Google has one main aim: providing users the best possible search experience. The search engine is constantly optimizing its algorithm to improve its understanding of user signals from websites. The main focus here is on signals that indicate how satisfied a user is with the content they find. If your content creates a positive user experience, then Google will reward you with high rankings. In this article, Thomas Gruhle, Founder and CEO of the Traffic and Conversion Agency LEAP/, answers your questions on SEO, UX and the role that user signals play in determining search engine rankings.
If you’re looking for support in developing your SEO strategy, then our Digital Strategies Group consultants are here to help:
User Signals FAQ
The following questions were asked by participants at a recent (German) webinar we held with Thomas Gruhle, Founder and CEO of LEAP/. The webinar dealt with the topic of SEO and User Experience, looking at the impact of user signals on search engine rankings. Thomas has been kind enough to answer the most interesting questions asked. Of course, if you speak German, you can also watch the full recording of the webinar on-demand. Otherwise feel free to ask any questions not covered here by using the comments below.
How do you define non-brand SEO in Google Analytics?
The primary source for the definition of non-brand keywords is the Google Search Console, as this contains all relevant keywords. Google Analytics isn’t a source for keywords.
We start by filtering the pages that almost exclusively generate brand traffic and add these to the segment “brand SEO”. All the other pages, that is those that don’t get traffic from keywords containing the domain name, are defined as “non-brand SEO”.
Which synergies are there between SEO and SEA?
An SEO analysis of keywords and trends can reveal seasonality insights that can be useful for your SEA bidding. You can also consider filtering out informational SEO keywords and booking them as SEA keywords later.
Another synergy is that SEO experts are often more experienced in improving conversion rates (e.g. for a product). And any improvements to the conversion rate can also help your SEA traffic to convert better. On the other hand, Page Titles can be more effectively tested in SEA, but lessons learned from tests can also be applied to search engine optimization.
Does Google use Analytics data to help compile rankings and do they analyze Matomo?
Google’s official position is that they don’t use analytics data when determining rankings. User signals are gathered using data from Google Chrome. Matomo isn’t analyzed by Google.
Why does Google Analytics exist if Google doesn’t use the data?
Google Analytics data isn’t used for rankings – these are based on data from other sources. However, Analytics was originally used to assess Google AdWords performance and has, over time, extended its reach into other spheres.
Is there any point doing A/B testing if the data is insufficient to provide statistically significant results?
For low-traffic pages, we use the Bayesian statistics provided by our A/B testing too. This uses another method to determine the statistical significance of the results.
Which tools do you use for A/B testing and which would you recommend?
We have worked well with Kameleoon and A/B Tasty. However, the data basis you are using is much more important than the choice of tool
How many keywords should you optimize each page for?
There is no fixed rule that applies to all cases, but it is important that you have a clear focus and that you define one topic per page. And topics should be clearly separate from each other. If cannibalization does occur – meaning that two pages rank for the same keyword – then these pages should be combined.
We observe that most pages have two or three keywords that are responsible for around 80% of the traffic. These are often the main keyword, the main keyword plus an adjective or questions related to this keyword. They could, however, also include synonyms, in which case these should always be targeted too.
How is scroll depth measured?
To measure scroll depth in Google Analytics, you just have to set up an event. This is most easily done with Google Tag Manager.
Here’s how you do it:
- Create a new trigger: Choose the Trigger type: “Scroll Depth” and give the trigger a recognizable name e.g. ScrollDepth
- Configure the trigger:
- Choose the scrolling direction, which will normally be vertical, and state at which percentages the trigger should be fired (e.g. 25, 50, 75, 100). This means that if the user scrolls halfway down the page, the trigger will be fired twice: once at 25% and once at 50%.
- You can also indicate whether you want the event to apply on all pages or only on some pages.
- Activate the scroll variables: Under “Built-In Variables” in the menu, click on the button “Configure” and select all the variables that are listed under “Scrolling”.
- Create the tag:
In the menu under “Tags”, click on the button “New”.
- Tag type: Universal Analytics
- Track type: Event
- Category: Scroll Depth
- Action: {{PagePath}}
- Label: {{Scroll Depth Threshold}}%
(Note: Action and Label are both variables that can be selected using the “building block” button.)
- Set the “Non-interaction” field to “True”, as otherwise the Bounce Rate will be affected.
- Google Analytics Settings: In the drop-down menu, either select the existing Google Analytics setting variable or use “New variable” to create a new one (add the Tracking ID from Google Analytics and save).
- In the trigger area, select your “ScrollDepth” trigger that you created in step 1.
- Use the preview to look at all your changes and if everything looks right, send the tag.
- You’ll find the event in Google Analytics under: Behavior > Events > Most important events.
How reliable is the data from Search Analytics for Sheets, i.e. the Google Docs interface for the Google Search Console?
We use Analytics Edge as our API. Analytics Edge is built for Excel, but uses the same API. The most important thing is that you pull data for a “Page” and not for the keywords, as this can show discrepancies. That said, you’ll find the same discrepancies directly in the Google Search Console.
You should also add a country filter. We work with these numbers to identify trends and developments. Sometimes we do see small deviations from the data in the GSC, but these are not significant.
Will the ePrivacy Directive have an impact on how Google analyzes user signals?
As the ePrivacy Directive has not yet come into force, it is impossible to say at this stage which impact it will have.
What is the traffic ratio between A and B users?
Testing variants are displayed randomly, so the number of users is usually almost identical.
How do you differentiate your analysis and tasks between Google Analytics and the Google Search Console?
Google Analytics provides data about your website. The Search Console provides Google data like clicks and impressions.
How much user data is lost from Google Analytics by using a cookie consent tool? Are there any alternatives?
We haven’t experienced any problems here. There is currently no alternative to the consent form.
Which challenges to you see when trying to get clients to move away from individual keyword research and focus on more holistic content based on topic clusters?
We explain to our clients that keyword research and briefing are the most important part of the content creation process. A thorough briefing should include additional information like a general structure, questions and a range of perspectives on the topic.
The text should be aimed first and foremost at the user and their challenges – not at Google. Clients have to understand that good content is an investment. That is the only way to be taken seriously by Google and by users, and to improve your search engine rankings.
Should landing pages that rank on the second page of the search results be optimized for better user signals or is it more important to work on the three pillars (mentioned in the webinar)?
This is a difficult question, for which there is no single answer. In most cases, it is likely that the other pillars should first be set up properly. User signals are at their most effective when aspects of the page related to its technical performance, content and links are already in place. That’s why we normally start with these three issues when conducting our website audits