Is Dwell Time Really a Ranking Factor?

Google uses over 200 ranking factors and one such factor is “Dwell Time”. But is “Dwell Time” a ranking factor? The truth is that “Dwell Time” is a very confusing and often misunderstood criterion.

What is meant by the term “Dwell Time”? How is Dwell Time measured? Is “Dwell Time” really a ranking factor? And if so, how to optimize it? In this article you will find answers to all these questions and much more.

Also in the article, we link to some well-known experts in the SEO industry to get to the bottom of this issue. Let’s get started!

What is Dwell Time?

“Dwell Time” is the time that elapses from the moment the user clicks on a link from the search results until the user returns to the same page of search results.

Example:

Suppose I enter the phrase “white hat link building” into the search box. I click on the first result and spend a few minutes poring over the content (5 minutes 14 seconds to be exact).

Since I have an unhealthy thirst for knowledge, I realized that I wanted to learn even more. So I return to the search results page (via my browser’s back button) to find more information.

Note: This intermediate process between a user entering a site by selecting it in the search results and returning to the main search results page is also known as pogo-sticking.

I was on this page for 5 minutes 14 seconds. But why is this important for SEO?

A Brief History of Dwell Time

Dwell Time was first mentioned by Duane Forrester in a blog post on the Bing Webmaster platform in 2011. At that time, Forrester was a senior project manager for Bing.

Here is what he wrote:

[…] The length of time between when a user clicks on our link in the search results and when they leave our site can tell us a lot. A minute or two is a good indicator, as it indicates that your content was useful to the user. If he spent a couple of seconds on your site, then this can be considered a bad indicator.

Duane Forrester, Senior Project Manager, Bing

But still, why is it really important for search engines?

Here’s what Dwayne had to say:

Your goal should be to ensure that the content is relevant to the needs of the user when they land on your page. By doing this, you will help the user stay on your site. If your content does not interest them, they will leave the page. Search engines can track this by watching Dwell Time on the site.

Duane Forrester, Senior Project Manager, Bing

Fine. It makes sense. The basic idea behind this is that the longer a user stays on your site, the more likely your content is to be useful.

Now I will give some examples of Dwell Time on the site and how they can be interpreted:

• 2 second Dwell Time: The user didn’t find what they wanted/expected to see on your site, so they quickly returned to the search results page to find something better.
• 2 Minute Dwell Time: A user found your content quite useful and stuck for a couple of minutes to review it.
• 15 minute Dwell Time: The user found your content very helpful and went to great lengths to learn the information you provided.

Thus, we can assume that search engines can use the waiting time as a ranking factor. Seems like a good way to gauge the quality and significance of a given result, doesn’t it?

You’ll learn more about time on site as a ranking factor a bit later in this article, but first, let’s clear something up.

Dwell Time, Bounce Rate, and Time on Page: What’s the difference?

It is impossible to combine these 3 terms into one word, because of this, confusion may arise. In fact, I’ve seen these 3 terms used interchangeably in SEO.

These 3 terms are not interchangeable!

Now I will explain to you in simple terms what each of these terms means:

• Dwell Time is the amount of time that elapses between the moment a user clicks on a search result and before they click on the back arrow to return to the search results page.
• Bounce rate is the percentage of one-page sessions (i.e. users who go to only one page on your site before leaving). These people may have returned to the search queries or simply closed the page. It doesn’t matter what exactly happened. It also doesn’t matter if they spent 2 seconds on the site or 2 hours, it’s still a “bounce”.
• Time on page is the amount of time a user spends on your page before heading to another location. Perhaps he returned to the search queries, went to another page on your site or to a bookmarked page – anywhere.

It’s also worth noting that time on page and bounce rate can be viewed in Google Analytics:

You will not find such information on Dwell Time. If Google uses any time factor as a ranking factor, then it does not post such information (or other data).

Is Dwell Time a Google Ranking Factor?

There has not yet been an official statement from Google on whether Dwell Time is a ranking factor. But at the end of 2017, the head of Google Brain, Nick Frost, said this at a conference:

Google is now using machine learning to [determine the relationship between a search and the best page in that search]. The training models track the user’s transition to the page, the user’s stay on the page, the user’s exit from the page, and try to determine the relationship between these actions.

