“The key is to measure and analyze lots of signals at once, including comments, scroll depth, time spent, shares and other things. If you over-optimize for one variable, then everything becomes the same.”
- Adam Hansmann, Co-founder, Athletic to Digiday
Ever since the advent of the internet, the media world got split between the brand marketers and the performance marketers. Even though they spoke different languages back then, the one word that united these two was - Audience. Audience engagement is and has always been vital in marketing. But unlike the initial days of online publishing, owning this audience and earning its loyalty has become a hard win for both the marketing divisions today.
Engagement has historically been defined as a state of being involved, occupied, retained and intrinsically interested in something.
An audience is the set of people that are committed to pay attention and trust your brand enough to remain engaged.
For publishers, audience engagement is the process of consistently involving readers and earning their loyalty while walking them through the visitor conversion funnel:
Why audience engagement matters:
According to News UK, a brand needs to engage with its audience at least seven times before a successful subscription. But, in a world where information is available immediately and abundantly, the audience attention required for this engagement gets distributed amongst thousands of your likes. And in order to grab the biggest slice of this attention, a reliable identity (brand) needs to earn their trust. It is difficult but it is necessary.
Media publishers are affected by this scarcity the most. They need an audience that prioritizes them over the noise. And the only yet sure-shot way of getting there is consistent audience engagement.
Seems like a jargon? Don’t sweat. Here’s an exhaustive guide on audience engagement that'll help understand this better.
Having an owned audience is a blessing as it is easier to track a registered user. Based on the journey of a reader, the audience experience can be personalized to a larger extent which favors retention.
But building this loyal and engaged audience is critical as it requires a publisher to look for a revenue model beyond advertising. Knowing the fact that it ends up converting your site visitors into recurring paid subscribers, this guide has been put together to help you define, measure and benchmark metrics for measuring audience engagement.
Before we get to the metrics, it is important to assume that you as a brand are doing enough. So here are a few assumptions that we are going to make:
- you are engaging your audience across multiple touch points
- you are being consistent with that engagement
- you are delivering significant value to all your audience segments
Alright. So we know you are talking. But is anyone listening?
Measuring Audience Engagement:
Measuring engagement helps you understand and expand your audience. But, when performed using vanity metrics or misinterpreting the right ones, it could adversely hurt the marketing plan, and ultimately affect the loyalty of your audience.
Different Stages, Different Metrics:
If you’re a new publication, GA traffic report would be the one tab always opened on your device. Traffic reveals the impact created by the content for a new website. But, how relevant would this traffic be for New York Times while measuring engagement? Different website stages have different engagement metrics.
Metrics become vanity when measured at a wrong stage. A website at the "prototype" stage of its growth curve (refer to the graph above), would have entirely different set of engagement metrics compared to the website that has entered the "scaling" stage. A prototype website would target metrics like new users, bounce rate, etc.
However, a scaling or expanding website would also look further into exit rates, active time on page, paid subscribers, rpm, and the list is never ending. Engagement metrics of a website, change with the audience type, audience size, and the engagement level of that audience.
But what is that one thing that changes every time a visitor enters the next level of its conversion funnel - it has to be commitment. Audience engagement is measured in terms of commitment given by a visitor to the brand. This commitment increases as the visitor goes deeper into the conversion funnel and craves for more content.
The ultimate objective of audience engagement remains converting a new user into a recurring paid subscriber. According to Andrea Boetti, head of PayRead, publishers while measuring engagement today, need to “look at the engagement metrics in aspects of their business” which are new sign-ups and paid subscribers.
Both these metrics require you to look at the user interactions: “how many users initiate the registration (or payment) process; how many of those who initiate complete it; which are the steps where users leave the flow and whether those steps can be removed or improved”.
1. Page views and pages per session:
Traffic is the keystone metric for a website. An increase in traffic shows that the website is growing. It confirms that people read what you write about. Also, Google uses traffic to rank sites in the SERPs which makes it enormously vital for websites that are at a prototype or a product stage.
But traffic gets deceiving when we talk about engagement for websites that have started scaling (200+ web pages). “A website that receives traffic in millions is expected to drive high engagement and have high-quality pages” - is marked as one of the common misbeliefs of present-day marketers. Though it could be true the other way around, it is not vice versa. Take your local newspaper, for example. The traffic generated by the digital handle must be reaching tens of millions every month but due to bombardment of so many ads, the audience would be getting minimally engaged. The number of users coming to such websites defines their reach, yes. But, does it tell us anything about the audience they engaged? No.
