109 indicates that high revenue generators use promotion vehicles more so than the aggregate.

 

Figure 109. All newspapers versus. high revenue generators promotion vehicles (percent).

There are also marked differences in the marketing tactics used to bring traffic to the Web site. As shown in Figure 110, URL searching, updating and free content are the only areas in which these groups have similar patterns.

 

Figure 110. All newspapers versus. high revenue generators marketing tactics (percent).

The Web has had little effect on the core product of the four media groups. When asked what increases or decreases in advertising dollars have been experienced in the core product, as a result of the Web, 93% indicated they saw no change. Only 1% saw a decrease in the core product, but a number of increases were seen by the remaining 6%, as shown in Figure 111.

Figure 111. Percent increase in advertising in core product as a result of the Web.

When it comes to competition, media companies generally indicated that they do not have a lot of competition. Figure 112 shows the ranking of various competition, by industry, on a scale from 1-10, where 1 is no competition and 10 the highest competition.

Figure 112. Ranking of media competition (1 lowest, 10 highest competition).

Showing this graphically, it can be seen that newspapers may be underestimating their competition. As shown in Figure 113, newspapers view other local newspapers as their biggest competitor in the on-line business. Magazine, television, and radio were all perceived as very little threat as indicated.

Figure 113. Competition ranking, 1 = least competition, 10 = most competition.

Cross industry conclusions

The newspaper, magazine, television, and radio industries are all in the on-line business to some degree, and the levels of success throughout the industries varies. It is tantamount to anyone’s success to understand who the competition is and why. As revealed, newspapers seem to think that the other three industries are not much of a threat to their business. This may be a very big mistake as other industries consider newspapers to be more competition than newspapers themselves. If fact, both radio and television rate newspapers twice as competitive than the newspapers rate them.

Newspapers may be underestimating their competition. Television and radio are already providing more community information and original content, and, they have similar numbers of users visiting weekly. Measurement by magazine and television are also at similar levels.

As a measure of success, the behavior of the highest revenue generating newspapers was studied. It was determined that proper content mix, measurement, alliances, cross promotion, standardization, and increasing on-line budgets were all elements these newspapers had in common. If this is an accurate measurement which can be applied in other industries pursuing banner and classified advertising, all four industries should take note.

Other

Negroponte (1995) believed that "The Daily Me" or personalized edition will be the future of news distribution. Of all media respondents, only 8.9% (not shown) indicated that they are providing this type of service. On the other hand, a number of media companies are capturing user names, addresses, and e-mail as shown in Figure 114.

Figure 114. Media companies that capture user names and addresses (percent).

These addresses could be used for a variety of purposes, including future sales, promotional opportunities, marketing research. This information could also be sold to outside vendors, with the knowledge or permission of the user.

Figure 115. Plan to use user information for future sales and promotion opportunities (percent).

Overall, it appears that media companies are doing very little with this information (Figure 115). As can be seen, magazine companies far out-pace their competitors when it comes to marketing opportunities. Although the possibilities for selling this information to outside companies are great, few media companies plan to do so. It should be noted, however, that magazine companies are twice as likely to sell their database information (Figure 116).

Figure 116. Plan to sell user information to outside sources (percent).

By capturing user information, demographics can be determined, especially in the area of site visitation. For example, sites will know who their local, national and international customers are. For those Web sites that capture user demographics, 50% found their customers to be local, while 25% were national, and 15% international. The remaining percent are suspected to be unknown .

Web site comparisons

Based on survey results, as well as findings from the research, step four of this dissertation required the examination of media Web sites by looking at ways in which they disseminate the news, advertising and other forms of information. With thousands of media

Figure 117. A television Web site, KTIV4, Sioux City Iowa.

Web sites providing hundreds of options for users, it becomes impossible to analyze media Web sites in aggregate. However, based on specific survey responses, various cites were reviewed and analyzed as follows.

