How Web Analytics Work in 2019


Over the past two decades, the internet has completely transformed how businesses and customers interact. Every type of business — from local startups to eCommerce giants — require an online presence to attract, convert, and engage with new customers. This means it’s essential for marketers to understand how users are interacting with web content.

This is where web analytics comes into play. Originally designed to measure and interpret website traffic, these tools have expanded to fulfill a broader range of marketing goals. Today, companies use web analytics to:

  • Analyze consumer behaviors
  • Aggregate market research
  • Optimize websites for mobile and desktop devices
  • Conduct A/B testing for online systems
  • Develop and implement core business strategies

In 2019, web analytics plays a role in everything from customer engagement to business performance. Understanding how it works is crucial to sustainable growth in today’s online age.

As part of PostFunnel’s Nuts and Bolts series, we’ll delve into the world of modern Martech to shed some light on tools and best practices being used by you – our fellow marketers – in your day-to-day initiatives. Every month, our experts will sink their teeth into another aspect of this fascinating field, hopefully inspiring you to elevate your business through smart marketing.

Check out our features section with special projects and articles for your reading pleasure.

What is web analytics?

Web analytics is a blanket term referring to the various processes of measurement, collection, analysis, and reporting of web data. At a very basic level, web analytics measures online traffic by compiling visitor and page view data for any given website. Modern analytical tools, however, are capable of far more comprehensive outputs. Today’s businesses can use web analytics to aggregate market research data, generate insights on popular trends, or optimize website interfaces.

How does web analytics work?

For all their complexity, most web analytics processes boil down to four key steps:

  • Data Collection: Web analytics begins with basic data collection. This can include page views, behavioral activity, user demographics, or any other useful dataset. Some data can be ingested through user forms, while other automated processes collect data in the background.
  • Data Processing: Once a dataset is collected, it must be processed into usable metrics. This can include percentages, ratios, or count totals that are understood in a human-readable form.
  • KPI Development: Next, businesses translate their processed metrics into key performance indicators. These can vary wildly depending on your industry or field, but all require that metrics be mapped alongside a particular business strategy (for example, if the goal of a website is to convert leads into customers).
  • Generating Online Strategies: With KPIs in hand, businesses can develop or optimize strategies to achieve broader business goals. If, say, your business goal is to generate revenue from converted users, your KPIs will inform you of progress towards achieving this goal. Once implemented, these strategies will change your analytical data, requiring a new dataset that starts the entire process again.

Some organizations will include additional steps as necessary to improve their analytical processes. Perhaps the most important is A/B testing, where site owners conduct a comparison experiment to determine an optimal path towards achieving a goal.

Which web analytics metrics are the most important?

The most important website metrics vary depending on industry, consumer markets, and other factors. From a marketing perspective, however, the primary goal is typically to attract visitors who become engaged with multiple pages on your website. On this front, are often the most useful.

  • Clicks: Instances of a user explicitly clicking on a hyperlink on your page, either to open another page or visit another website.
  • Hits: Requests to view any file on a web server.
  • Page views: Requests to load a single HTML page that will be displayed in a web browser.
  • Unique visitors: The number of individuals who spend time visiting your website. This metric is especially useful to determine the number of actual visitors, as opposed to running tallies of clicks and hits. Unique visitors can be measured by day, month, or year depending on the nature of your report.
  • Visits: A set of page views that represent a single browsing session for a unique visitor. These begin when a visitor first arrives at a website and generally ends when no requests have been made for 30 minutes.
  • Average page depth: The total number of page views divided by the total number of visits.
  • Session duration: The amount of time a unique user spends visiting a website.
  • Bounce rate: A percentage of users who immediately leave your website after viewing a single page.
  • Exit rate: The percentage of users for whom the visit ends on a specific page.

How do data platforms approach and manage web analytics?

No matter what type of web analytics process you follow, all web analytics technologies fall into the following two categories. Some platforms offer a hybrid combination to both approaches.

  • On-site web analytics: These common tools measure activity once a user reaches your website. They often collect data from cookies, server log files, or embedded page scripts to create in-depth reports. Most metrics can be obtained through this category.
  • Off-site web analytics: Off-site tools analyze and estimate a website’s performance without accessing on-site data. For example, an off-site analysis could determine potential audiences, overall visibility and reach, or buzz generated on social media. In many cases, these are the only tools available for analyzing competitors.

