posted by Caitlin O'Connell
Late last year we explored the state of push notifications and found that over the course of 2015, push notifications continued to improve app user engagement and retention. We also saw early signs of companies shifting away from broadcast messages sent to all of their users to more personalized messages geared toward specific audience segments.
This trend has continued into 2016, with many apps are driving toward an even more “individualized” approach. Individualized messages take into account a user’s actions within the app as well as if the user fits into a particular audience based on their profile attributes. Instead of basic “Hi [first name}” personalization, individualization is about truly delivering the right message to the right user based on their previous actions.
As we look deeper into the state of push notifications in 2016, we’re encouraged to see many positive signs.
As mentioned above, we continued to see the volume of broadcast messages decrease in 2016. Just last October we reported 65% of push notifications incorporated some sort of audience segmentation while 35% of messages were still being broadcasted to an entire app’s user base. Today, our data shows 75% of push notifications are segmented and only 25% are broadcast. That’s great progress for app marketers.
Segmented, targeted messages see success in 3 very important metrics: click-through rates, engagement, and click-through conversion rates. Segmented messages average a click-through rate of 7.4% within the first week of the message being sent, as compared to broadcast messages which only yield a 4% click-through rate in the first week. While click-through rates can be viewed as a vanity metric, it’s still a useful way to gauge how enticing the message was in getting users to respond.
Engagement, which we define as the average amount of sessions a user has within the first week after receiving a push message, is also much stronger from segmented messages. We see an average of 6.4 sessions for segmented messages versus 2.8 sessions for broadcast messages -- that’s more than twice as many sessions just within the first week.
Lastly, click-through conversion rates, which are measured as the percentage of users who clicked on the push message and then also performed the determined conversion event (ie. completed checkout, watched video, etc), averaged 3% for segmented messages and a measly 0.6% for broadcast messages.
When it comes to addressing app users through push messaging it can be done it two ways: personalizing the message and individualizing the message. We touched on the idea of personalization a bit in our previous in-app messaging report, but as a refresher it includes low hanging methods such as including a user’s first name in the message or writing the creative in a specific language.
While this level of personalization is a step in the right direction, these messages tend to generate below average click-through rates and engagement. Messages with the user’s first name generate 5.5% click-through and average 2.7 sessions within the first week, and messages in a specific language generate click-through rates of 5% and average 3.5 sessions within the first week.
These messages likely see lackluster results because they do not take into account any actual information about the user’s habits or activity within the app. If a push message talking about the Yankees is sent to a Red Sox fan, but includes their first name, that minimal level of personalization will not be enough to entice the user to engage with the message because it is not about what they really care about.
This is why the move to individualization is needed -- it’s necessary that app marketers incorporate profile information about their users as well as behavior information from activities they’ve taken within the app in order to be successful.
There are 3 ways to create push notifications with a more individualized mindset:
Let’s look at how each of these types of segmentation effect our 3 core metrics: engagement, click-through rate, and conversion rate.
Measuring the average amount of sessions users have after receiving a push message is a great way to judge the satisfaction of those users. The more sessions had off a push message proves it provided those users with value and gave them reason to return. Not surprisingly, the more segmentation added to the message leads to more sessions.
Apps are sure to see increases in engagement by creating specific audiences based on user information, but sending a message to an audience based on both their profile and behavior yields the highest payoff. These messages average 4x more sessions than broadcast messages.
Click-through rates are helpful in measuring the success of the creatives of a push message. As click-through rates are the first line of response, the information conveyed in the creative will persuade users to react immediately or push it to the side. Once again we see a positive ascension in click-through rates as segmentation increases.
The more specific an audience, the more specific the information displayed in the push can be. If a user interested in red sneakers receives a push message detailing a sale on red sneakers at the store in their town, the chance of them reacting to this immediately is high. These types of messages have an 11% click-through rate and prove the ability to detail the most relevant information possible is crucial to drive that immediate reaction.
While click-through rates are great for measuring how successful the creative of a push was, click-through conversion rates help to determine how effective the campaign was as a whole. Conversion rates detail how impactful the push was on getting someone to the desired finish line.
Profile-driven and behavior-driven messages both illicit impressive 2% clickthrough conversion rates, but combining those attributes doubles conversion for an average of 4%. This is is because as the audience becomes more specific, so too can the conversion event. When apps are able to first identify which users are most likely to convert and then figure out the when, why, and how of it all, they are able to create this individualized message strategy that is hard for users to ignore.
We are encouraged to see positive signs of individualized push messaging strategies playing out across vertical industries, with Retail and Travel apps leading the way.
Retail apps have seen a significant lift in our 3 core metrics for these individualized messages compared to when they send broadcast messages. Engagement more than doubles, click-through rates more than triple and they see a 9% increase in conversion. As mobile becomes a bigger story for the eCommerce space, push messages are a proven way for this industry to tackle the mobile world.
Travel apps have also seen impressive results for their messages segmented by both profile and behavioral attributes. These messages average just over 7 sessions in the first week after a message is received compared to 1.06 sessions for broadcast messages. Their conversion rates also triple and they see an 8 point increase in click-through rate.
There are a lot of great things happening in the realm of push messaging, but in the ever changing world of mobile, the work will not stop anytime soon. But we are encouraged to see many marketers taking the right steps and focusing on sending the right messages at the right time with creatives that match the behind-the-scenes segmenting.
We’re also excited about what the introduction of rich push will mean for push messaging campaigns going forward. With new features and innovations like rich push being added all the time, marketers need to jump on board and keep driving toward the best user experience possible.
Localytics is the leading analytics and marketing platform for mobile and web apps across more than 1.5 billion devices and 28,000 apps. Localytics processes 50 billion data points monthly. For this study, Localytics looked at apps that have integrated push notifications across both iOS and Android. Localytics also analyzed the sessions, clickrates, and click through conversions within a 7 day period for these push notifications. The timeframe for this data was July - Oct 2016.
Thanks for signing up. Look for your first email shortly!
We’ll reach out shortly to schedule a time to talk.