Measuring Website Performance

Measuring performance of government websites is a complex issue. Many web managers look at usability testing and customer satisfaction scores as significant measures of website performance. Indeed, they do measure how effective your website is. But is that really the measure we should be looking for? From a budget point of view, it seems to me that we have to look for ways to quantify how the website helps the agency further its mission. It’s entirely possible to have a website that people like and that they can use, but it doesn’t advance the mission of the agency. Is that really money well-spent? The problem is: how do you get that data? HUD’s mission is to increase homeownership and affordable housing and strengthen communities. The problem is that we never could find a good way to show that people were buying homes or finding affordable housing, because of the website.

Gerry McGovern, who spoke at our government Web Manager Workshop in the spring of 2005, pointed us to measuring completion of tasks. Are website visitors able to complete the tasks they want to complete? I like this idea because it really focuses on the bottom line of any website: do you have the right content?

I still have this nagging concern, though. In all these measures – what is the impact of reader’s expectations? I remember getting a message from an outraged website visitor just last summer. He had come to our website to find a foreclosed home in his area, and we had no foreclosures available in his area at that time. His reaction was, “I’ve seen ads that the government has thousands of foreclosed homes for sale, and you don’t have a single one listed in my area. Bad website!” He completed the task – he was able to find and search the listings. But he wasn’t satisfied with the outcome; therefore, he wasn’t satisfied with the website. I don’t think he’s an exception – I think expectations play a big role.

The bottom line – the reason we measure websites at all - is this: is the agency/are taxpayers getting adequate return on investment? Maybe it depends on where you sit. Take the case of the guy who couldn’t find a foreclosure listing. From the agency’s point of view, he was successful. He found the information and was able to use it, without having to draw on more expensive human resources. From his point of view, the experience was not a success. He didn’t feel he was well-served. Does the latter trump the former? Does the former trump the latter?

Hmm…what are we really trying to measure?