Where can analysts collect useful web data?

Having a web analytics platform is great, but you still need data to input into it. Analysts constantly require reliable, up-to-date metrics to establish KPIs and accomplish overarching business goals. In most cases, web data comes from the following sources:

  • Direct HTTP request data: This data comes directly from standard HTTP request headers once users request access to a particular web page.
  • Network and server data: While this category is necessary to transmit HTTP requests, it only manages associated request data. For example, IP addresses of users come into play at the network and server level, as opposed to the direct request.
  • Application data: Web-based applications and plug-ins often capture and process their own internal data independent of HTTP requests. Web browsers might create cookies storing relevant metrics, while embedded scripts like JavaScript generate specific log files.
  • External data: There is a variety of supplementary data points that exist outside of standard website requests. Some common examples includes geographical region data from internet service providers, email click-through rates, and more.

Is there a difference between web analytics and mobile web analytics?

Since today’s smartphones are generally capable of browsing websites, mobile web analytics bears many similarities to traditional web analytics. That being said, several analytical data points will differ when collected from mobile devices. Metrics like page views remain identical to traditional web analytics platforms, but behavioral data might be vastly different.

Some notable examples of mobile-specific data include:

  • Device model
  • Device manufacturer
  • Device capabilities
  • Screen resolution
  • Mobile service provider

Mobile web analytics is also tailored to address mobile-specific business goals. Any processes that relate to mobile marketing, mobile advertising, or even SMS marketing can benefit from mobile web analytics. In some cases, desktop promotions for mobile services can also apply to the goals of mobile web analytics.

What obstacles do web analysts face when obtaining or processing data?

For all the capabilities of modern web analytics, its processes are not perfect. Marketers often encounter knowledge gaps, data discrepancies, and even problems with data collection itself. Here are a few common obstacles and potential solutions for overcoming them.

The hotel problem

The first obstacle most analysts-in-training encounter is that identical data counts can produce different results simply by adjusting your range. Dubbed “the hotel problem”, this refers to a phenomenon where the total number of unique visitors per day doesn’t match with the equivalent unique visitors per month.

The hotel problem highlights a comparison with hotel chains where unique guests return throughout the month, throwing off the daily and monthly unique visitor counts. Analysts who are processing data manually will need to make separate calculations for each data category. Thankfully, most web analytics tools can address this problem, even though discrepancies will appear when comparing category totals.

Data security

Since web analytics is based on data collection, security is an important issue for analysts. The hazards of privacy breaches and security flaws increase exponentially as businesses collect high volumes of data, even if the data is anonymized to protect individual consumers.

Take cookies, which are a primary data collection channel for many organizations. Some analytics solutions rely on third-party cookies distributed by web domains instead of first-party cookies from specific websites. That means private data stored in cookies is accessible across a wide range of sites, increasing the potential for security breaches. In response, some browsers will block or delete cookies, limiting their tracking potential.

Web analysts have various solutions at their disposal — they could emphasize first-party site cookies only, or turn to alternative sources like IP addresses. But cookies are just one specific security obstacle, and IP addresses pose their own risks. Constant vigilance is required to meet these ongoing challenges.

Unfocused data, or “too many metrics”

One surprising issue occurs when analysts go too far with data collection and processing. Instead of focusing on a few core metrics, some business will collect every single data point, even when they aren’t relevant. This compounds the aforementioned security concerns, and generally wastes time by processing useless information.

Jay Baer of Convince & Convert put it best — just because you can measure something, doesn’t mean you have to. Analysts should streamline their processes by focusing on metrics that specifically meet predefined business goals. Beyond that, being selective is always the wisest choice.

At its core, web analytics is about measuring engagement. In 2019, the tools have become more complex and our hardware is more advanced, but the basic principles remain the same. These tools help marketers understand the drivers behind web-based trends and how users ultimately engage with the internet. Whatever online advances we see in the next half-century, web analytics will be the process we use to understand them.

The post How Web Analytics Work in 2019 appeared first on Post Funnel.


Online enterprenuer. Lean leadership consultant.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.