Ahead of SES London we’ve been lucky enough to interview a few of the speakers at the event. When we saw the agenda though there was someone who sprung out as someone we really wanted to interview, Brian Clifton.
We asked Brian about how he measure’s SEO ahead of his session Meaningful SEO Metrics: Going Beyond the Numbers
Q. What’s the biggest mistake people make when measuring their SEO?
A. Focusing only on SEO! Seriously, you would be surprised at how many people still treat SEO as a silo. As for all an organisation’s marketing efforts, SEO needs to be compared and understood along side other channels. For example, there is a great deal an SEO expert can learn from the results of PPC, email, social media and even offline marketing campaigns. Although the reasons for arriving at your website may be different if you came from say, organic search, the visitor is still a potential customer seeking you out. Hence, the performance of a visitor from a PPC keyword click-through, for example, is valuable research data when planning an SEO strategy. Answers to questions such as, did the PPC visitor purchase on first visit? Conduct an on-site product search? If so, using which keywords are all important pieces of information for the SEO expert.
I also find people make the mistake of not looking beyond the referral data. In other words, what visitors do once on the site. They can be too focused on achieving the top positions at Google. For example, you might be number 1 for fifty keywords and it all looks good, However, these may not be the keywords that convert visitors to engaged visitors, or engaged visitors into customers.
Q. How do you measure link building?
A. Unless a visitor actually arrives on your website from a referral link, you cannot measure who is linking to you on the web using an “on-site” web analytics tool such as Google Analytics, Yahoo Analytics, Omniture etc. That is the realm of off-site tools. However, if visitors do follow those links you have from other websites, then you can track them. The trick is to group or categorise your referrers so you can compare performance.
For example, if a travel site is receiving referral visits from tripadvisor.com, www.guardian.co.uk/travel, www.virtualtourist.com etc, these should be grouped as “Review referrals” or similar, and compared with “Social network referrals”, “Partner referrals” and so forth. That way you avoid missing key data spikes that can be buried under the plethora of referrals that most sites receive (typically thousands per month). In Google Analytics, you achieve this either by using Advanced Segments or re-write Filters.
Q. What’s your favourite over looked SEO metric?
A. Farming for new relevant keywords from visitors that use on-site search i.e. internal search. OK, this is not a metric per se – more of a process, but it is a great opportunity that is often over-looked. On-site search is an excellent feedback mechanism for your search marketing department. That is, visitors telling you exactly what they want to see on your website.
In addition to discovering new keywords, perhaps there are variations in usage or spelling you can take account of in your page content. Perhaps visitors are using relevant keywords after they are on your site that you are not aware of. For example, visitors looking for books may also use keywords such as “how-to guides,” “manuals,” “whitepapers,” and “tech sheets” on your internal site search. This is a perfect opportunity to build and optimize your website content for those additional, related terms.
Q. You’re a big fan of Google Analytics is there ever a reason to chose a paid option?
A. Yes, its true I have drunk a lot of Google cool-aid in the past and I am a big fan. But I have worked both sides of the Google fence and their products have strengths and weaknesses just like any other. In addition, its important to bear in mind that no one tool can achieve everything. Some reasons for looking at an alternative web metrics solution include:
- ISP/Hosting providers
These require large scale automation – particularly for account setup. Google Analytics is just not setup that way. For this, the better solution is to run licensed software on your web servers.
Google Analytics requires an open Internet connection to communicate with Google data collection servers. If that is not possible, behind a firewall/intranet for example, then licensed software is the alternative.
If you are a CMS, CRM vendor, or you are a large portal, you may wish to embed a custom analytics tool into your products. For example, including only certain metrics or creating your own product specific ones. There are now customisation abilities with the GA Export API, though that still queries pre-processed data. Sometimes you want the raw data.
Despite those caveats for using Google Analytics, Google does have a product that can also meet those needs – you just can’t get away from them – and its called Urchin Software (www.google.com/urchin). My thoughts of Urchin v GA are discussed here: http://www.advanced-web-metrics.com/blog/2010/01/29/what-is-urchin-6/