How is your web business doing? Are you stuck in traffic or riding in the fast lane?
It’s a simple question that is often difficult to answer accurately. All too often, Internet business performance is measured, by the media and marketers, in terms of traffic.
Most of you are probably aware of the dizzying array of audience measurement products and services available on the market today. (If not, send me email, and I’ll cover this topic in a future column!) The next challenges are what to do with this data and how to go beyond it. There are at least three basic limitations to the value of traffic measurements.
Many factors influence traffic, and most of them are not under your control. Suppose that an online retailer of computer games gives its web site a face-lift in November. In December the number of unique visitors to its site increases 15 percent. Does that mean the face-lift was a success?
Not necessarily. The increase might have resulted from the site redesign, but an equally likely explanation is that a rush of holiday shoppers boosted traffic. Maybe it was a mixture of both. Maybe it was neither. Many outside forces can affect trends in traffic: Internet-wide growth (or shrinkage), seasonal effects, and changes in competition, to name just a few.
The first step toward creating usable metrics from traffic data is to put the traffic data in context. Panel-based audience measurement services, such as Media Metrix, allow you to compare your site usage to a relevant comparison group’s.
If the comparison group is affected by the same outside factors that your site is, differences between the sites reflect interesting trends. You might be happy with slowly rising site usage until you discover that all your competitors’ traffic is skyrocketing. Depending on your business goals, you may want to compare your traffic trends to those of your competitors, all e-commerce sites, or the entire Internet.
Your own historical data can also provide a useful reference point. One way to sort out seasonal effects is to compare usage trends in one period to the same period in a previous year. For example, our video game retailer could gain some insight into the 15 percent increase from November to December by comparing this change to last year’s percent change from November to December.
Quantity Versus Quality
A serious drawback of traffic statistics as a measure of performance is that it misses the element of customer value. A web site measuring success with hits is like a brick-and-mortar store measuring success by counting people who walk through the front door. Sure, it’s part of the story, but that count by itself is very misleading.
Looking more closely at click-stream data can provide some insight into customer value. The number of hits to your home page roughly measures the volume of potentially interested customers. Those visitors who are actually interested in your site clearly represent greater business value. Customers who are actually interested may identify themselves by reaching certain pages (for example, transaction-related pages) within your site.
Tallying the proportion of users who reach key pages within your site adds an extra dimension of value to raw usage statistics.
The Why Behind the What
A fundamental limitation of traffic measurement is that it only tells you what site visitors do — how many people visited a certain page, where they came from, and where they went next. Traffic statistics are much less clear about why users did what they did. But why customers did what they did is a critical piece of information for making decisions about your web site.
What would you do if you saw your site traffic steadily declining over time? Unless you know why visitors do what they do, you’re only guessing. To obtain actionable information from users, you need to listen to what they have to say. There are many ways to do this on the web: online focus groups, web-based surveys, and email surveys are just a few examples.
It is not enough to count customers’ heads — you have to get inside their heads. In this column we will cover all the aspects of this process, from sampling strategies to survey design to analysis ideas.
Next week we’ll talk about sampling methodologies for customer research projects.
Sponsored content in collaboration with Marchex. When it comes to brand keyword bidding, most tests show that it makes sense to bid ... read more
Emily Ma, product director of Tencent’s advertising platform products department, was a keynote speaker at ClickZ Live Shanghai where she discussed the ... read more
The terms that customers type into your site search function can help you to gain an understanding of user behaviour and can be used to optimise ... read more