Performing SEO is hard, and quantifying SEO efforts can be even more difficult. Google Webmaster Tools (GWT) helps us regain the keyword level granularity lost with Google’s switch to secure search. In the last article, we looked at what it takes to compare Google Analytics to Google Webmaster Tools, and found that when taking into consideration limitations and filters, it’s surprisingly accurate. Digging further (looking across almost 40 different implementations using the methodology described in the previously mentioned article), we’ve found it can also be used as a tool to ensure:
- Ensure proper organic attribution.
- All Top Organic Pages are getting tracked in the proper analytics profile.
In this article, we’ll first be walking through how to quickly find potential mis-attribution of organic search. Then we’ll wrap up by quickly grabbing a list of GWT Top Pages to make sure traffic is getting logged in the proper analytics account.
Proper Organic Attribution
Comparing the Top Pages in Google Analytics (GA) versus GWT seems to be a great way to identify traffic getting improperly bucketed as organic search. Keep an eye out for strange URLs that don’t seem to fit the canonical profile; landing pages for paid campaigns have been common.
Step 1: Export Proper Data From GWT and GA Reports
Follow this tutorial to match up GWT and GA reports for the same date range, then export the data into two tabs in the same spreadsheet. One for GA, and one for GWT.
Step 2: Consolidate GWT and GA Data
Using the exported Google Analytics report, make sure the Sessions are in descending order and pull in GWT Clicks from the tab over.
Note: By default Google Analytics reports have relative URL paths (a seen in column A), if using VLOOKUP you’ll need to make sure the values match up exactly. To do this we just inserted a new column for absolute URLs and used this formula, =”https://www.example.com”&A8 .
When looking at Column D, notice that the values of #N/A represent URLs that don’t actually receive any organic traffic. This was the case across multiple sites and was confirmed using the BrightEdge ranking database.
Note: BrightEdge is an enterprise ranking database, SEMrush might be a good free/cheap alternative to check URLs.
Confirm Organic Landing Page Analytics Tagging
This seems like it would never happen, but we were able to find multiple instances of pages receiving organic traffic without proper tagging (sometimes none at all!). This is devastating to quantifying SEO efforts. Pages need to be tagged for tracking, in the same way they need to be crawled for indexing!
Step 1: Create List
This is as simple as exporting the GWT Top Pages report.
Step 2: Crawl List of GWT Top Pages for Analytics Tag
We’ll be using Screaming Frog to crawl the list we just exported. Using list mode and specifying the file we just downloading should be all it takes to load what we need.
Next we need to identify the text snippet of the analytics tagging to use for the custom filter. Screaming Frog will mark all the pages that contain this snippet, from there, we can figure out which pages do not have proper tagging.
In the upper right-hand corner of Google Analytics, a search box exists that works great for quickly grabbing the UA tag code from profiles.
Use this number for your custom filter.
Once specified crawl the list, then export the results from the Internal and custom tabs.
Similar to the consolidation method above, use VLOOKUP to see which URLs do not contain the specified analytics tag snippet specified in the Screaming Frog custom filter.
Add a column in the Internal Tab export, use VLOOKUP to determine if the URLs exist in the Custom Filter Tab, and make sure the status code is 200 (in case you’re looking at an outdated date range in GWT).
These pages are receiving organic traffic and are not accounted for in your analytics profile!
Especially when beginning an SEO engagement, it’s helpful to ensure tracking is as accurate as possible. Having the ability to compare analytics to external sources can give us something to point to when tracking doesn’t seem right.
We’ve only just begun looking at the Google Webmaster Tool data set and hope you’ve found these initial insights helpful.