An SEO Theory metrics retrospective

Since this blog started collecting analytics data in May 2007, the average referring site has sent SEO Theory about 45.8 visitors. Exactly 105 referring sites (including both SEO Theory locations) have provided more than 45.8 referrals. 261 referring sites have provided at least 10 referrals. More than 1,000 sites have sent visitors to SEO Theory (or perhaps fewer sites than that, as the analytics data tends to distinguish between sub-sites more than it should).

StumbleUpon has been the leading source of referral visitors, outperforming even Google (most of whose referrals come from the /reader/ service). Of course, search referrals have provided us with more than double the amount of traffic received from StumbleUpon.

The long-running experiment this blog represents tests whether natural search engine optimization can outperform natural Web site referral. The data is only relevant to this blog, but you could certainly look at your own analytics and make similar comparisons.

For example, the average search referral visitor spends an average of 2 minutes and 13 seconds on SEO Theory, whereas the average Web referral visitor spends an average of 1 minute and 29 seconds here. Nonetheless, Web referrals outpace search referrals by a factor of 2-to-1.

Direct Traffic (people using bookmarks or simply typing in the URL) provided about 82% as much traffic as search referrals.

The ratios change if I just look at the last six months’ worth of data. For example, Direct traffic drops to about 58.6% of Search Referral Traffic and search referrals now equal about 69.5% of Web Referral Traffic.

Although there has been growth in all three areas, Direct Traffic and Web Referral traffic have been inconsistent. Only Search Referral Traffic shows fairly steady growth over the months. SEO Theory now receives about 11 times as many search referrals as in May 2007 (the first month we installed Google Analytics on the site). By contrast, the site only receives slightly more than 4 times as much Direct Traffic as in May 2007 and only slightly more than 7 times as much Web Referral Traffic.

Over the past six months, StumbleUpon has dropped to second place behind Google Web Referrals (most of which come from the reader service). Sphinn, Bloglines, and Netvibes all follow behind those two sites, none of them even coming close to matching StumbleUpon.

I have done two things over the past year to help promote this blog: I have written at least 5 posts per week and I have left comments where I felt it was appropriate on other blogs. Nearly all if not indeed all of those comments included nofollowed links back to SEO Theory. I would say that a large number of the sites that sent small amounts of traffic to SEO Theory were blogs where I left comments.

There are a few blogs that have sent a respectable number of visitors to SEO Theory through their blogrolls. At least, the only links to SEO Theory I can find on them seem to be in their blogrolls. Every referring blogroll I have discovered contained fewer than 20 blogs — most contained fewer than 10 blogs in the list.

And there are a LOT of blog posts out there that refer to SEO Theory. According to Technorati, more than 1200 blog posts have linked to SEO Theory over the past six months.

The linking has been good and I haven’t had to resort to any sort of linkbaiting ideas to garner links. In fact, none of the most popular posts were intended to be link bait. The most popular post of all time remains the “20 Hard Core SEO Tips” article I wrote last year, mostly out of desperation because I couldn’t think of anything else to say at the time. It was supposed to be a fluff post, filler designed to keep up the pace.

Which is not to say I don’t stand by the tips. Of course I do. I wouldn’t have suggested them if I didn’t think they would be good training techniques. I just didn’t expect to see so much traffic for them. I’ve occasionally thought about writing “20 More Hard Core SEO Tips”, but what if that post failed to attract as much attention? I’d rather give in to paralysis-by-analysis for at least another few months. Maybe I’ll try to do a follow up post in October and make it an annual thing.

There have been a few controversial posts over the past few months, one of which garnered almost as much traffic as the “20 Hard Core SEO Tips” article has in the same time frame. I think that goes to show that the more helpful and less controversial I am the more traffic I can attract. Someone once explained it to me as, “You’ll draw more flies with honey than with vinegar”.

In my typical obsequious fashion I pointed out that I wasn’t really interested in drawing flies, but I understood his point.

