TechSoup.org The place for nonprofits, charities, and libraries

A Few Good Web Analytics Tools

Measure the success of your nonprofit or library website

 
Laura S. Quinn and Kyle Henri Andrei - May 19, 2011
Web analytics tools help you track your site's statistics, which let you see how many people are looking at each page, what sites they came from, and other information to help develop a picture of who your audience is. But which web analytics tool should you use? This article, completely updated in 2011, summarizes what six different nonprofit experts recommend.

Web analytics tools help you track your site's statistics, which let you see how many people are looking at each page, what sites they came from, and other information to help develop a picture of who your audience is. But which web analytics tool should you use? This article, completely updated in 2011, summarizes what six different nonprofit experts recommend.

How many people visit your website every day? What do they do there? Which features are most popular? Was that big redesign worth the money? If these questions keep you awake at night, you're a good candidate for a web analytics tool. Such tools track your site's key metrics to help you measure traffic, understand your visitors' needs and behavior, and gauge click-through rates to new content or features — in short, they help you answer these questions and others like them.

Data-tracking needs are similar for all organizations, including nonprofits, libraries, small businesses, and corporations. But given the vast array of analytics tools out there, selecting the right package can be overwhelming.

Idealware talked to six nonprofit experts about the web analytics tools they've seen work well. We also consulted postings on a number of nonprofit listservs and scoured reports on the topic. In this article, we summarize what we've learned to help you understand what to consider when choosing an analytics package, and identify free tools and applications to help you better monitor your site's visitors.

Which Data Should You Analyze?

There's no point in looking for a tool unless you have a sense of what information you want to track. Needs can vary from simple traffic-monitoring to complex analyses on the behavior of specific user groups, support for multivariate testing, and more.

What important metrics and figures should you keep in mind when selecting a web analytics package? We've broken them down into three overarching areas to track, with multiple metrics to help you understand different aspects of each.

The first thing you want to track is an accurate measure of how many people are using your site, which is neither as easy nor as clear-cut as you might think. Metrics which address this statistic include:

  • Hits. This metric probably doesn't make sense to track, but you'll hear a lot about it - a hit measures the total number of requests for text, images, and files your web server receives for a given page. Despite what software packages may lead you to believe, hits are virtually meaningless when it comes to actually understanding what users are doing on your site. Because the number of hits a site receives depends on how it's organized rather than how visitors interact with it, this metric is useful only in evaluating such information as server load.
  • Visits. The most common unit of measurement in site analytics is the number of visitors to a particular site or page. The trend in the overall number of visits to your site over time can give you insight into your site's popularity. Comparing the number of visits to each page is also a good way to identify which parts of your site are most useful to visitors.
  • Unique Visitors. This is the number of site visits by different users. If two people visit the site three times each, you'd have six visits by two unique visitors. Comparing visits to unique visitors can help you understand whether users are returning to your site over and over, or whether you are attracting a large number of users who only visit once or twice.
  • Page Views. This is the number of times any page was viewed by any visitor, and is often divided by visits to give a page-views-per-visit figure that represents the average number of pages each visitor viewed on a single trip to your site. Increased page views can indicate a more interesting site, or simply one that requires people to jump through hoops to find what they need.

Next, you'll want to track who the visitors to your site are, in broad terms, and what they're doing when they visit - in other words, what site features and pages engage them? Which ones go ignored? Metrics to help understand these factors include:

  • Bounce rate. Bounce rate is the percentage of visits where the visitor left your site after viewing only one page. This metric is typically used to measure visit quality. For example, a high bounce rate might indicate that site entrance pages are either not relevant or compelling enough to your visitors. On the other hand, if you have a blog or article-based site, it may make sense for visitors to come, view one article, and leave
  • Top Entry and Exit Pages. This refers to the pages on which most visitors enter your site — don't assume it's the home page — and leave it. These pages can be good places to begin when you are optimizing your site.
  • Visitor Information. You can discover a lot about your visitors through analytics tools, including how many are new to the site, the country or region where they're located, the web browser they're using, and much more.
  • Click Paths. Also called click tracks, or click trees, these are graphical representations of typical journeys through your site. For instance, a click-path chart might show you that 20 percent of your home page visitors go on to click the Resources link, while 15 percent visit the About Us page — and that 60 percent then leave the site and 10 percent go on to the Board page.
  • Conversion. This is a complex-but-valuable statistic that typically needs to be customized in a tool or calculated by hand. Conversion tracks the number of people who did what you wanted them to from a given starting point - for example, the number of users who went from a Donate link on your home page all the way through the donation process, or the percentage of people who viewed your home page and then signed up for your newsletter.
  • Tracking Registered Users. If parts of your site require users to log in, a web analytics tool can track exactly what they did during each visit to the site. (Without a login, it's not practical to link up data for a particular person from one visit to another.) This can allow for more detailed analyses and understanding of what different types of visitors are doing on your site.
  • Site Search. Some packages allow you to see what people search your site for. This can help you understand what visitors are looking for, and what they are having trouble finding.

