Advanced Segmented Push is here and together with insights from Google Analytics, this new segmented push is a powerful tool for alerting, informing or even catching consumers before they churn. Google Analytics Segment Builder is now integrated into the Visiolink Hub, and this allows for almost infinite different ways of segmenting readers and devices. This article will explain how it works and how to create segments using the Segment Builder. Furthermore, it will also explain how to send push messages via the Hub.

Segmented Push in 3 parts:

1. Push Token Collection
2. Segmentation
3. Send Push Messages 

1. Push Token Collection

A Push Token is a unique identifier which links to a specific device and app.

It is important to note, that the Push Token does not tie a device to a specific user, but instead only links to a specific device and app combination. Maybe one user has several devices and in order to tie these together, it would be necessary to implement user tracking into the solution.

When loading the app first time, users can choose to opt out of push notifications. If they choose not to receive push notifications, the Push Tokens will not be collected. If they say yes to notifications, the Push Token is collected and stored for later use.

Users can also turn on/off notifications at a later point via the settings menu. 

2. Segmentation

Users can be segmented into groups in almost infinite different ways. A good segmentation creates an opportunity for efficient and differentiated communication. Google Analytics provides a powerful tool for segmentation called Segment Builder. The Segment Builder works by segmenting users based on filters in Google Analytics. 

The 6 main categories for segmentation

Custom (Advanced -> Conditions)

• Use or combine Visiolink’s key metrics for segmentation
• Find groups with high interest in “cars” or “investment” section
• Target only readers with a high or low reading engagement
• Target readers who spent more than 30 seconds on a specific article

Demographics

• This information is obtained by Google and can often be lacking
• With user tracking enabled, a more detailed segment can be built using CRM data

Technology

• Sort based on OS, browser, screen size, phone model, software etc.
• E.g. useful for targeting users on an old version of the app

Behavior

• Sessions in total, days since last session, session duration
• E.g. can be useful for categorizing inactive users

Date of First Session

• This date can e.g. be used to track campaign performance
• Track number of active users since their onboarding last month

Traffic source

• Sort on how users found you e.g. from a specific campaign etc.
• E.g. find users who followed a specific link in an emailing ad

Based on wanted results, the segmentation can be performed on user level or on all data.

When segmenting on user level (user tracking only), data from users with multiple devices (pc, phone and tablet etc.) will be aggregated and used to create more precise segments.

It is important to use the UserID view in Google Analytics when performing segmentations on the user level. This view is created by Visiolink if user tracking is implemented.

Segmentation with and without user tracking

The segmentation is handled differently with known readers than with unknown readers:

1. Segmentation WITH user tracking (known readers)

With user tracking implemented it is possible to send messages to unique segments based on knowledge from the individual readers. The collected Push Tokens can be tied to specific users and thereby create much more detailed behavioral data and a more detailed segmentation.

User tracking also opens for the possibility of implementing the company CRM database in the segmentation process, which allows for a more detailed and rich segmentation based on e.g. age, gender, subscription type, residence or historical behavioral patterns from the company website.

Here are some examples of this type of segmentation:

• Message users who read specific articles about the royal weeding
• Message females between 35 and 55 about today’s special home decoration section
• Message students on a trial subscription
• Message users in Copenhagen to inform about a delay in the printed paper
• Message males who entered the company website's camping section

2. Segmentation WITHOUT user tracking (unknown readers)

Segmentation is still possible without user tracking, but the knowledge is instead based on information from the devices instead of the individual users. The segmentation is still very useful, but less detailed than the segmentation with user tracking. 

Here are some examples of this type of segmentation:

• Message only devices which have been inactive for more than 3 days
• Message only devices with more than 7 downloads in a week
• Message only devices with freebie downloads (users with potential)
• Message only devices with specific article downloads (content-based)
• Message only devices who did not download newspaper on release (behavior-based)

This can be useful for retainment campaigns, loyalty programs, surveys about specific content (e.g. only send a survey to devices which downloaded the publication about a new local bike lane).

When you know how you want to segment your readers, you can proceed to build segments using the Segment Builder in Google Analytics. The Visiolink Hub will then fetch these created segments and use them for push messages. Next section describes how to use the Google Analytics Segment Builder. 

How to create segments in Google Analytics

First step is to login to the Google Analytics account of your organization.

Then navigate to the Admin page (STEP 1) and press on Segments (STEP 2).

After this is done you will see all created segments for the account.

Below is a picture illustrating this screen, including four custom segments.

Press New Segment (STEP 3) to begin the process of segmenting.

As an example, we will target readers who have been inactive for more than three days but have been reading articles about Books and Culture within the last month.

Below the conditions have been set to capture these specific readers. 

A filter with “days since last session > 3” has been applied together with two custom conditions. First custom condition includes only sessions from last month. Second custom condition includes only sessions within the topic “Bøger & Kultur”.

After the filters have been set, the segment has to be shared with collaborators.

This is done by pressing “Change” in the top right corner of the window (STEP 4).

Below is an illustration of this new window with the proper settings.

Press ‘OK’. The segment has now been created and you are ready to send push messages. Next section will explain how to send push messages to the created segments using the Visiolink Hub.

3. Send Push Messages

When the segmentation is complete and satisfactory, next step is to send out messages.

The push messages are sent to the reader’s device and shown on the lock screen or message center of the device. Push messages are often used to inform the reader that a new issue is ready for download and for telling the most important news stories of today’s issue. When the reader opens the push message, the app will open (or a website if this is specified).

First make sure that you have Segmented Push enabled for your e-paper. Then login to your Visiolink Hub account and make sure you are on the top level of your organization.

Then you will see a screen similar to the one below.

Navigate to the feature “Push Messages”.

Choose “Segment by Google Analytics” and choose the relevant view in the first drop down menu. Below, the view is called “all data [PUSH]”, in your case this will be named differently.

When the relevant view for your e-paper has been chosen, your created segments will appear in the second drop down menu. Earlier we created a segment for inactive users with an interest in books and culture. As an example, we will choose this segment in the second drop down menu. 

Then the message can be written. It cannot contain more than 255 characters.

It is possible to include an URL to open when clicking on the push message. By including an URL, the user will be directed to the website instead of the app.

Furthermore, it is possible link directly to the front page or inside a publication using Deep Linking, a guide to Deep Linking can be found here. If you want to use Deep Linking, make sure you do not use the same link for iOS and Android users.

A date and time can be set, and several push messages can be created ahead of time.

When everything looks good press ‘SEND’ and the Hub will start processing the message.

After pressing ‘SEND’, a 5 min delay has been implemented in order to regret a decision or modify the message. Meanwhile the status will be yellow as seen below.

After 5-10 minutes, the status will turn green, and the message has been sent to the specified users/devices. When the status has turned green, it is no longer possible to delete the message. Each push message will include the number of recipients.

Restrictions and limitations

Push messages are limited to one message per title, every 10 minutes. If a push message is sent to a title and you try to send a second message within 10 minutes to the same title, you will get an error. Also, you cannot send push messages back in time. This will also result in an error. The Segment Builder also offers segmentation based on a date range. This date range is limited to 31 days. If a larger date range is needed, instead use two conditions with “Is on or after ..-..-..” and “Is on or before ..-..-..”. This has no limitations.

Impact

If user tracking is enabled on your e-paper, then it could be interesting to check how many users were activated via the push message and their behavior in the following days. This could be done via your Google Analytics account.

________________________________________________________________

If you would like more information or guidance, 
feel free to contact our Business Consulting department.