Generally, it can be seen that television and radio Web sites offer little depth. Most offer news, weather and sports, providing only a few stories. Overall, few offer archival information, but some are affiliated with national providers. As an example, KTIV4 in Sioux City, Iowa (Figure 117) is affiliated with MSNBC. By linking to the MSNBC site, the user feels as though they are on the Sioux City site, even though they are looking at the MSNBC site (Figure 118).

Figure 118. MSNBC link from KTIV4.

The Sioux City site also offers a "Sky Cam", or view of the city. This is something that a number of media sites offer. Though they claim to be "live views", most are still photos that are updated periodically. This seems to be an interesting feature for someone out of the area, particularly for someone who really want’s to see what the area looks like. It does not however, appear to be a big draw for someone local.

On the Web, it is often difficult to tell the difference between a newspaper, magazine, television or radio or other type of Web site. Figure 119, as an example, at first appears to be a newspaper Web site, when in reality it is a Web site that provides culinary information for the newspaper.

Figure 119. WWW.culinary.net Web site.

This unique site offers free co-branding for on-line media. In this way each newspaper is given their own URL on the culinary site, under their on-line masthead, with any other navigational controls that they wish to employ. This provides a seamless presentation to the user, making it easy for them to return to the electronic publisher’s front page.

Advertising on the culinary site is negotiable. If a media company wishes to place their banners on the culinary page, they can do that. Alternately, the culinary site reserves the right to place their own banners on the page. To this point, many of the newspapers using the culinary site do not have banner advertising, so this has not been an issue. Those that do have advertising have not pursued running their ads on the culinary site.

How does the site make money? All of the content is sponsored by food companies and commodity boards. While the content is advertorial in nature, the site is paid for by sponsors to drive the audience to their materials. The site also offers a small classified section related to the culinary industry, and allows users to post classified ads for free. Overall, the site is accomplishing their client's goals as well as their revenue goals, without banner advertising on the co-branded product.

Television and radio sites tend to run contests and giveaways much more than their

Figure 120. A Web site contest.

newspaper and magazine counterparts. This may be related back to their core product in which this type of behavior (contests) is typical. A contest is shown in Figure 120. Users are able to simply fill out an electronic entry form and send it.

A number of sites do not accept banner advertising. Quite often, these are public radio or television stations. Generally, these sites have sponsors to help fund the site.

Niche products are abundant on the Web, ranging from mountain climbers information, to Florida environmental, to the space agency (NASA), to The Black World Today, a unique, electronic only version which targets African Americans and blacks around the world (Figure 121).

Figure 121. Black World Today, a niche product Web site.

 

With registration, some sites offer users the ability to customize their news, and receive e-mail updates daily.

Figure 122. Newspage, an interactive news service that allows users to create the "Daily Me".

 

NewsPage (Figure 122) is an interactive news service from Individual, Inc., providing users with current, filtered news across a broad array of topics. NewsPage utilizes Individual's SMART agent technology to sift through and personalize up to 20,000 news stories each day from over 600 sources. Relevant news articles are sorted into 2,500 news topics and displayed on NewsPage in an intuitive, hierarchical structure for quick access.

The interesting feature of the service is just that, it is a service, not a newspaper, magazine, television, or radio station. The service gathers information from multiple sources (some 600) and allows the end user to set up a personalized issue, where information can be collected and stored in one efficient interface.

Newspage does not create or edit any news stories. Instead, they have relationships with over 600 information providers who send their content electronically to Newspage. The content comes directly and verbatim from these providers.

One of the Web features trying to attract more users is streaming video. Some sites have begun experimenting with this, and, have shown good results (Figure 123).

Figure 123. A streaming video Web site.

As noted in the survey results, only 15% (Figure 17) of newspapers currently provide video. This is due to several factors including cost and speed. Video files are quite large, and, as a result, they can take a long time to download to the user’s computer. This results in user frustration, often causing the user to give up, possibly switching to another Web site.