Websites with over 200 published content pages need to measure their pages per session to get a fair idea of their audience engagement.
Here’s why - an increase in page views per session implies that the visitor didn’t just land on your website, consumed the content it was there for and left. It conveys that the website, the content, and the interlinking were so engaging that the visitor just didn’t want to stop exploring.
How to Measure Page Views:
Traffic or page views are defined as the number of times a webpage is loaded or reloaded. To calculate page views, open your Google Analytics dashboard. From your left sidebar, go to Behavior > Site Content > All pages. Pageviews with respect to the webpages will be shown in decreasing order.
How to measure pages per session:
Pages per session are calculated as the number of unique pages viewed by a visitor in the entire session. To calculate pages per session, go to Audience > Overview.
Tools recommended - Google Analytics
Benchmarks for Pages per Session -
From a survey conducted by Littledata on 3623 websites in July 2019, the average pages per session was found to be 3.0.
If your website receives a 4.7+, you would be under the best 20%, whereas a 6.0+ would put you under the best 10%.
If pages per session turn out to be less than 1.8, your website is under the worst 20%, whereas something less than 1.5 would put you under the worst performing websites.
2. Session Duration and Active Time:
Session Duration or the dwell time is another common metric used by media companies measuring engagement. A higher session duration implies that a visitor can be prompted to subscribe. As per a whitepaper published by WNIP, it was found that 17% of the viewers were able to recall the brand when they saw an ad for 5 seconds or less whereas this percentage hiked to 30.5% when the dwell time was kept more than 5 seconds.
Session duration is calculated as the cumulative sum of time a visitor spends on all the pages of a session and can be represented as:
Let’s understand this with an example:
A visitor came to a website at 10:00, consumed the content for 5 minutes and then hopped on to page 2 at 10:05, spent another 5 minutes there, went to the next page (page 3) at 10:10, spent 5 more minutes on page 3 and then left the website.
In Google Analytics, this time on page 1 is recorded when a user clicks (hits) on page 2. To calculate its time on page, GA subtracts the page 2 hit time from page 1 hit time, which would be 5 here.
Similarly, page 2 is recorded when the user hits on page 3, and time on the page is calculated after subtracting page 3 hit time from page 2 hit time, again 5 minutes. Do you realize that time spent on the page n (last page of the session) would never get recorded as there are no further hits?
The dwell time on site here would be 10 minutes which is incorrect. But, that can be fixed by measuring average session duration hence, is not the major setback here.
Put on your time traveling helmet and enter a universe where you’re promoting a content piece on your social handle. A follower sees your post, clicks on it, visits the page, consumes the whole content, and after 10 minutes of engaged reading, leaves your website.
Now, it was a well-engaged and referred site visitor but the session duration recorded for it was 0 in Google Analytics. All we know that the visitor didn’t click on any other page within the next 30 minutes (default session timeout).
Use tools to calculate the active session time:
Due to the lost time in Google Analytics, session duration alone won’t be able to give you the most accurate engagement stats of your audience. Also, for visitors that website park i.e. leaving a web page opened and hopping off to another website or simply leaving the system unattended, there’s a fair chance of getting erroneous data.
That said, when measured along with tools like Momently and Riveted, session duration becomes a gem for marketers. You can track what your readers are doing in-between the hits by tracking the cursor movements, keyboard activity, mouse clicks, etc, which paints a clearer picture of the active vs the hidden time of that visitor.
How to Measure Session Duration:
Session duration is calculated as the cumulative sum of all the pages and can be represented as (page 1 + page 2 + _ _ _ + page n), where ‘n’ is the last page of that session. To find the average session duration for your website, open the Google Analytics dashboard. From the left sidebar, go to Acquisition > All Traffic > Source/Medium.
How to Measure Active Session Duration:
Active session duration is measured with events in Analytics and plugins like Riveted. Riveted helps build a better understanding as it checks if the user is active and pings Google Analytics in every 5 seconds. It tracks whether the tab is currently visible and if the user spends 30 seconds without registering an action, it is considered idle. The active session duration can be found in Behavior > Events > Riveted. whether the tab is currently visible.
