Now that we’ve discussed how you can boost CTR, it’s time to identify the metrics that tell Google whether or not your content satisfies users and makes them happy.
Because yes, Google does measure this.
How? By looking at a variety of engagement metrics including:
- CTR (as discussed in Part 1 of this series)
- Bounce rate
- Dwell time (what we’ll be unpacking here).
Here we’ll review how Google uses machine learning to determine how satisfied a user is after clicking on a result (via their dwell time).
What is dwell time and does Google really use it?
Dwell time is a metric that uses a combination of SERP CTR and session duration to calculate user engagement. In simple terms, dwell time is the length of time a user spends reading your content after clicking on your site in the SERPs.
The longer someone spends on your site, compared to your competitors, the better.
An example of dwell time:
A user lands on your site and bounces back to the SERPs after 10 seconds. This suggests to Google that your content didn’t match the needs of the user’s query. And if this pattern occurs often enough, your site will gradually drop down the rankings. This is all logical, and it makes sense for Google to use machine learning to look at things like this.
But don’t just take my word for it. Here’s an excerpt from a patent released in 2015 explaining how Google modifies ranking data.
“One method includes receiving quality of result data for a query and a plurality of versions of a document, determining a weighted overall quality of result statistic for the document with respect to the query including weighting each version specific quality of result statistic and combining the weighted version-specific quality of result statistics, wherein each quality of result statistic is weighted by a weight determined from at least a difference between content of a reference version of the document and content of the version of the document corresponding to the version specific quality of result statistic, and storing the weighted overall quality of result statistic and data associating the query and the document with the weighted overall quality of result statistic.”
The patent basically tells us search results may be ranked based on scores they are assigned by search engines (which we already knew). It also reveals Google measures how users engage with a result and compares this data with similar sites to determine rankings. And while it doesn’t explicitly say that ‘dwell time’ is one of these metrics, data and studies do support this theory.
Backlinko carried out a study on 1 million search results and found that low bounce rates are strongly associated with higher rankings.
So, a high bounce rate is always bad… right?
Well, no. Not exactly. It’s all relative to the type of query a user is searching for and how competing sites are performing.
I get asked this a lot. People often panic when they see bounce rates of 90% on some pages. Most site owners assume that anything over 60% is poor and that something isn’t working or needs to be changed. It’s easy to jump to conclusions and assume the content isn’t performing how it should and users aren’t engaging with the site.
However, this isn’t always the case. In fact, it’s extremely common to have bounce rates as high as 90% on certain pages – even if there’s nothing wrong with the content.
Bounce rate should be looked at on a case-by-case basis because it all comes down to intent. The big question is: What does the user need from the page? (And is it possible for a user to bounce after getting what they needed?)
For example, let’s say a user is looking for address details or a phone number and lands on a page like this:
All the user may want to do here is note down the phone number and leave the page. In many cases, this means there’s no need for a user to interact with the page, which means they will be counted as a ‘bounce’ when they leave the session.
Would you consider this a ‘bad’ experience? Not at all. And neither will Google, because the user has shown their query has been satisfied by not returning to the SERPs to find an alternative result.
Bouncing vs. pogo sticking
If a user lands on a page and does return to the SERPs to click on another result (unlike the example above), this is known as “pogo sticking”. I believe (contrary to what Google has recently announced), that in many cases, this is used to signal a bad user experience and query dissatisfaction.
Again, context is everything here. As John Mueller pointed out in this Google hangout, there may be lots of reasons a user might go back and forth in search results. This is why your performance relative to how competing sites in the SERPs are performing is paramount, as this is the only way Google can use metrics like dwell time to fairly determine the best result for a query.
For example, if your dwell time is significantly lower than a competing site for the same query, which site has delivered the better experience in the eyes of the search engine? Yep, the competing site has won that battle.
Since that’s not what you want, let’s look at some ways you can increase dwell time and outperform the competition.
How to boost dwell time and minimise pogo sticking
There are many methods that focus on boosting dwell time, but I want to concentrate on the ones that are easiest to implement.
Make sure your site loads in under 2 seconds
The rise of mobile search and app usage has led to many users expecting an ‘app like’ experience when visiting your website, meaning that every second counts.
In fact, a study carried out by Kissmetrics shows 47% of users expect a page to load in 2 seconds or less, and 40% of users will abandon a site that takes more than 3 seconds to load.
You could have the best content on the net, structured perfectly for readability and engagement. But if users are abandoning your site because your speed isn’t up to scratch, you’re sending negative ranking signals to Google before you even give your content a chance to perform.
Avoid intrusive pop-ups
If pop-ups are used well, they can have a positive impact on conversion rates. But it’s essential any pop-ups you use are not intrusive or having an adverse impact on the user experience.
Interstitials are a type of pop-up that most users (including me!) find extremely irritating. A good example of a site that uses these is Forbes.
Quick tip: AVOID these! Most people hate them. And Google isn’t a fan either.
If you plan to use pop-ups, consider the following:
- Timing: Make sure your pop-ups don’t interfere with the browsing experience, especially the step between the SERPs and on-page load. If your pop-up is triggered by a certain scroll percentage or after a couple of minutes, this shouldn’t have a negative impact. For example, if a user scrolls through an entire product page and is then presented with an offer, this might be something a user would respond to better.
- Interaction: This one is simple. The pop-up needs to be easy to close. If a user can’t quickly determine how to close it, there’s no doubt it will cause frustration.
- Measure the impact: If you decide to add a pop-up, make sure you’re measuring the impact it has on user experience. You could look at metrics like bounce rate and time on site after the element appears to track this. If you see the pop-up is having a negative impact, it’s probably time to change your approach. Don’t make any rash decisions here though; collect enough data to make an informed decision first.
Improve content structure and readability
If you really want your audience to engage with your content you need to structure it so it’s more palatable for the reader, particularly if you’re writing long-form content.
Here are some elements to consider:
- Short paragraphs: Reading a lot of content is difficult and you’ll lose a user before they even get started if they’re presented with huge blocks of texts. Short paragraphs help you break the content up into digestible chunks.
- Use media and images: Images and videos help break the text up and add visuals to the page. It’s also important to consider that users consume content in different ways; some like reading text, but a study from Hubspot demonstrated that 4X as many consumers prefer to watch a video about a product than read about it.
- Compelling subheadings: Just like short paragraphs, subheadings help break the content up and allow a user to scan through your page and find elements they want to read.
- Links to other content: This is particularly important for blog content. You need to make it easy for users to find other content you have written that might be a good fit for them. Your internal linking structure should always consider what users could find helpful.
Nailing all these elements is essential to keeping users engaged with your site for longer and increasing dwell time.
However, we haven’t discussed a key factor: making sure the content you write matches the user’s search intent. None of the above is going to do anything for you if your actual content is wide of the mark and doesn’t satisfy all of the user’s queries. This is what we’ll cover in part 3 next month.