Many media sites provide weather reports and updates, and some provide live Doplar radar. However, none compare to the niche weather site known as weather.com. A spin off of the television channel (The Weather Channel), weather.com offers a comprehensive look at up to date weather information. Figure 124 shows a page from the weather channel site.

Figure 124. Web site of the weather channel.

 

In reviewing various Web sites, the researcher found several things annoying. One of the biggest problems was in Web sites trying to load cookies to the end user’s PC. A cookie is simply an HTTP header that consists of a string that gets entered into the memory of your browser. This string contains the domain, path, lifetime, and value of a variable that is set. If the lifetime of this variable is longer than the time the user spends at the Web site, the string is saved to file for future reference. Many users do not like cookies being loaded on their computers for a variety of reasons, the most common being privacy.

Often, when moving from one page to another on the same site, the host name disappears. Thus, if the user forgets where they are, or where they have been, they can become lost quite easily.

When it comes to what a particular Web site has to offer (depth), some sites are far ahead of others. Figure 125 shows ohio.com, which has a variety of items to choose from.

 

Figure 125. Ohio.com Web site, with a variety of options.

 

And sites like boston.com provide not only depth within their own site, but also provide links to other partners. Boston.com (Figure 126) boasts over 40 partners that provide reciprocal marketing.

Figure 126. Boston.com, Web site of the Boston Globe.

 

As an example, the Museum of Fine Arts lists boston.com as their Web site on all of their materials. Likewise, ticket sellers advertise boston.com on the back of tickets they sell to concerts.

Boston.com is one of a few sites that offer an abundance of services to their audience. As shown in Figure 126, boston.com gives the user the ability to subscribe to the print product, place a classified ad, get advertisement proofing, send a letter to the editor and contact customer service, among other things.

A number of Web sites offer some sort of free services. As an example kjeo.com (Figure 127) shows a marquee in typical banner ad fashion, from the National Center for Missing and Exploited Children. While the banner stays in position on the page, the child’s photo and personal information rotate about every 5 seconds.

Figure 127. kjeo.com, a Web site that offers free community services.

Another free service offered by many Web sites is free Web pages for clubs, organizations, non-profits and government agencies. This attracts readers to the site in addition to reading the news. Figure 128 shows charlotte-florida.com, a site that is doing just that.

The idea was to find a way to involve the community in the Internet. Judges from across the country and around the world have repeatedly named Sunline's sites as the very best (source www.charlotte-floridaohio.com/internet/winner.htm). And each time, the judges have said they are fascinated with the community site concept. Very few papers have promised to put the entire community on-line for free. Sunline charges nothing to create and host the Web pages for non-profit groups, community organizations and government agencies.

Figure 128. A true community newspaper Web site, www.charlotte-florida.com.

Residents can create their own Web pages in minutes through the use of Sunline's Create-Your-Own-Web-Page site. The result is a comprehensive community site that changes daily, if not hourly, and gives residents reasons to sign on every day. Sunline's sites have caught the attention of judges around the world, including those from the Pulitzer Prize competition, according to information available on the charlotte-florida.com site.

Sunline offers different ways for their customers to create their own Web pages.

Personal Web Pages - limited to Sunline customers only. This allows customers to upload text and photos in creating their own personal Web page. They can also choose a background color, a background image, text color and even the style of a Web page counter. In Memory Pages - On these pages, a photo and text can be placed as a tribute to a deceased loved one or friend. People who know their time on this Earth is short can also leave their own personal good-bys, including sound files. These pages are free for anybody to use. Baby Pages – Instead of carrying around pictures of children or grandchildren, users can now create pages on the Web in tribute to them. These pages are free for anybody to use. Pet Pages - Pictures of pets can be placed on Web pages. These pages are also free for anybody to use.

In an effort to become and remain profitable, a number of sites offer various services, for a fee. As an example, the Aurora News Register offers Web design,

 

Figure 129. A Web site that offers electronic coupons.