The continuous production of content makes a difference. You don’t have to provide a consistent quality of content as it all tends to even out over time. It’s hard to find blogs that complain about the quality of SEO Theory articles, when you look at how many blogs either compliment my work or just cite it. Sure, there are occasional expressions of dissatisfaction but no one has (so far as I can tell) written anything along the lines, “Good lord, Martinez is waving his wand about content again!”

As best I can determime, the largest number of negative comments from other bloggers (not people commenting on other blogs) have referred to three topics: the Google Supplemental Results Index (some bloggers still believe that Google actually did away with it); Google’s anti-paid links campaign (some bloggers feel I strike too hard and too often at Google); and use of rel=’nofollow’ for so-called PageRank sculpting (most bloggers seem to think it’s at least worth trying even if it doesn’t seem to work for most people who have reported actually trying it).

With or without the links, the largest source of growth in traffic has been Search Referrals, and that growth represents the core concept of this blog: if you write optimized content without building links, you’ll get relevant search referral traffic.

Of course, you have to get indexed to get the search referral traffic, but since this is a blog all it really has to do is ping the blog search indices and it gets crawled. Okay, all your social media bookmarks out there are helping, too. I’ve found dozens of pages linking to SEO Theory that many of you have created by submitting SEO Theory posts to social media aggregators. Your link profile pages have turned out to be very interesting sources of information.

Do those pages help with crawling? I don’t know. This blog was not set up to test that hypothesis.

It could be argued that SEO Theory articles can grab top rankings for long tail keywords simply because it has what some SEOs call “authority” — lots of PageRank (or, a fair amount of PageRank). However, that argument won’t hold water. Some of the SEO Theory posts were designed to trump very link-rich sites in the SERPs by simply out-optimizing those link-dependent sites.

Some SEO Theory posts have also been written to help build new query spaces. To show that your query space development works, you have to populate the query space with your content and analyze search traffic for that content. SEO Theory’s query space development is limited because this blog focuses on a very narrow set of topics.

More than half the articles you find on SEO Theory were written in response to search referral data. I’ve answered questions, redefined concepts, and added topic areas based on the search referrals. In some of those articles I pointed out what I was doing.

For an SEO blog, SEO Theory doesn’t seem to be doing badly. It was never my intention to go after SEO competitive expressions with this blog since most search traffic is actually in less competitive queries. The obscure queries which many of you found this blog with were not always targeted. Some really bizarre search expressions have been used repeated to find SEO Theory (and, I have no doubt, other sources of information or opinion).

One recent experiment I performed was to write two posts per day. It took a lot out of me (especially since I’ve been struggling with severe bronchitis for almost a month) but I found that daily traffic almost doubled when I doubled production. Although the people who subscribe to SEO Theory RSS feeds were undoubtedly coming back to read the second articles, the search referral data shows that the articles were drawing in both targeted and random traffic.

If you’re hoping to build up a popular blog, you need to understand that your production rate and consistency in performance are two very vital factors for success. That’s what being the resource is really all about. That’s why multi-author blogs do so well over time. When people quickly see that they can depend on a site to be a source of information, they’ll keep coming back to it.

One article I’ve occasionally thought about writing is “SEO blogs you may not have heard of”. There are a lot of nicely written blogs out there that occasionally catch my eye. But whenever I’ve started to write that article (and tonight was one such occasion), I always find that the blogs I like most are updated least often. It’s good that people out there write some really thoughtful posts. I just wish they wrote more often.

Of course, if they did publish more often I probably wouldn’t want to include them on a little-known blogs list — they’d be well-known.

One mediocre blogger who writes something every day will achieve more than ten great bloggers who write only once a month. In the end, the quality of the blog will be measured in terms other than how link-popular any particular post may be. Consistency and dependability are key factors in blog-based search engine optimization.

But don’t ignore your analytics. Search referral data tells you more about what people want to read than any other SEO research data available. Even a mediocre analytics package will pay for itself if you just scour the referral data at least once a week.