Tools recommended - Google Analytics, Riveted, Momently
Benchmarks for Session Duration -
According to a Databox, a good average session duration lies between 2-3 minutes which means that anything above 3 minutes would put you under the best performing sites.
3. Bounce rate and % exits:
A high bounce rate reflects that readers are not getting enough value out of the information that you have published. It implies that visitors are coming to the website via engaging social posts or SERP titles but are instantly closing the tab due to either the low quality of content or a misleading SERP title, right?
A bounce is a single page session on your website. It is calculated as the number of visitors that only read the first page (no further page loads or hits) divided by the total number of visits on that page. Marked the ambiguity here? If a visitor came to the website, consumed the content, and left after let’s say 15 minutes, the bounce rate recorded would remain 100%.
Evaluating the exit rate for conversions:
The exit rate of a specific page is defined by the number of visitors that left the website after visiting that particular page (examples include the contact information page, sign up page, etc).
For conversion, exit rates could be a better way of analyzing engagement as it impacts the conversion funnel of the website. For example, the exit rate of a page that has a serious job of taking either a return visitor to a sign-up page or a subscriber to a premium content page (behind a paywall) ought to have a low exit rate to assist conversion.
Though, the average bounce rate doesn’t help with the actionable insights as some pages, like the “contact us” page, benefit from a high bounce rate, a high bounce rate of a page can be a strong indicator of problems with user satisfaction - in the context of content, site quality, loading speed, etc.
How to Measure Bounce Rate and % Exits:
To find the exit rate of a page on your website, open the Google Analytics dashboard. From the left sidebar, go to Behavior > Site Content > All Pages. Tools recommended - Google Analytics
Benchmarks for Bounce Rate -
As per the stats of Gorocketfuel, an excellent bounce rate lies in the range 26 - 40 percent. A bounce rate in the range 41 - 55 percent is considered average, whereas, something between 56 - 70 percent is considered higher than average. This might not be a cause for alarm depending on the growth stage of a website. Anything over 70% puts the page in the worst performing ones.
Benchmarks for Exit Rate -
As per TrewMarketing, a good exit rate remain under 50%.
5. New and returning visitors:
Let’s begin with how Google Analytics calculates the new and returning visitors.
When a new visitor lands on a website, the GA checks for a tracking cookie and uses the dimension “user type” to distinguish between a new and a returning visitor.
In the case where no cookie is found, GA creates one and sets the “user type dimension” to new visitors and the “new user metric” to 1. Whereas, if a tracking cookie already exists, GA sets the “user type dimension” to returning visitor and the “new user metric” to 0.
For measuring audience engagement, we stick to measuring returning visitors as they show higher engagement as compared to the newbies. They tend to exhibit low % exits, view more pages per session and have higher session durations. They take part in the discussions, have a higher tendency to turn into a loyal visitor - the end objective.
How to Measure New and Returning Visitors:
Google Analytics report the “returning visits” and not the returning visitors. This means that a visitor might get counted multiple times if there were 1< visits within the reporting period. New and returning visitors can be located in the form of a standard report along with the necessary stats. From the left sidebar in your Google Analytics dashboard, go to Audience > Behavior > New vs returning.
Tools recommended - Google Analytics, GoStats.
Benchmarks for New and Returning Visitors -
There’s no single answer to this. The benchmark here would completely depend on the industry type, audience targeted, and the objective you are trying to achieve with your website. In general, having 30% returning visitors is good and having 50%+ is marked great. A decent influx of new visitors, 50 to 70 per cent, is also vital as that’s what keeps the engine running.
6. Paid subscribers:
A hike in the number of paid subscribers brings the widest possible smile on a publisher’s face. Getting more paid subscribers is the ultimate objective that a publisher hustles for - every day.
A visitor enters this conversion segment when it craves for more quality content. And, the only way to get there is to pay for the subscription. Though these stats would be mentioned on top of your paywall solution dashboard, what keeps these numbers from growing is where we need Andrea’s views again.
According to him, most publishers are still sticking to the traditional means for user registration. These include email-based accounts, credit cards, PayPal payments, etc.
In order to accelerate the subscription process, he suggests that “publishers need to simplify the checkout process and use mobile operators to capture the pockets of audience segments that do not want to pay with their credit cards.”
