Measuring Data: Whadda Ya Got?

Why Measuring Website Behavior Matters

The late‑nineteenth‑century retail insights of Paco Underhill taught us that shoppers are a curious mix of habits and impulses. When those habits were transposed onto a website, the same curiosity - plus the power of digital data - opened a new world. Suddenly, every click, scroll, or pause could be quantified, not just observed. The magic of the web is that you can measure behavior with precision that a retail floor can never match. That realization can feel almost like a revelation to a marketer: “Here’s a way to see exactly what the visitor is doing, where they’re stopping, and why they’re leaving.”

However, turning that insight into actionable information is a different story. In practice, a marketing lead will walk into an IT office and say, “What can you tell me about what’s happening on my site?” The IT person, accustomed to metrics like uptime and throughput, may ask back, “What do you need?” The exchange can feel circular: “What’s got?” The root problem is that the marketing team often has no clear specification of the data they truly want. Without a focused question, the IT team runs a list of generic metrics that might be useful, but they might miss the specific insight the marketer needs to drive campaigns or optimize the checkout flow.

To avoid this mismatch, start with a simple question: “What does my visitor do before they arrive, what do they do on my site, and where do they go next?” From there, you can build a data architecture that tracks those steps, using a blend of performance monitors, logs, and user‑tracking techniques. The goal is to shift from a vague “I want to know what’s happening” to a precise “I want to see the path of every visitor from landing to conversion.” With that clarity, the IT team can pick the right tools and deliver the right data.

Once you have that roadmap, the rest of the article walks through the key tools that can help you collect and interpret that data. It covers everything from basic server logs to sophisticated client‑side surveillance, explaining how each piece of technology contributes to the big picture of visitor behavior. The sections that follow are designed to give you a practical, step‑by‑step guide to building a measurement ecosystem that works for your marketing goals.

Essential Tools for Capturing Site Performance and Visitor Data

Performance monitors are the first line of defense against a slow, unresponsive website. They continuously watch server response times, CPU usage, memory consumption, and the health of e‑commerce applications. If a server starts lagging or an API call fails, the monitor flags the issue in real time. That early warning is critical because a slow page can cost you conversions before your visitors even notice anything wrong.

Server logs capture every HTTP request made to your web server. With a log‑file analysis tool, you can turn raw data into useful metrics such as total visits, page views, bounce rate, entry and exit pages, and traffic by country. These logs also reveal the browser type and operating system, giving you a demographic snapshot of your audience. Though the data is reliable, remember that logs come with a grain of uncertainty; a single request can be miscounted if the user’s network drops, or a bot can inflate the numbers. Always cross‑check with other data sources when you see sudden spikes.

Referer logs, or more correctly “referrer” logs, show where a visitor came from before landing on your site. If someone clicks a banner ad or follows a link from a news article, that source URL appears in the log. This data is invaluable for evaluating paid campaigns, affiliate marketing, or earned media. For instance, if banner A generates 200 sessions and banner B only 75, you can re‑allocate budget accordingly. Similarly, if a press release leads to a surge of traffic from a specific keyword, you know your outreach worked.

Tagged URLs are a lightweight way to track click paths across a session. When the server delivers a home page, it adds a unique identifier to each outbound link: “/page2.html?127”. If the visitor clicks that link, the server logs that the user with ID 127 requested page2.html, and the next page’s links carry the same ID. Over the course of a session, this produces a breadcrumb trail of where the user went, which is harder to extract from standard server logs. It’s a simple, effective method that works even when you can’t install JavaScript or other client‑side scripts.

Cookies give you continuity between visits. A cookie is a small piece of data stored on the visitor’s browser, which can be read each time they return. By pairing a cookie ID with any personal data the user provides (like a newsletter sign‑up), you can link behaviors across days and sessions. Cookies also allow you to segment users - new visitors versus repeat buyers - so you can tailor messaging or offers. A quick look at CookieCentral will give you the basics of how cookies work, but for most sites, a single analytics cookie is enough to start tracking a user’s journey.

Packet sniffing takes a step back into the networking layer. Tools like

Tags