Secrets of AirSafe.com:
How Visitors Use the Site

Date: 14 July 2008; Length: 9:08

AirSafe.com creator Dr. Todd Curtis shares his observations of how visitors use the site, and how these traffic pattern continues to influence the content of the site.

Audio: MP3

Transcript

Welcome to the Conversation at AirSafe.com, with your host Dr. Todd Curtis.

This is show #54 - Secrets of AirSafe.com: How Visitors Use the Site

In this show, the third of the "Secrets of AirSafe.com" series, I'll talk about what parts of the sites generate traffic, and how my study of past media coverage of aviation safety and security events shaped the design of the site.

What makes visitor behavior important to the success of AirSafe.com is that visitor behavior, especially the patterns that are evident after months or years of observation, gives me insights into what works, what doesn't work, and the directions the site should take to improve the visitor experience.

Paying attention to visitor behavior is part of the marketing efforts I discussed in the first show of this series. Before the site was launched, I made educated guesses as to what visitors wanted to see. Those guesses were based on research that I did in the months leading up to the launch. That research was very revealing about what aspects of safety drove media coverage, and those insights determined what information would be featured on the site.

In the early days of the web, before alternative media options became abundant online, the most important sources of news for most people were the major broadcast television networks, CNN, and major newspapers.

I'd been involved in aviation safety for almost a decade before I started the site, focusing at first on aviation-related public policy issues. It was during that time that I realized that although television had many more viewers than newspapers had readers, some key newspapers, especially the New York Times, were highly influential with respect to what kinds of stories were considered newsworthy by other newspapers and especially by the television networks.

Based on this reality, I set out to measure what events got covered and how heavily they were covered by the Times. Fortunately, the New York Times made finding the data very easy. For decades, the Times has published a very detailed annual index of news stories, and it is a newspaper index you will most likely see in the reference sections of university libraries and in many larger public libraries.

I took a 17-year period, from 1978 to 1994, and analyzed the coverage of airline events that involved the death of at least one passenger. I chose this period because it coincided with the deregulation of the domestic U.S. airline industry. With deregulation came a substantial increase in the number of flights and significant changes in air carriers and their business strategies. Along with these changes came an increased interest in issues related to air safety on the part of the flying public, the airline industry, and the federal government.

I'll spare you the details of the analysis. If you're interested, you' ll find a link to the study at secrets.airsafe.org. The highlights are easy enough to digest. During that 17-year period, there were 525 fatal events, but only 307 had at least one reference in the Times. A statistical analysis of the 2,752 references revealed that four factors were the best predictors for coverage by the Times: events involving a jet airliner, the number of passenger fatalities, a connection to New York City, and a political aspect to the event.

For the last factor, political could mean something like a politically motivated hijacking, or a civilian airliner shot down by a military organization.

The New York Times disproportionately reported fatal airline events that involved jet aircraft that were hijacked, sabotaged, or destroyed by military action. In fact, six of top ten events in terms of the number of associated references were non-accidental jet events, including five of the top six. These five represented 39.7% of the 2,752 total references of the 17-year study period.

Another way to look at this was that just under one percent of the total population of events were represented in about 40% of all the articles that mentioned fatal airline events.

This study ended in 1994, two years before TWA Flight 800 and seven years before the events of 9/11. Given the four factors associated with coverage by the times, it should come as no surprise that the number of articles that the Times has published about these two events dwarf the number of articles from even the most reported event from 1978 to 1994.

When I finished this study about four months before the site was officially launched, I had no idea what the future would bring, but based on my research, I made the assumption that media coverage was closely associated with public interest, and that I'd have an easier time attracting visitors if I focused on those elements that were associated with extensive media coverage.

Based on what my research revealed, I decided to narrow my focus to those events that killed passengers on airline flights, to include fatal events due to both accidents and deliberate actions, and to closely track events that occurred on jet airliners or propeller driven aircraft models commonly used by airlines in North America and western Europe.

This was a radical departure from what was and is the norm for the FAA, NTSB, ICAO, and other organizations in the aviation safety world. Typically, these organizations would include events involving cargo aircraft, would be concerned with deaths of either passengers or crew members, and would not include events associated with sabotage, hijacking, or military action.

I focused specifically on events with passenger deaths because those are the kinds of deaths that get the most media attention and that also were most likely to lead to public debates or policy changes around airline safety and security issues. The feedback I received during the site's first year confirmed that this was an approach that most visitors approved. There were certainly those who disagreed and thought that such an approach was biased. The argument, one that I agreed with, was that events were very rare, with most airlines having no fatal events in their history at all. It was also a fair criticism that the presence of a fatal event doesn't imply that an airline is excessively risky, nor does the lack of an event implying that an airline is risk free.

These were fair criticisms, but the reality for my site was that the behavior of the site's visitors mirrored the kind of results I saw in my study of the New York Times. Events that resulted in passenger deaths or that were covered heavily covered by the major media were the same ones that consistently got high traffic numbers on the site. Over time, I became convinced that to the average person it really doesn't matter if passenger deaths are from random accidents or from premeditated or deliberate acts. The threat of death, not the cause of death, is the thing that concerns the general public.

While I made many efforts to put the data into context, and to give visitors insights into what the data could or could not say, the bottom line is that by following the site's philosophy of giving the visitors what they want, and just a little bit of what I thought they needed, I've been able to attract and expand the audience for the site.

The original site was built around a collection of pages that detailed fatal events by aircraft model and by airline. Over the years, I included other sets of information concerning airline safety and security. Some of these additions were specialty pages such as a page listing fatal U.S. and Canadian bombing events or a list of jet airliner ditchings. Others covered broader subjects and sometimes expanded well beyond a single page.

The most prominent of these subjects on AirSafe.com is that of celebrity plane crashes. I had my doubts that it would get significant traffic, and even questioned whether it would reduce traffic to the other parts of the site. I was happily mistaken on both counts. It has proven to be one of the more popular sections of AirSafe.com. When I first started this section, the event that had by far the most traffic was a December 2000 crash involving Sandra Bullock. More recently, the January 2008 incident involving Senator Barack Obama has only grown in popularity as his presidential campaign has gained momentum.

In the last few years, the site expanded to other areas dealing with passenger issues such as fear of flying, air rage, baggage issues, and security rules. This was an ongoing process over the last 12 years and this evolution will continue in the years to come. As other areas become relevant to the traveling public, these areas may be added to the site.

I learned early on that it was difficult or impossible to predict what kinds of content would attract an audience. My experience with the celebrity plane crash section was the main reason why I rely on audience reaction to determine if something will work for the site. If the content in a new subject area doesn't generate traffic, then I won't add any more content. If it generates significant traffic, then I'll simply give the audience more of what it wants. If there's a little something extra that I think the audience needs, then I'll throw that in too.

In the next episode in the Secrets of AirSafe.com series, I'll talk about what it takes to become an internationally recognized airline safety expert, or for that matter an expert on any given subject, and how the Internet makes it easier than ever to become an expert.

For more information about this show, and about other AirSafe.com podcasts, please visit secrets.airsafe.org. There you'll find links to related resources, as well as AirSafe.com contact information.

Thanks for listening, and I'll see you next time.

Additional Resources
Secrets of AirSafe.com Special Podcast Series
Other AirSafe.com Podcast Episodes

Contacting AirSafe.com
Want to leave a voice mail or email? How about mailing a letter? Go to the AirSafe.com Contact Page and send us your message.