Design of Museum Web sites

Methods part 1 || Methods part 2

a Quantitative method:

Measuring how Users behave

Before I presented Griffiths' article in order to show some of the information that the museum web development community "knows" about museum web site users. The methods used for understand and measuring users of museum web sites in the Use, Usefulness and Value of Museums in the U.S. study were primarily qualitative. There are however many quantitative methods of measuring and understanding users.

Web analytics are a traditional way of collecting and analyzing information about how users interact with a website. "Data collected almost always includes web traffic reports. ...user performance data such as click heat mapping, or other custom metrics as needed. This data is typically compared against key performance indicators for performance, and used to improve a web site or marketing campaign's audience response." (Wikipedia) I will not get into the varieties of web analytics but if you want to learn more read wikipedia's article on web analytics or check out the company Haynes and Zambonini write about in their article "Why Are They Doing That!? How Users Interact With Museum Web sites." Below is a synopsis of their findings as an illustration of the quantitative measurements available through collecting web analytics for a web site. Or in Haynes and Zambonini case five web sites.

Click Data and Web Analytics

Haynes and Zambonini detail the results of a data collection effort involving five museums.
|| the Imperial War Museum || Museum of London || Powerhouse Museum ||
|| San Francisco Museum of Modern Art || the Science Museum ||
The authors focused on the potential of web analytics, particularly click-data, ability to inform developers' understanding of the average museum web site user. Such an understanding is central any user-centered design effort. Click-data offers a new way to automatically collect user behavior quantitatively. Particularly, click data allows a modeling of site "hot spots" and accompanying graphical models of data that communicate details about user to the development team. They sought to define the average profile of a museum web site visitor and compared to the typical profile of a non-museum web site visitor. (Haynes)

Figure 11 from Haynes and Zambonini's article titled Why are They doing That!?

"Where visitors click on links, text boxes and buttons" -(Haynes figure 11.)

Haynes and Zambonini's findings

Platform differences: Visitors of museum web sites are five times more likely to be mac users than pc users. (Haynes figure 1)
Of the Mac users they are five times more likely to use Safari than Firefox.
When users visit: Visits to both museum and non-museum web sites are at their peak mid week and their low on the week end. But, based on the percentage of visits to museum and non-museum websites, museum web sites bring in two times as many visitors during the weekend compared to visitors to non-museum websites. (Haynes fig. 3.)

Museum web site user activity:

The Visiting section (museum hours of operation): is the most clicked on area of a museum web site homepage. "An average of 17% of all museum homepage visitors clicks on the links for visiting or open times." (Haynes) These links are also the quickest to be clicked on from the when the page loads to clicking a link. In the Visiting section of a museum website, maps are the most clicked area, still visitor copy the street address information from the webpage. So, street addresses should always be available to users. Search boxes are less used in the Visiting section, as compared to the Museum Home page. On a museum web site Exhibits, Galleries and Events are the next most clicked on links after Visiting.

Figure 10 from Haynes and Zambonini's article titled Why are They doing That!?

"Distribution of when users click on the homepage"-(Haynes figure 10.)

Lessons from Haynes and Zambonini

The measurement and alaysis of click data allows web delvelopers to make better assumptions about user behavior. Their study sought to understand generic user behavior that would be common to almost any museum web site. The authors through their study found that there is a distinct difference between the standard web site user and the standard museum web site user. Museum web site users use Firebox more, and their museum web site user is more likely to use a Mac than the standard web site user, i.e. internet user. Museum web sites attract more traffic on weekend than the average of non-museum web site traffic. This concurs with the Griffiths finding that the majority of museum web site visitors visit for recreation or personal enjoyment. The Haynes and Zambonini article also taught use that the Visiting section is very popular with museum web site users. The authors conclude with "analysing the use of museum Web sites- and importantly the reasons for such use- is a neccesary but complicated process."(Haynes) Thus implying that web analyitics is not the only important method for analyzing user behavior.

Move on to the next page.