Example Google Analytics statisticsOne of my very favorite bloggers (Philocrites) recently requested average rates of daily traffic. He specifically is wondering about Unitarian Universalist blogs, but this takes me down a parallel path: statistics for church websites.

If you’re like me, you think of analyzing web statistics in the same breath as visiting the dentist. Actually, that’s not quite fair. I like my dentist a lot. Nonetheless, I wish I didn’t have to go so often. So like visiting the dentist, gathering web statistics boils down to figuring out how to do it as painlessly and efficiently as possible.

But let me back up. The first question is, if it’s that bad, why bother at all? I actually have a few sites where I don’t care about the statistics and make little-to-no effort to gather them. Heretofore, this very site has been one of those. It’s so new. But Philocrites has asked, so I’d best get moving.

For the sites I maintain professionally, statistics are a must, but in my opinion, they aren’t as crucial for the average church website. Don’t get me wrong. The figures can be incredibly useful, and there are likely to be times when they are really needed.

So When Are Statistics Needed?

In my experience, you’re going to need to know your congregation’s website numbers if:

  • A major stakeholder (e.g. the minister) asks you for them.
  • You are about to do redesign and need to know what browsers to code for or what parts of the site are most used.
  • You enjoy puttering around with this kind of thing. Some people do. I do when I have time. Similarly you might be curious and have the time.
  • Your site is large.
  • Your church wants to grow in a big way.

You don’t need two of these reasons. Any one of them is enough in and of itself.

What Are the Options?

Digging a bit deeper, it’s helpful to understand the basics of how web numbers are crunched. Actually, the more formal term for this is “web analytics,” which can be broken into two basic methods.

Logfile Analysis. The first, more traditional method is to use software that analyzes the “log files” on the server. Anytime someone or something visits your site, a very particular set of data is automatically recorded in a set of lines in a log file on the server. Typically it’s the IP of the user, the date and time, the files they accessed, where they came from and their browser and platform.

On any given day, a site with moderate traffic will have thousands of these entries. They can be parsed quite easily by software and number-crunched. The catch is your server administrator or host has to give you some kind of access to the log files. Some do; some don’t. Also the log files mount up over time and can have a huge impact on the server. They’re a bit like kudzu. Server administrators have to prune them, so you may only have access to a limited time’s worth of data that’s less meaningful.

Page tagging. The other approach is to include a kind of bean counter (typically JavaScript) that’s embedded in the pages you want to track. Webmasters have more control over their pages’ code than their servers, which often makes this method more viable. In addition, the range of statistics gathered can be broader. However, you can’t append this code to all files, so you’re likely to miss some of the important data.

But What Are Your Best Options?

In practical terms, this means that your best bet is to first check out the logfile analysis your host provides. If you can’t find it, check the host’s documentation or ask customer service. What particular analyzer they provide to do the number crunching varies a great deal. Mine uses Analog, but alas, I find Analog close to useless. Analog is venerable — one of the first and in its time one of the best — but, in my opinion no longer so. AWStats (which I find easier to interpret) is another common analyzer used by hosts. Of course your host may use something else entirely.

If your host’s analytics don’t meet your needs, then it’s time to consider page tagging. Google Analytics and StatCounter are among the most popular in this category. I’ve used both professionally, and have been amazed at how helpful they can be. About a month ago I added Google Analytics to my church’s code, and plan to report back soon on how I did that and how it’s working.

One side note: blogs, because of the unusual ways they are accessed, are a somewhat different beast. FeedBurner is generally considered to have the best statistics for blogs. But that’s beyond the scope of this post, since church sites aren’t usually blogs – even mine where we use WordPress.

Once You Have the Statistics, What Do You Do with Them?

In the wealth of numbers, charts and graphs generated by analytics, the items webmasters tend to look for most often are:

  • Unique visitors. This tells you roughly how many people are coming to your site, which can be a particularly helpful statistic over time.
  • Page views. This is not quite as useful in my opinion, though again, it can be a helpful over time. These are the number of pages or files each unique visitor accessed. So if 5 unique visitors each looked at 10 pages, that would be 50 page views.
  • Most accessed pages. This can be quite useful in figuring out where to put the most energy into the site. For example, my church discovered to our horror that our calendar was one of the most visited pages. It was in terrible shape, so we made it a priority to improve.
  • Search terms. How are people finding you? From this you can learn things like keywords to feature.
  • Browser and platform. This can have an impact on redesign in particular. For example, no one is using Netscape 4 to access our site any more, so I no longer have to code for that bear.

So that’s a crash course in web analytics. Over time I plan to cover related topics in more depth, but until then, here are some….

Further Resources