Nick Frost, Head of Google Brain

This supports the notion that residence time is a ranking factor. But, as Keir Shepherd pointed out on Twitter, this is not entirely true.

To clarify, no one at Google said that Dwell Time + bounce rate were ranking factors, but they are machine learning channels. Google Brain is Google’s research project to study artificial intelligence based on deep learning.

They don’t deal with ranking algorithms. As such, there is no evidence that Dwell Time is a ranking factor. Although, it’s also worth remembering that most people never make it past the first page of search results.

So if your site ranks on the second page (or even beyond), almost no one will visit your page… not even for a second.

This means that Dwell Time will only work as a (possible) ranking factor for top 10 results only.

Is your site on the last pages of results? If your site isn’t on the first page yet, don’t worry about timeout optimization. Better spend your time optimizing other (more important) factors that will help you get into the top 10.

For those already on the first page, there are three reasons why Dwell Time is important as a ranking factor. Let’s search for “paleo diet for beginners”.

Anyone looking for this is clearly new to paleo. They are looking for a guide for beginners. In first place (at the time of this writing) we have just that – NerdFitness’ Beginner’s Guide:

If you have ever read this guide, you should know how extensive it is. It covers just about everything you can find about the paleo diet on one page. The article is well written and quite presentable.

In general, the result is fully consistent with what we wanted to find. Most people will probably spend 15+ minutes reading this article (probably) before returning to the search results page.

For comparison, let’s look at this page (ranked #6 for the same query, at the time of publication).

It doesn’t take long to realize that this site is way worse than NerdFitness’ guide.

Several remarks can be made immediately:

• The content is rather scarce: about 500 words;
• The information is too simple;
• The site is full of ads.

Because of this, it can be assumed that the average Dwell Time for this page will be below 30 seconds. Conclusion: waiting time seems to be a good indicator of the relevance and quality of a given result.

It’s (possibly) a fiction that bounce rate is a ranking factor

To put it mildly, considering bounce rate as a ranking factor is highly questionable. Because “rejections” from people can come for a number of reasons, such as:

• There is no need to visit more than 1 page – everything they needed, they found on one page;
• Standby for 30+ minutes starts a new session in Google Analytics;
• Poor content – users don’t like it and they leave.

If the system were only trying to meet user goals in terms of bounce rate, it would make it difficult to distinguish between good and bad interfaces. To illustrate this, let’s go back to our previous example (“paleo”).

Both pages received a “failure” due to technical reasons, but the experience was completely different for these sites (positive on one, negative on the other). Here’s what these two visits look like in Google Analytics:

This is a 15 minute session at NerdFitness:

And here is a 30 second visit to the FitnessMagazine:

Not only do these two (in fact, very different) sessions appear to be the same, the user’s time spent on the page is also listed as 0 seconds.

Clearly this is incorrect information. Why is this happening? In order for the Google Analytics service to calculate the time spent by the user on the page, it needs two clicks: an entry click and an exit click. If there is no exit click (for example, the user went to another page on your site), GA (Google Analytics) cannot perform the calculation.

Here’s a great explanation from AnalyticsEdge.com:

For sessions in which the user viewed only one page (“bounce”), the time on page and session duration are 0. This is not because Google knows they left immediately, but because they had no indication when it is the user who left, so they cannot calculate the time on page, and due to lack of information, they consider the time spent on the site as zero.

Further, specialists from Google Analytics write:

The time spent on the page can be 10 seconds or 10 minutes; they probably don’t know how many, so they put 0. Has the user read your page? They don’t know it. Maybe yes, maybe not. All they know is that users haven’t looked at another page on your site for the next 30 minutes (that’s how long a session is by default).

Another big problem with using bounce rate as a ranking factor is that then Google would have to use Google Analytics data to rank.

But Google’s official position is that they don’t use any GA data for ranking purposes. And there is no other way to determine the page bounce rate.

So, based on Google’s admission, bounce rate is not a ranking factor. You may not want to take all of Google’s words as truth, but in this case they are most likely telling the truth.

