I have been in the SEO business for more than six years, and I always chuckle when a client or colleague says, "Oh, so you're an old dog in the SEO biz, huh?" While many recognized names in SEO have been at it for years longer than me, I've been around long enough to see the landscape of my profession revolutionized again and again. But one client question has remained constant since the beginning, and that is, "Can you get me to rank No. 1 on Google?"
Usually, this question is brought about by another -- sometimes unspoken -- question, which is "How much traffic can I expect from a No. 1 Google ranking?" It's a fair question, but historically it has been tough to answer, not just because there are a ton of variables, like keyword variations and seasonal business factors, but also because up until recently, we could only guess. But until moving forward with any client, I try to get to the root of these questions because the truth is that ranking No. 1 on Google for many types of keywords is not all it's cracked up to be.
You see, traffic by keyword position is not something that the search engines really wanted us to know about. It's tough to say why they kept it a secret for all of these years. The cynic in me says that it's because they wanted to sell more paid search (PPC) inventory because the metrics for PPC have razor-sharp precision compared to SEO. My inner optimist tells me that it's because the search engines were more interested in optimizing their algorithms for relevant search results than being marketing tools.
But of course, we all know that organic search, and search engines themselves, are nothing if not marketing tools. Whatever the reason for the secrecy, the fact remained that we SEOs had to at least attempt to answer the rank/traffic question for our clients. I think that we, as an industry, did a pretty good job of estimating what could be expected assuming that our client worked closely with us and gave us access to their analytics.
One cool hack was the Wikipedia traffic stats hack. Let's say your client is trying to rank for the keyword term "Dwayne Johnson." A Google search reveals that The Rock's Wikipedia page ranks second for that search term. The Wikipedia traffic stats application shows that the English version of this page was viewed approximately 227,000 times in June of 2010.
This is hardly perfect data because we only have the total number of page views, not the number that came from Google. But it's a starting point. In March 2008, Search Engine Land wrote a great piece on how to use Wikipedia to reveal traffic stats.
Also, in late 2006, AOL accidentally leaked click data percentages based on ranking position. As you might imagine, this was one of the biggest pieces of news in the history of the SEO game. We finally had solid evidence to back up many of the claims we were making. No longer were we depending on our clients to believe only our professional expertise and hypotheses. We now had numbers, directly from one of the search engines.
Understandably, the SEO blogosphere exploded with countless extrapolations of this data, like this one from SEO Black Hat. And using publicly available market share data (what percentage of total searches belong to each search engine), we could estimate traffic, by ranking position, across the major search engines.
With all due respect to AOL (how else was my grandmother supposed to have an email address?), it's not exactly a big gun in the search engine lineup. SEOs wanted Google data, and we wanted it bad. About six months ago, our wish was granted. Google, inside of Webmaster Tools, added data on rank, impressions, clicks, and click-through rate for common keyword queries. To get to this data, log into your Webmaster Tools account, expand the "Your site on the web" menu in the side bar, and click on "Search queries."
It would be wrong to call this the Holy Grail of data because, after all, SEO is largely about making recommendations based on estimated results. And this data isn't predictive, it's reactive. But it tells us a lot about how well a client's site is converting, and it can be extremely valuable for any SEO or agency that has developed niche expertise.
For example, let's say you are an SEO with a lot of real estate clients. For searches like "homes for sale in Los Angeles," you can tell a new client, with a high amount of confidence, that you know how much traffic is available for that particular search term because you have seen it before. Over time, you can make highly educated guesses about how much traffic can be expected from different search positions, assuming that you are collecting and trending all of your data.
Image courtesy of SEO Book
But the real value of this new Google feature isn't predictive anyway. There are three basic types of organic searches: navigational, transactional, and informational. And all of them behave differently. Earlier this year, Aaron Wall, arguably the best SEO in the business, wrote an excellent article on his site, SEO Book, about how to interpret the data in Google's new feature. In the above example, Wall shows that ranking No. 1 on Google for the keyword "SEO" drives only about half of the traffic that he receives from his top two branded search terms, "seo book" and "seobook." He also points out that branded searches tend to bring traffic that is more likely to spend money on your products and services.
The conclusion here is that brands need to be careful about spending time and money on vanity keywords because ranking No. 1 isn't the end all of SEO. Perhaps the most valuable point Wall makes in the entire article is, "Rather than sweating trying to rank well for the hardest keywords first focus on more niche keywords that are easy to rank for."
Google has provided an easy-to-read interpretation of what different searches actually mean. In other words, ranking No. 1 for one keyword is not at all the same as ranking No. 1 for a different keyword. Therefore, SEO strategy has a lot more to do with the keywords themselves than their ranking positions. But the final, and perhaps most important question related to this topic is, "If I can't rank in the top 10, is it worth trying to rank at all?" The short answer is "no."
On Google and other search engines, most searchers don't navigate past the first page, and the percentage of traffic drops sharply after the first result and almost disappears after the tenth result. Add to this the fact that Google's universal search (blending video, image, maps, and local alongside standard organic results) is pushing search results, even from the top 10, off the bottom of the first page.
Of course, there are exceptions to this, including robust long-tail search strategies, but that's a different topic altogether. But the fact that top 10 rankings are so important should be informing new variations on traditional SEO strategies. For example, strategies that pursue local search, video SEO, and image results can put your pages directly at the top of the page for targeted search queries. And in many cases this is more valuable than traditional organic results.
In addition to being a nice little history lesson about measuring how traffic is affected by rank, there are some important takeaways, including:
- Ranking in the top 10, especially for branded terms, is important.
- Brands should carefully evaluate the ranking impact of their top organic keyword terms; are people actually clicking the result?
- Remember that searcher behavior is heavily influenced by the type of query: navigational, transactional, and informational.
- Once an SEO strategy has been started, brands should closely monitor the results with Google's excellent Webmaster Tools.
- If you are interested in SEO and you aren't already reading Aaron Wall's blog, you should be.
Drew Hubbard is a search and social media consultant.
On Twitter? Follow iMedia Connection at @iMediaTweet.