Complete Web Monitoring

Chapter 4. The Four Big Questions

As you saw in Chapter 3, there’s a dauntingly long list of metrics you can track about your website. All of those metrics answer four fundamental questions about your site’s visitors:

• What did they do?

• How did they do it?

• Why did they do it?

• Could they do it?

Without the answers to these four questions, you can’t tell where to focus your efforts. You don’t know what’s working well or what’s hopelessly broken, and you can’t tell how to improve your business. Most companies that can’t answer these questions try new strategies randomly, hoping to hit on one that works.

Armed with the answers to the four big questions, you can make informed decisions about where to focus your efforts. Would better design or better infrastructure improve your conversion rates? Are your competitors winning against you before prospective customers ever have a chance to see your offers? Do visitors prefer to search or browse? Are buyers leaving because they didn’t like your prices, or because they came to the wrong site altogether? Are your biggest subscriber’s users’ complaints your fault—or theirs?

These are the kinds of decisions that make or break a business. To make the right decisions, you need to answer the four big questions.

What Did They Do?

The first question concerns what your visitors did, and it’s answered through web analytics.

Visitors’ actions speak for themselves. Analytics shows you what worked best, but won’t tell you why something worked. The only way analytics can help improve your site is by showing you which content, messages, designs, and campaigns have the best impact on your business. This is because analytics lacks context—it won’t show you what was on a visitor’s mind, or whether the site was fast during her visit, or how easy she found it to use.

Analytics was once relatively simple because web transactions were simple. Three things have changed in recent years that complicate matters, however:

Visitor interactions aren’t just requests for pages

Gone are the days of straightforward page-by-page interaction; instead, visitors stay on a single page, but interact with page components through DHTML, JavaScript, or plug-ins.

Visitor impressions start long before users visit your site

To get a complete picture of a visit, you need to know what people are saying about you elsewhere that led a visitor to your site and set the tone for his visit.

Visits don’t follow a set path

Instead of browsing a predictable sequence of pages to arrive at a result, visitors explore in a haphazard fashion, often relying on searches or recommendations to move through a site.

While the basic building block of analytics is an individual visit, analysts seldom look at web activity with this much granularity. Instead, they look at aggregate analysis—metrics grouped by geography, demographics, campaigns, or other segments. You’ll only look at individual visits when you’re trying to reproduce a problem or resolve a dispute. Web analytics is more focused on large-scale patterns of interaction.