Here are three reasons why this is so:

• Not everyone uses GA: In 2012, an estimated 10 million websites used Google Analytics. Even if that number has grown tenfold since then (which it probably isn’t), it still only represents 10% of all sites. Can Google really decipher everything that matters by using bounce rate analysis on just 1/10 of the world’s websites? Maybe, but I suspect the data will be mostly useless;
• GA is often installed incorrectly: Anyone who has ever done an SEO audit knows how important it is to make sure GA is installed correctly. This is a common problem and can often lead to increased failure rates. This would be inaccurate data for Google;
• Ad blockers often block GA code: it is estimated that 30% of users have an ad blocker, and almost all of these users have Google Analytics blacklisted by default. They also block any actions that come from the JavaScript GA library. If Google secretly used the GA code, it would be incomplete data – about ⅓.

So even if Google secretly analyzes bounce rate data from GA, chances are it won’t matter much. Therefore, wait time (theoretically) trumps bounce rate as a ranking factor – wait time data is easier to collect, especially for Google.

Here’s why: Let’s say you type “iPhone 8 review” into the Google search bar.

If/when you click a result, Google will launch a “virtual stopwatch”. And when you return to the search results page, Google will stop it. Now they know exactly how much time you spent on this site (i.e. Dwell Time).

If you’re wondering how Google will know when you return to search results, we have a couple of ideas:

1. Chrome Browser Data: According to the latest data from W3Schools, 72.4% of users now use Chrome. Chrome is Google’s own browser, so they seem to know when you hit the back button and return to search results.
2. “Next click” analysis: If you go back to search terms, it’s supposed to be only a couple of seconds before you move on to another result. Google can wait for this click and thus decipher the approximate wait time for your previous click.

It is clear that with a small amount of mined data, Google can reveal some useful data for Dwell Time analysis. Try doing this:

Open Safari on your iPhone and type “link building strategies” into the search bar. The best result for this query is “Moz” (it should be in the top 3):

Wait for the page to fully load, then click the back button. Did you notice anything? Google added a scrollable list of related queries below the result you clicked on.

Note. If that doesn’t work for you, then try clicking on any other result from the first page – it seems to work that way for most. Also, I’ve only tried this on my iPhone, so I have no idea if it will work on Android devices.

This is quite reasonable. Since you’re back to the search results, you probably didn’t find what you were looking for. So Google offers some related queries to help.

This proves that Google controls the so-called “pogo-sticking” (at least on mobile devices). And if they control that, then they probably control Dwell Time as well.

But before you get too carried away, let me say one obvious thing:

This does not prove that Dwell Time is a ranking factor.

This only proves that Google controls it to improve the user experience. However, this does not prevent us from suggesting that this data may be useful in influencing the ratings.

Why You Shouldn’t Rely on Dwell Time as a Ranking Factor

You now think that Google’s idea of using Dwell Time as a ranking factor is quite logical and plausible, doesn’t it? As the saying goes: it sounds too good to be true, and it probably is.

Here are a few possible problems with using Dwell Time as a ranking factor:

This does not work for simple queries in the form of a question.

For example, we want to find the answer to the query “when did ”Robot Wars” come out?”.

At the time the search results page loaded, we still hadn’t found the answer. So, we need to go to one of the results to find the information we need. If you go to the best result (from Wikipedia), you will find the answer in the very first sentence.

Since the answer was quickly found, you will most likely return to the search results page in a few seconds. So the timeout will be quite low for this request, less than 10 seconds on average.

But this has nothing to do with bad UX (user experience); you found the information you were looking for in a couple of seconds. Here is what Eric Enge said about this:

There are many variations where a shorter Dwell Time is an indicator of quality. For example, someone is constantly looking for short reference information such as a zip code or phone number. For information queries like these, pages are created to help users find the information they need instantly.

Eric Enge, founder of Stone Temple Consulting

Thus, in this example, a short Dwell Time is not consistent with a negative experience. Result #1 was the best and most relevant result for this query.

