Site Stats Share the Bad News, Too
This chapter is from the book
This chapter is from the book
This chapter is from the book
Site stats share the bad news, too
- Just as analytics tools and site stats are an essential aid in optimizing a site’s searchability, so too can these tools flag warning signs, bad news, and other signals that the website in question is not meeting searchers’ expectations.
In such cases, bad news is not necessarily a bad thing. After all, we’re talking search engine optimization. By definition, SEO is a continuous, ongoing process. Without any negatives, what would you optimize?
In the broadest sense, the goal of SEO is to increase organic search engine referrals to a given website. Therefore, tracking unique visitors from search engine referrals as a percentage of a site’s total unique visitors is an essential part of measuring the success of any SEO effort. You want unique search engine referrals to rise, or at the very least to hold steady. It’s worthwhile noting that unique search engine referrals are measured not in their totality but as a percentage of traffic because nearly every business is affected by cyclical variance, ranging from news cycles to seasonality. If the only thing you’re selling is snow tires, you’d expect a decrease in traffic (as well as in snow tire-related searches) in the summer months, wouldn’t you?
The traffic to your site is rising steadily. Great news, right? Well, not so fast. If traffic to the website is going up, but sales, lead-generation, or the other goals of the website are flat or in decline, something’s wrong. It could be site issues, of course, but if flat performance is directly linked to organic search engine referrals, it’s all but a sure sign that the site has been optimized for the wrong keywords.
Keeping an eye on those pages users visit—and those they don’t—is another important set of stats to track. Search engines drive traffic to specific pages in a site, not just to the homepage (see Truth 8, “You don’t have a homepage anymore”). This holds advantages both for the user and for the site owner. Through search, users are able to navigate directly to the page that most specifically addresses their search query. They needn’t bother to poke around a site to navigate to relevant information.
The edge for site owners is that each and every page of a website affords new and specific opportunities for optimization. Each page has a specific and individual set of keywords that can be adjusted and optimized. Copy can be revised and rewritten. Both outbound and inbound links can be improved. Certainly no site gets consistent levels of traffic across all pages. Some are simply bound to perform better than others. But that doesn’t mean underperforming pages deserve to be ignored. They should be considered optimization opportunities.
Even the most rudimentary site analytics tools provide more data than many site owners are prepared to deal with. What matters is determining which datasets are important, and then carefully monitoring and tracking those statistics on a regular basis—daily is great; monthly at a bare minimum.
Overall, the most fundamental site elements to track include the following:
- Visitors—What could be more basic than knowing how many visitors have come to your site in a given time frame? Be careful to differentiate between unique (or individual visitors) and overall site visits, which account for the total number of visits to the site, combining unique visitors with people who visit multiple times. Although knowing the number of visitors to a site won’t help increase conversions, it’s a good indicator of how search engine rankings and links from other sites affect traffic.
- Referrers—Where’s your site traffic coming from? This metric lets you know which search engines send how much traffic to your site, as well as which links to you on other sites impel users to visit.
- Keywords—Ranking well for keywords is one thing. Getting traffic from them can be something else altogether. Keyword stats—those words and phrases searchers are using to find you—let you know exactly how much traffic keyword rankings actually generate.
- Time spent and bounce rates—Traffic is only part of the goal. Encouraging visitors to actually stick around and do something, such as buying or otherwise converting, is a much more critical benchmark of site performance. Assessing how much time visitors spend on a given page, or on the site overall, is a strong indicator of how well it fulfills the expectations of users who followed a link from a search engine, or any other referrer, for that matter.
- Exit pages—Exit pages can be a strong indicator of flaws in a website. This stat indicates pages, processes, or other flaws in a site that frustrate, annoy, confound, or disappoint visitors—enough so that they leave. A site page that does not indicate the end of a desired process (for example, the “Thank You” page that following a completed transaction) is likely not a good place to lose visitors. This metric is highly effective at revealing fundamental problem pages on a site.
After collecting the preceding data over several weeks, patterns should begin to emerge that should inform both the search optimization and site strategy. Here’s a hypothetical example. Say a site was optimized for the phrases “video production services Cleveland” and “video post-production Cleveland.” The site ranks well for both keyword phrases, and both generate a healthy amount of traffic. Yet while the first page generates leads (phone calls or filling out a form leading to a “Thank You” page on the site), the second phrase generates traffic, but no conversions.
You’ve got a problem. It could be a site issue, or perhaps a subtle optimization tweak is in order. Seasoned SEOs will tell you that even an apparently meaningless change, such as “video post-production services Cleveland” or “digital post-production Cleveland” or “video production Ohio,” can do the trick.
The number of tweaks, adjustments, and calibrations you can make to any search optimization initiative approach the infinite. But without consistently tracking results with a web metrics program, it’s simply impossible to know what works. And what doesn’t.