A Simple Way to Re-Visualize Your Key Sources of Traffic (Part 1 of 2)
We all know that things change very quickly with respect to the Web. For instance, the way that people find your site has changed a lot. Think of the days when Yahoo was the site of choice for people to do research and find a site that would help them accomplish their goal; whether it was to plan a family vacation or to get forms for income tax filings. In recent months and years, it seems that when we do informal surveys in rooms of 20 or 30 people, virtually everyone raises their hand when asked “who has done a Google search today?”.
What is a Referrer?
Knowing that Google has a huge percentage of the North American and global market share, it shouldn’t be a surprise that it is typically the number one referrer to a Web site. Before talking too much about referrers and some related benchmarks, it’s important to highlight a quick definition. What is a referrer? In its simplest form, a referrer is a Web site which sends traffic to another Web site and can be a search engine, a social media site, a privately run site, a partner site, etc. When a person sits at their computer and enters a search term in to a search engine, a list of results is returned (called a ‘Search Engine Results Page’ or SERP for short). Upon clicking on one of the results, the keyword they entered in the search engine is captured in the server logs as a ‘keyword’ and the site which sent the visitor is the ‘referrer’. For example, when searching in Google for ‘government of canada’, the top listing on the first page is a link to ‘Government of Canada Official Web Site | Canada Site’ which, when clicked, takes the visitor to ‘www.canada.gc.ca/home.html’. In the log data for this Government of Canada Web site, the ‘referrer’ would be tracked as ‘Google’ and the keyword would also be captured as ‘government of canada’.
Who Are Your Top Referrers, and Who Are They Sending To You?
Getting reports on ‘top referrers’ to a site is relatively straight forward and is a standard out of the box report from Web analytics tools such as WebTrends, NetInsight, Omniture, Google Analytics, etc. It is common for clients to ask for the top 10 or top 20 referrers, and while we can provide that, we prefer to provide them information which is far more valuable; referrer groupings with benchmark data. Rather than looking at the top referrers, usually out of curiosity more so than for practical decision making, we group referrer traffic in to the following major categories (listed with their benchmark values):
- Google – without question, Google deserves a category of its own. Across the sites we studied (which will be summarized at the bottom of this blog), we see that Google makes up 37.7% of the traffic. On the low end, some clients receive about 23.1% from Google while on the high end they can receive as much as 57.8%. Sites on the high end of this benchmark are traditionally extremely well search engine optimized.
- None – at 29%, this reflects traffic which comes direct to the site (i.e. the visitor knows the URL), they click on a link in an email signature or email newsletter, their session on the site times out and they visit yet another page (i.e. they do not view a page on the site for at least 30 minutes while doing other things on their computer), they frequent a site and therefore have it bookmarked in their browser, etc. Depending on the type and frequency of any branding exercises and outreach campaigns, this can vary from a low of 19.7% to a high of 37.4%.
- Other Search Engines – when combining all other search engines and directory services such as Yahoo!, Bing, MSN, Ask, and so on, cumulatively they represent just a fraction of Google. Overall, one can hope to get about 3.8% from all other search engines, but shouldn’t be too concerned if they are near the low end of 2.5% and should be thrilled if they achieve above the high end around 5.2%.
- Social Media – this is a hot topic recently with many clients and an important type of referrer traffic to monitor. Whether your organization is actively pursuing social media campaigns or not, others may be blogging, tweeting, or sharing information about your organization online. At 0.4%, this can be easily impacted by external activities or issues, such as the H1N1 virus or political uncertainty leading in to an election. At equilibrium, this can be as low as 0.1% (usually for sites with no social media strategies) and at times of heavily debated issues, as high as 1.1% (typically for sites with an active social media strategy).
- Other Web sites – all other sites that are not included in the 4 categories above can be easily combined in to a single grouping. Ranging from a low of 16.8% to a high of 38.1%, this section represents a relatively significant piece of the big picture and averages about 29.2%. It’s important to note that this could be literally tens of thousands of sites, however, by process of elimination from the aforementioned categories, it involves simple math to calculate this percentage.
- A sixth and final grouping of referrers could be considered on a site by site basis to reflect partners, likeminded organizations, etc. For example, a Government of Canada site may be interested in creating a group to reflect all traffic which is referred from other Government sites, including municipal, provincial and federal. If this sixth grouping were created for a site, it may steal a small fraction (likely below 5%) from the ‘Other Web sites’ grouping.
As we can see from this data, Google is a predominant source of traffic for the Web sites analyzed. Given the size of the data set, we believe this trend would be similar for most other sites similar in nature to those in this study (i.e. Government Web sites). Given this trend, we strongly support that clients search engine optimize their content to meet the needs of Google, first and foremost, while also considering other avenues for traffic generation (i.e. links on partner Web sites, Pay-per-click campaigns, etc.).
Now that we’ve touched on some major benchmarks with respect to referrer traffic, part 2 of this blog will outline some of the major reasons why it is important to understand this aspect of your site.
The data within this blog was derived by analyzing several Web sites from within our client base. Cumulatively, the data set consists of over 200 Million visits and 1.2 Billion page views and ranges from January 2007 through December 2009. The data was considered on an annual basis and future efforts will be invested to creating similar benchmarks on a monthly basis. Clearly, when analyzing this data on a more granular level, such as weekly or monthly, these numbers would potential have much greater ranges and we will be pursuing this research to include in the near future.
For a more elaborate definition of a ‘referrer’: http://en.wikipedia.org/wiki/Referrer.
We update our data regularly and will be producing a more granular report exclusively for our clients who choose to be involved in this benchmarking program. Participation ensures confidentiality of the data specific to each client, however, provides mutual benefit across all those who participate. If you would like to be part of our benchmarking program and have your site’s data included, please contact Tyler Gibbs at 613-232-8500 x 102 for more information.