This does not work well for “AFA” pages (random false advertising)

“AFA” are those results that, at first glance, offer exactly what you were looking for. But upon further inspection (which increases the time spent on the site), you realize that this is not the case.

Example:

Recently I was looking for a template for Google Sheets that can import Google search results. I searched for “google results scraper google sheets” and selected the top result which was a site from SEER Interactive.

At first glance it looked perfect.

I clicked on the added Google Sheets template (which opened in a new tab so I didn’t leave the previous site), made a copy, and tried to test it out.

But after a couple of minutes, I realized that the spreadsheet had stopped working and was throwing an error.

So I went back to the previous tab (which I didn’t close) to check the comments and see if other users were having the same problem. And they were.

I then returned to the search results page to continue my search. So, while Dwell Time on the site performed well (5+ minutes), my experience was still negative. The page didn’t meet my requirements and, frankly, doesn’t deserve to be in the top 10 at all until the errors on it are corrected.

This is not suitable for the “Purchases” section

Another interesting point that Mark Trafagen noted:

Another version where Dwell Time can be a false factor in determining content quality and user satisfaction is shopping. Often, when I’m shopping and comparing products by price and other parameters, I can quickly move between several results by constantly clicking on the back arrow.

Mark Traphagen, Senior Director of Brand Evangelism

Everyone I know does the same when comparing purchases. You click on a result, check the price, go back and click on another result. You repeat these steps over and over again until you find the best price.

But the “pogo-sticking” process in this situation creates a scenario where Dwell Time is not a good way to determine quality or relevancy. This is because there is no problem with the results – you are just a savvy shopper and you are comparing prices.

So, does this disprove Dwell Time as a ranking factor? Not really. Google is smart. He can discern questions with commercial intent. How? Google only shows us the results of purchases in such queries.

Example:

Here are the results for “when is the best time to drink protein powder”:

And here are the results for “best protein powder”:

Google is automated enough to realize that although these queries are similar, only one of them has commercial intent. And if Google is smart enough to figure that out, then it’s almost certainly smart enough to ignore Dwell Time for commercial queries.

So even if Dwell Time is a ranking factor, it’s probably ignored for shopping section queries.

Should you try to improve Dwell Time? (And if so, how?)

Google looks at many user engagement metrics (e.g. Dwell Time, CTR, etc.).

Don’t be obsessed with these criteria.

Instead, focus on creating quality content and providing good UX. By doing this, you won’t have to worry about upgrading Dwell Time – it will take care of itself. I reached out to Danny Sullivan (founder of Search Engine Land) to ask his opinion on all of this.

Here’s what he said:

I think Google is probably trying to analyze and use user activity as part of their ranking algorithm. Exactly how this happens is not yet known, it is my opinion that most search engine optimizers prefer response rate as a ranking factor. It doesn’t matter in general. As marketers, the first thing you want to do is engage users with your content. So focus on that and you will most likely be Google compliant.

Danny Sullivan, founder of SearchEngineLand

This is definitely true, but how can you attract users? Here are some options for you:

1. Create BETTER content… Roughly 2k to 10k characters

Pretty obvious, right? But this does not always mean creating more voluminous content. Sometimes the most appropriate result for #1 spot is the most concise article.

Here’s what Eric had to say about it:

I bet if you were to run an experiment that measured the average Dwell Time across millions of sites and their placement on a search results page, you would see a strong relationship between Dwell Time and rankings. Does this mean I consider it a ranking factor? Of course not. It only means that there are a large number of queries where a longer Dwell Time means that the user is more satisfied with the search result than in queries where the Dwell Time is short. There are probably also many examples of queries where Dwell Time doesn’t make any difference as a measure of quality.

Eric Enge, founder of Stone Temple Consulting

Eric’s point that there are “many queries where Dwell Time is not important as a measure of quality” is significant. For example, result #1 for the query “is it Christmas?” This site is isristristmas.com. This is a one page site with one word. – NO! This is by far the best result for this query, although the Dwell Time will be very short.

2. Use the right keywords (and DO NOT use “Clickbait”)

Let’s say I’ve written a blog post called “Advanced SEO Guide”. This is how it will look on the search results page:

Pretty appropriate and appropriate result, right? Not so fast, in this article I will cover things like:

• Adding keywords to title tags and meta descriptions;
• Link Building Guide;
• And so on.