Lastly, it can be beneficial to track where visitors to your site are coming from. This can help you find similar sites or better understand the types of things that lead people to you. Metrics include:

  • Referrers. These are the external links people follow to get to your site. For example, if TechSoup links to Idealware's site, TechSoup would show up as a referrer in Idealware's web stats. This metric can be very useful in tracking a big influx to your site or just in staying on top of who's talking about you.
  • Search Keywords. Many packages can show the words or phrases people typed into search engines like Google or Yahoo! that led them to your site.

These metrics should be enough to get you started, but powerful web analytics tools support even more sophisticated analysis. There are people who make a living analyzing web statistics — if you have a large site and the desire for deep usage analysis, you may wish to consult with one of them.

The world of analytics is complicated by the fact that not every software tool handles metrics in the same way. Determining what sequence of web actions to interpret as a "visit" or a "unique visitor" is complex, and somewhat subjective. Different tools calculate these figures differently. Some types of software — called "log analytics" software — look at traffic based on a log of what pages your web server provides, while others rely on what's reported back by "cookies" — pieces of information sent back by each user's browser. Don't be surprised if your metrics vary somewhat among tools.

Software Offered by Your Web Host

So now that you know what to track, what web analytics software should you use to track it? You may already have some of the tools you need. Many shared hosting companies, like DreamHost or LunarPages, offer web statistics through the same control panel you use to administer email addresses and check available file space.

AWStats and Webalizer are the two most common, and both are relatively basic "log analytics" packages that offer information about visits-over-time, most-visited pages, referrers, search strings, and some data about your visitors' browsers and locations. Webalizer is a bit more popular, but AWStats's reports are generally considered somewhat easier to understand.

Because these built-in tools are purchased and maintained by your Web host, there are no fees or installation required. While basic, they are perfectly adequate options if you simply want to keep an eye on your site.

Site Counters

You may have seen site counters on the bottom of web pages — basically, they're a numerical display that counts, and shows, the number of visitors to your site. A quick word about site counters: Don't use them. All the free tools listed here will give you the same information without interfering with the look of your site.

Google Analytics

In a class by itself, Google Analytics offers substantially more functionality than the basic tools, and is free — unlike most of more-advanced tools listed below.

Unlike Webalyzer or AWStats and other tools provided by hosting companies, you need to install Google Analytics on your site. This involves pasting a chunk of HTML code provided by Google into every page. This piece of code sends back "cookie" information to the Google server. Adding the code requires basic HTML know-how, but Google's directions are pretty detailed and clear, and the process shouldn't require a huge amount of time. Depending on the size of your site and how it's set up, installing the code might take anywhere from a couple of minutes to a few hours. But once it's added, Google displays your statistics in a custom reporting interface you can view online.

In addition to the reports offered by tools like Webalizer or AWStats, Google shows how often visitors come to your site, tracks visitor conversion across a series of pages, compare the behavior of different types of visitors (such as new vs. returning, or those from different referring sites), and much more. A selectable date range allows you to analyze any given time period instead of being limited to a monthly view (as you are with AWStats and Webalizer). Almost every set of metrics can be sliced and diced to drill down to exactly what's of interest to you.

New features continue to be added every few months, creating an increasingly powerful tool. The powerful reporting and filtering tools let you set up custom reports and segments to view specialized subsets of your data. It won't easily let you track advanced Flash sites, and does not automatically track downloads of files like PDFs, but otherwise is a great free option for analytics.

More Powerful Analytics Packages

Large organizations may want to look beyond Google Analytics for more powerful features. There's no shortage of available options. More advanced tools like Lyris HQ Agency Edition (formerly Click Tracks), WebTrends and SiteCatalyst provide substantially more: more control, more powerful metrics, much more freedom to perform detailed user segmentation, the ability track detailed patterns, and in many cases, sophisticated data charts such as trees or interactive layouts that make it easier to track complex sites.

They'll also support complex or Flash-based sites that Google Analytics will not, and offer professional technical support, as well. More advanced features come at a cost. The lowest-end packages start at about $100 per month, and you'll likely need to pay $500 per month or more to get features that rival Google Analytics.

Choosing a Package

The first question to ask yourself when deciding on an analytic package is, will Google Analytics meet my needs? If it will, there's no point in spending money on a more advanced tool. In fact, Google Analytics is a good default option for a lot of organizations.

If your needs are even simpler, check to see if you have a control panel and analytic tools available through your Web host. Or, if you're familiar with web statistics tools and want more than the analytics and limited control that Google offers, picking a more powerful analytics package might be the way to go.

The right package can make a big difference in your ability to understand visitors' needs and your site's traffic. Choosing the right option means you'll be able to track exactly what people are doing on your site, get all your questions answered, and maybe even sleep a little better.

For More Information

Thanks to TechSoup for their original financial support of this article, as well as to the Annie E. Casey Foundation for funding its update. We also thank the nonprofit technology professionals who provided recommendations and advice on this article and the 2009 version of it:

Image: Charts, Shutterstock