These segments include audience “who don’t have access to the payment method, who don’t have it at hand and those who do not want to share their personal information due to privacy concerns”.
Tools recommended: Paywalls like PayRead, LaterPay, Piano, etc.
Benchmarks for Paid Subscribers -
Well, the more the better. According to WNIP, stop rates are the very strong predictor of the overall subscription sales. Stop rates are the % unique visitors that are stopped by a payment gateway. The most successful publishers stop 5-8 percent of their digital audience. And, here’s what the trend for successful pubs suggest.
7. Retention as a Metric:
Google defines retention as the process of continued possession, use or control of something. In terms of audience engagement, it is the ability of a website to retain its subscribers over a specific time period. There are two key features of retention -
- It tends to drop really fast at the beginning (unless you’re one of the NYTs or the WSJs of the media industry, this shouldn’t depress you)
- Over a period of time, it either settles at a constant value or hits the zero mark. (hitting zero signals that either you have stopped engaging with your audience or there’s something terribly wrong with the engagement practices)
The objective remains to keep the retention curve from dropping at the start. Brian Balfour from HubSpot told Popcornmetrics that retention in week 1 "has one of the biggest impacts on your retention over time" because improvements made in week 1 retention are carried through the entire retention curve.
Retention of Email Subscribers:
Based on interactions with 3.2 Billion users, Sumo reports that 1.95% is the average opt-in rate that the emails receive. So if your website is getting anything above 2%, your engagement mission accomplishes, right?
For a website receiving 100K unique monthly visitors, 0.5% average weekly subscription, and 3% average weekly subscriber churn, here’s what the retention looks like:
Subscription rate tells you the number of visitors that choose to be engaged by you. Retention rate, whereas, filters the most loyal ones and informs how many of those visitors chose to remain engaged. Retention as an engagement metric, matters. Looking at the table pasted above with the active subscribers getting almost halved, we now understand why.
Retention of Push Notification Subscriber:
Churn measures your audience building failures. It tells you the number of loyal visitors that left the conversion funnel of your website. While working on our annual report on the state of push notification ads, we went deep with the data that we accrued from the performance of 45B+ notifications. It was found that churn gets a gigantic boost of up to 25% as the number of notifications increases from 3 to 7.
Knowing that a few of our publisher customers were still not sailing the “less is more” boat, this 90-days retention count hit us in the sore spot.
Subscription count is what most publishers address as their core metric for higher engagement. Taking a step further and considering supporting metrics like retention and churn while measuring subscriptions didn’t just paint a clearer picture of their leaking conversion funnel but also helped them understand the kind of content that worked.
8. Conversation rate of the brand:
For returning visitors, the conversation rate has to be the next metric you should be looking at. It doesn’t require you to go through the complex GA reports or find solutions to patch the loopholes. It shows the real engagement on your blog in terms of discussions had in the comment sections.
The conversation rate for a defined period is calculated as the total number of comments received by the total number of blogs or posts published by a website.
CR - Number of comments / Number of Posts
Tools recommended: Conversation rate depends on the comments, likes, and shares. It is manually calculated for a set period of time and doesn’t require you to have a solution in place.
9. Trends in Analytics:
Trends in Google Analytics does the job of measuring the overall engagement of the audience over a period of time. So, if your GA says that the average session duration is going up, it means that the audience, in general on your website, is consuming more content these days.
Therefore, if content page X shows more “time on page” whereas, page Y and Z drive low engagement, you need to create more content around X and less around Y and Z.
Tools recommended: Google Analytics
One metric doesn’t narrate your entire engagement story. To measure how engaged your website audience most accurately is a cumulative analysis of the above-mentioned metrics would remain a recommended approach.
Optimization: Google vs Audience
Webpages are like weapons for a publisher to compete in the Google SERPs. In order to get a higher rank, optimizing these webpages at times appears to be necessary. But let’s take a step back and understand this by keeping an eye on our end objective. Is ranking higher in the SERPs your ultimate goal? Each effort that a publisher makes is to engage its audience. And, giving it an engaging experience is where the competition needs to essentially remain.
Remember - Optimizing for Google is crucial but optimizing for your audience ensures long lasting results. Let’s avoid compromising the latter to win the former.