Google analyses behaviour data as a ranking signal
User experience is an important ranking factor in SEO in 2015. But how do Google measure that entity? In fact, Google collects behaviour data about user to measure user experience.
The way users navigate on a website translates their interest or not for your brand or product. Different datas like click-through rate (CTR), time onsite, frequency, type of devices used, navigational paths give insights about how visitors perceive and enjoy the experience offered by your website.
1# Click-through rate
Click-through rate is an important data for user experience and quality. That rate relies on different metrics and factors.
Your position in the SERPs influences your CTR value. Actually, people will be more likely to click on top results, which refers to “position bias”. This metric can appear quite unfair as first results will get an higher CTR.
A good CTR is relative since it won’t be a surprise to have a 25% CTR for a top result while the same value for a competitive term would be pretty high if located close to appealing search features (like answer box, rich snippets, etc). In fact, CTR is deeply linked to its position, quality of snippets, relevance of the result, brand recognition and other features.
2# Duration, frequency, trajectory and URL access
The time a visitor spend on a website is a key metric. But it has to be measured correctly because for instance you could open a page, navigate on it but go do something else while the page is still open. That data wouldn’t be accurate. Engagement onsite, as a non direct ranking signal, can be detected through keyboard, mouse, tablet, pen, touch screen, etc. Actually, John Mueller stated that time on page, clicking or filling out forms for instance was not regarding as a direct ranking signals but they admit that
Another direction for future work is to incorporate active learning in order to gather a more representative sample of user preferences.
Chrome is editing a system that collect user log data. It delivers reports with user activities like fetched URLs, opened and closed tabs, maximized windows, etc.
Use in rankings
Google also collects duration data using nodes (URLs), edges (links) and labels (user behaviour data) and page engagement like session duration are used to measure weight of nodes.
Dan Petrovic from Moz explained that if
you link to pages that people spend a lot of time on, Google will add a portion of that “time credit” towards the linking page. This is why linking out to useful, engaging content is a good idea. A “client behavior score” reflects the relative frequency and type of interactions by the user.
Likehood to click
The likehood of a link being clicked depend on:
- Location of the link on the page (menu, sidebar, footer, content area, list)
- Position of the link on the page (top, bottom, above/below fold)
- URL characteristics (external/internal, hyphenation, TLD, length, redirect, host)
- Number of links on page
- Image link, size, and aspect ratio
- Font size, style, and colour
- Size of anchor text
- Commerciality of anchor text
- Topical cluster match
- Words around the link, in title, or headings
Also, Google may gives more credits and weight to a link that has been clicked more often than another on a page.
Bounce rate is an important signals for search engines. Indeed, if after clicking on a result, the user quickly go back to search results, it shows that the results didn’t answer their needs.
The way the user fill out an URL has an impact on its signal. Actually, URL data can refers to both an URL into an address field of a web browser, a user access an URL by clicking on a hyperlink to another web page or a hyperlink in an email message. The URL signal will be stronger if the visitor has directly type the URL than if he comes from a suggest from the SERPs. There are 3 types of URL selection (by order of significance):
- Writing the full URL
- Writing partially the URL with auto-fill completion
- Following an hyperlink
The moment when users log into a page and when they log out gives more insights about their journey on a website. The login page has a larger impact on rankings for Google.
A login page can start a user on a trajectory, or sequence, of associated pages and may be more significant to the user than the associated pages and, therefore, merit a higher ranking score. (Moz)
Moz runned a test about login pages to see if repeated client access and page engagement impacts the search visibility of the page in any way and to check those signals:
- URL familiarity, direct entry for maximum credit
- Triggering frequent and repeated access by our clients
- Expected session length of 30–120 seconds
- Session length credit up-flow to home page
- Interactive elements add to engagement (export, chart interaction, filters)
Mixing indirect and traditional ranking signals
Google doesn’t consider all behavior data the same. Indirect signals like visit frequency of session duration won’t have the same importance as traditional metrics like anchor text or title for instance.
To sum up
Google gives more and more credits to user behavior data in its way to measure user experience. Any actions realised on a website does matter and send informations to both search engines but also us. Our goal is to foster users to engage on a page and to offer the best journey possible to rank higher.
Originally written on: http://www.oncrawl.com/google-analyses-behaviour-data-as-a-ranking-signal/