 

logo design, letterhead & business cards, scanning, photo prints & laser copies, illustration, brochures & flyers, as well as special packages.

Figure 129 shows another interesting idea, Web coupons. Users no longer have to clip coupons from the newspaper. They can simply use their computer, and with the click of a button, can save money on dinner tonight (Figure 130), or find other promotional offerings.

Figure 130. An electronic coupon.

 

Classified advertising is one area in which print publishers are in fear of their future profits. As seen from the survey results (Figure 69), 71.3% of the newspaper respondents have their classifieds on-line, as compared to only 5.3% for radio, and 9.8% for television. Smaller Web sites tend to just put whatever they have, on-line. This results in the user being required to search through all the ads to find what they are looking for. On the other hand, some sites offer dynamic queries, giving the user the ability to retrieve only those ads with the specific features they are looking for. Figure 131 shows a site that offers dynamic queries.

Figure 131. A classified search engine.

 

As shown in Figure 131, the user can search by geographic location, by year, maximum price, maximum mileage, and a variety of other options. Users will be more attracted to sites offering this or similar capability, as opposed to sites that do not. This was evident in the survey findings (Figure 72) which showed a 115% increase in page views of sites that provided dynamic queries.

Another feature that Web sites should provide in their classifieds is e-mail notification. In this way, a user could put in a request for a certain type of automobile, or range of automobiles, and every time a match comes into the database the user would be notified via e-mail. While this is a nice feature for the end user it could have a negative effect upon the Web site, because users would need to visit the site less often. There may also be an impact to net traffic as a result of e-mail notification.

Selling archival information is another area in which some Web sites are making money. These are generally sites that offer something unique, or something that can be found no where else. The survey found that 93.3% of all the media respondents do not charge for access to their archive, and unless consumers are looking for a hard-to-find article, they will probably try finding it on a free site. As a result, archival information will most likely remain free at many sites, for years to come.

Summary of Results

Despite the extraordinary increases in advertising revenues from 1995-97, media companies (newspapers, magazine, television, and radio) continue to lose large sums of money, and as a result are slowing their involvement in this on-line business. This can be seen in the decreases in promotion expenditures, staffing, alliances, and plant and equipment investment (for on-line product). Further, the majority of media companies are not providing the correct mix of content to their users (Figure 108). The predominant absence of outside user measurement prohibits the facilitation of matching content-to-user, traffic-to-site, and finally web site-to-advertisers. This trend continues in light of editorial and technological developments which are standardizing the on-line product. This standardization has made the on-line product a potentially more viable line of business by reducing costs and increasing efficiency.

However, a small cluster of media companies are responding to these changes and are capitalizing on the opportunities. These few media companies are consistently acting in a similar fashion which support a long term investment and return. Their efforts to defray costs and provide measurement has enabled them to quantify and qualify users, while at the same time minimize their financial exposure through outsourcing their Internet services (consulting, web site construction and design), media alliances, and outside content providers. This process has induced a higher measurable traffic and translates into advertising revenue.

In general, it can be seen that media Web sites are similar in many ways. They all offer news, weather and sports, and most have community interest sections. There are many differences and similarities that exist in the context of on-line products with media companies. These differences exist often because of the nature of these businesses as well as the need to use this medium to perpetuate their respective products. Often times it can be seen that there is no marked differences between any of the media industries, but that in and of itself may explain the level of involvement each may have in the on-line business.

If Web sites are to learn through other companies by example, perhaps the most successful firms can be identified and their behavior towards on-line products revealed. Identifying the commonalties and understanding the differences of these Web sites, may shed light that can be used by other Web sites. Upon close analysis of this group, it can be concluded that those companies that are among the "highest revenue generators" are reacting the same towards issues of alliances, promotion budgets, outside measurement, pricing, cross promotion, and content. Each of these activities shows a significant, positive relationship to overall revenue levels.

Previous Chapter III

Next Chapter V

Return to main menu