So, the site does not quite meet our needs, as it turned out… This is a vivid example representing the incorrect use of keywords. Something like “learning SEO” or “SEO for beginners” would be much better for this article.

Be that as it may, the content does not live up to its promise – the information in the article is standard, not advanced. Because of this, most users will quickly return to the search results page to find a better match.

Here’s what Eric had to say about it:

Ultimately, what Google wants to see on your site are the people who represent the best match for your content. Be useful to them, and then your goals and objectives will be as consistent as possible with Google.

Eric Enge, founder of Stone Temple Consulting

3. Is the site loading slowly? Intrusive adware? ENOUGH! Make UX/UI your top priority!

Have you ever clicked on a result from a search results page and the following happens?

Annoying, right? For me, this is an instant situation with a “reverse” button, which makes Dwell Time no more than a couple of seconds. But this is not the only version of bad user experience.

The example site above was terribly slow to load. This was confirmed by the speed test result.

Keep in mind that 47% of people expect a web page to load in 2 seconds or less. This way, if your site is too slow, people will return to the search results page before it has fully loaded.

If this happens, users will never interact with your content and Dwell Time will be zero.

Here are a few other common UX issues to keep in mind:

• Layout: The layout of your site should be designed to prioritize your content and make the information easy to read. Remember that people only choose your result once out of thousands of others. They won’t read your article if it’s written in yellow Comic Sans on a white background, no matter how good it is.
• Design: Ugly-looking sites rarely inspire a sense of trust that makes users stay for a long time. So, make sure your design is relevant to your users and appeals to your target audience.
• Mobile optimization: Even in 2018, many websites are still not optimized for small screens. 9 times out of 10 times it’s another instant “back” situation for me. Therefore, even if you are sure that your site is optimized for mobile devices, do me a favor and check again.
• Scrolling without pagination: Implementing continuous scrolling works wonders for sites with a lot of content. Why? So it is easier to continue to perceive information, there are more chances that people will stay. For example, look at Facebook/Twitter. How many times have you sat scrolling through your seemingly endless news feeds?
• Ads: Nobody likes intrusive ads, and Google understands that. In February 2018, Google Chrome began blocking such ads by default. So, don’t cover the entire page with ads – you won’t be making much money anyway.

4. A small number of internal links throughout your content is fine…but make sure they are relevant!

Internal links are not related to the PageRank link ranking algorithm on your site (although there are exceptions). It’s about improving the user experience.

Example:

Google Sheets is one of the tools mentioned in our list of 70+ free SEO tools. Below that mention, we link to our list of 10 Google Sheets every SEO professional should know.

This works because we are directing visitors to content that may be of interest to them. And appropriate in the context of the article. It also increases the likelihood that visitors will view more than one page of your site. This reduces bounce rates, increases Dwell Time and improves overall interaction.

5. ALWAYS keep your content up to date (Stale Content = Fail!)

All content is constantly updated.

Example:

In 2016, we published a huge list of over 200 SEO tips.

But SEO is one of the most updated topics. And, when we reviewed the article in 2017, we realized that a lot of the advice was outdated. We also noticed that many of the 200+ tips were complete bullshit.

So we decided to remove the junk and cut the post down to the top 75 tips.

We recommend keeping the blog articles up to date. For what? Because people are more likely to trust recently updated articles. This increases the CTR (link click through rate) on the search results page and provides a more effective interaction with the content in general.

Final Thoughts

Dwell Time is something Google has experimented with and researched on. But that doesn’t mean it’s a ranking factor. Or at least some reliable way to evaluate the quality and relevance of search results.

No. When it comes to Google, there is only one indisputable fact: they want to make their users happy. And that means showing the best, most relevant result in place #1.

Your job is to be that result.

So, create quality content that meets the desires of your target audience. And rest assured, your site will be visited with pleasure. Do these two things in sequence and you won’t have to worry about Dwell Time.

Translation of an article from ahrefs.com/blog/dwell-time/