One of the things I frequently get in trouble for is giving away too much information. I think that giving away too much insider information is what I am about to do below. My argument for sharing is that knowledge is power and by giving companies the skills and ability to understand what I do and how I do it, not only will they will better understand, appreciate and purchase services.
One of the things I get asked most often, next to “how do you choose an agency” is “how do you know if what they are measuring means anything”? Understanding what is being measured, how it is measured and how you can double check the measurements yourself if you choose to is an important part of understanding what your agency, or in-house SEO guru, is doing and how it is positively impacting your bottom line.
The below SEO ROI measurements can be used with different types of businesses. They are not restricted to one industry nor one type of measurement package. Where possible I’ve chosen measures which can be done easily and/or for free in order to keep costs low and understanding high.
In order for any measurement to have meaning, there needs to be a benchmark. Take the measurement you are going to track, record it and date it. If you then take the measure again, you have a point in time to compare it against. Always record your measurements at least monthly. For some sites weekly will be too frequently but less frequently than monthly may mean you miss something going wrong.
Visits from Organic Search Results – Measuring the number of visits in to your site purely from organic search results. To measure this, an on-site analytics package will give the most accurate representation of this number. Google Analytics is free and Omniture is my favourite but there are others available.
Number of Visitors from natural search Per Keyword – A breakdown of the number of visits from organic sources driven to the site via different keywords. The key here is that we’re breakingthis down by keyword and not just overall traffic. No other traffic such as direct or referred (from sites other than search engines or paid search) should be included. To measure, analytics must be installed and tracking incoming referrals. Most, if not all, analytics packages will break down incoming visits from search engines to give a per keyword value for visitors.
Conversions on Organic Traffic – This is a measure of the number of conversions made after a user clicked through from a natural search listing. More complex to measure as this does require the interaction between analytics and cookies. To measure, track user conversions where the last click is from natural search and leads to a sale or whatever you deem a conversion to be. To expand, measure where natural search is responsible for sending 2 or more visits by the same user to the site but where natural search is not necessarily the last click. Conversion rates = (conversions /visits)*100
Rankings for Keywords A ranking is where a particular site ranks for a particular keyword within a neutral set of search results. Neutral search results are where personalization and search history is not a factor in rankings – ie you must log out of Google and delete all cookies, close, reopen. Keywords are single or multiple words which are being targeted on page on the website being measured. To measure this, a neutral tool like Advanced Web Ranking which is able to gather non-personalised results. Failing the use of a tool, in Internet Explorer delete all cookies and history, close and reopen the browser and search for the term. This must be done prior to each search and if possible, a country-specific anonymiser should be used. Record that date this measurement was taken and link to the measured data.
These measurements, once you know what you are doing and have an action plan of execution, will help you see whether the changes are having an effect. Don’t worry if some changes take longer to show a difference – it can take up to 3 months for changes to be picked up by search engines.