I took some liberties with this assignment. Since I have been tasked to replace the current BBC website, I couldn’t bring myself to do a thinking aloud on a site that was essentially the walking dead. Rather, I decided to ask my users to pretend that they were either “New Cyclists” if they were inexperienced or non-riders, or that they were experienced riders moving to Seattle or Chicago. I chose Seattle, because it has the Cascade Bike Club, which is the most sophisticated bike club website that I know of. Chicago was chosen because it has a large cycling presence, but no dominant club.
Before I had my users go through their tasks, I contacted about 5 additional users regarding how they would get information about starting a new activity such as riding. These individuals were both male and female, and ranged in age from 19 to 54. Without exception, all said that they would prefer to ask someone they knew about the activity. If a friend was not available, then they would then try a variety of means. Several said that they would go to a bike shop first. The general bias seemed to be: “Talk to some people first to get a sense of things, then go to Google to find out more”.
The only exception to this was the two experienced riders. They already knew what they were looking for and went directly to the Web.
All users went to Google. All searched for “xxxx cycle club” or “xxxx bike club” where xxxx was the closest city. These included Annapolis, Columbia, and Baltimore. In the case where the users were already familiar with the local cycling scene, the tasks were set in Chicago and Seattle as described above.
Interestingly, one user started out by using Bing, but had terrible search results. A search for “Columbia MD Cycle Club” brought up results for Columbia MO, and Columbia TN, and USA Cycling, which is the USA bike racing organisation.
The main takeaway is that there are really three groups of users. New Riders want a simple overview of the club and rides. They want to see what kind of person is a member, and if the club looks like the kind of organization that meets their needs. Experienced Riders want to know what the rides are like first, and what the club is like second. Club Members need clear access to club activities such as leading rides, that the other two groups don’t need to know about.
Beyond that, an important point is the importance of simultaneously providing information in a clear, uncluttered way, while also providing numerous ways to look at that information. Some users like lists. Some like calendars. Some want ride descriptions while others want maps and ride profiles. The desire to find data by searching rather than running down links seems common. One user actually did Google site: searches to find information in the club website.
The most curious piece of information is the story of the Cascade Bike Club. The CBC is a technological marvel of bike websites, in that it actually has employees to maintain the site. As such it has a lot of capability, and has blogs, a Facebook presence, even a Wikipedia entry. Interestingly, all the ancillary websites actually rank higher than the main CBC website. This makes finding the CBC site from a generic search problematic. Allthough nearly half of the Google hits for Seattle had a mention of the Cascade Bike Club, only the Wikipedia entry had a link back to the home page. The Facebook page and the blogs did not appear to have one.
All data was entered in a spreadsheet to support dynamic aggregation based on sorting. The spreadsheet is too big(!) to take a screenshot of, so here’s a link to an uploaded copy. If this doesn’t work and you’d like to see the spreadsheet, let me know and I can email it.
All in all, I found this exercise to be extremely useful. It pointed out how people find this kind of information and gave me lots of good and bad examples of how to do things. In the process, I have made significant revisions about how the new BBC website should be organized, and how it should work with its ancillary Facebook/Twitter/Strava/RideWithGPS/Instagram web footprints.