For as long as the industry has existed, SEOs have been hung up on search engine rankings, with keyword tracking and position monitoring being regular website health checks. Most SEOs use automated software to check the rankings of websites for key search terms and use fluctuations in positions as a key indicator of performance.
However, is this really the best indication of performance in organic search? What does it really mean if you’re ranking first for “fast food restaurant” and second for “really tasty burgers” if your audience is just searching for “McDonalds”?
In addition, a number of changes in the industry, including the introduction of (not provided) in Google Analytics and the increased awareness of Google’s policy on rank tracking, will make it increasingly difficult to measure the value of positions in the SERPs.
Instead of monitoring your position in the SERPs for keyword terms, why not measure the activity of actual visitors who clicked through to your site, whether your page was listed in first position or on page 4?
Landing page analysis allows us to study the behaviour of visitors who actually visited your site from organic search. We can check the specific keyword they used (unless it is “not provided”), see how relevant the landing page is to the search and review bounce rates to see where you might be losing visitors. The following guide gives an overview of how to carry out the kind of analysis which I believe provides more value than position monitoring and keyword tracking.
To start, open up Google Analytics and navigate to the Landing Pages section (Content/Site Content/Landing Pages). You will be presented with a complete list of your website’s landing pages; pages which visitors have arrived at. In most cases, your homepage will be your top landing page.
Before you carry out any analysis, you will want to filter the data to show only landing pages from organic search. Your referrals from others sites or visits from any offline promotions which send visitors to a particular page will be listed as landing pages and the behaviour of those visitors will be included with that of organic search visitors. You can easily filter these out by choosing the Default Segment “Non-paid Search Traffic” from the Advanced Segments menu.
Next, sort by visits and you will see your most landed-upon pages from organic search traffic only. Already you can see information on organic search behaviour that you do not get from rank tracking alone: the actual pages returned in search. Sure, your site may be ranking first for “blue sandals” but what page is in the rankings? If you only sell blue sandals, you would probably be happy to have your homepage ranking for the search, but if you sell shoes, clothes and umbrellas with a specific page for blue sandals, you would likely want that page to rank for the keyword instead of your homepage.
To find out what keyword people used to reach your landing pages, you can now set “Keyword” as a secondary dimension and get a world more information.
If you are mostly interested in extracting keyword data from your landing page analysis, you may also want to exclude (not provided) keywords using Advanced Search:
Now you can see your top landing pages with the keyword the visitor used. Bear in mind that landing pages will appear more than once as the same page will likely be returned in search for multiple keyword terms.
Bounce rate is a particularly useful metric in organic landing page analysis as it gives you an indication of which pages are serving the needs and expectations of visitors well and which pages are causing more visitors to leave your site without navigating further.
This example shows that pages 4, 9 and 10 have lower bounce rates while pages 2 and 6 in the list have unnaturally high bounce rates. This should trigger some manual analysis of the pages and the keywords: is the keyword search what you would expect for that landing page? Is the page clear and easy to navigate? It might be you need to better optimise your pages or give users clear access to useful navigation or actionable options (Contact Us, Buy Now, etc.)
If you know your way around an Excel spreadsheet, the best way to digest and analyse the information is to export the information to a CSV and manage it in Excel. You can choose to show up to 5,000 rows in the Analytics window and export those rows easy enough.
If you need significantly more rows in the export (ie; Analytics is showing 1-10 of 100,000 rows) you can force Analytics to show as many rows as you need with a quick manual adjustment in the URL bar. When you have chosen to show a certain number of results in the Analytics window, you can see a “rowCount%3D” value at the very end of the page URL:
The example above shows the URL when you have chosen to show 25 results per page. Simply change the number following this code to the number of rows you need and reload the page.
Please note: forcing too many results to load on one Analytics page for export can take a lot of time or cause your browser to crash. In addition, Microsoft Excel 2007-10 supports a maximum of just over one million rows.
When you have the data in Excel, you can more easily analyse and annotate the data any way you wish. Consider sorting by high bounce rate or low pages per visit. You could also filter out or only show brand terms.
If you still want to track your positioning in the SERPs, you can easily do this alongside Landing Page Analysis and a number of other analytical metrics within Google Analytics. However, rather than using automated software to report on rankings or even search in Google to check your ranking position, you can actually monitor ranking positions within Google Analytics at the time of users actually clicking on your site in the SERPs.
There are a number of benefits to this, not least because Google does not support or approve of position monitoring. In addition, it means you can see the position of the actual keyword searched for at the time it was clicked. This shows you to compare the value of different keywords at different positions. For example, if “short black dress” was clicked on 20 times at position 4 but “long red skirt” was only clicked on 10 times at position 2, you know that the first keyword may be performing better and you can then try to work out the reason for this. It could just be that there is a greater demand and higher search volume for short black dresses, but it could be that the page’s Meta description is better constructed and better advertises your brand.
For details on how to set up rank tracking in Google Analytics, see our blog post from 2010. Despite being two years old, this post is still accurate and describes in detail how to monitor search engine rankings within Google Analytics.
Once you have a better understanding of your organic landing pages and visitor behaviour, you are better suited to consider landing page optimisation and refining under-performing pages to better meet the needs of your users.
My colleague Graeme Benge recently posted a video guide to Landing Page Optimisation on Koozai TV which discusses some Landing Page Optimisation techniques you should be aware of when reviewing your website’s landing pages.
If you have an opinion on these techniques or perhaps have other alternatives to rank and keyword tracking, please feel free to share them in the comments below!
Landing Pages Image from Bigstock