Introducing the OpenBack Periodic Table of Signals
Here at OpenBack, we have explored in previous Newsroom posts our unique data signals that trigger delivery of our push notifications. Our hybrid platform, which uses mobile edge computing and machine learning to process user data directly on the device, is able to leverage data at a far more sophisticated level than our industry competitors. All of our signals can be combined to pinpoint the exact perfect moment of delivery. And because of our platform’s highly reliable deliverability, mobile apps can trust that notifications will arrive at the device at the exact moment they’re meant to. We’re now very excited to unveil our official OpenBack Periodic Table of Signals, for those who would like them laid out all in one place.
Download our FREE Mobile Marketing Playbook to perfect your user engagement game!
This branch of the periodic table allows you to select timing for delivery of your notification. The Now signal sends your notification immediately. (Which is not advisable, as OpenBack studies showed this signal resulted in a drop in click-through rate of 111%, when compared to combinations of other signals.)
There is also the option to select a specific window of time for notification delivery, which performed slightly better than the Now signal. Still, the fact that both signals saw considerably lower CTR than the control group shows that determining the perfect moment for users to read your notification is a complex process. Simply selecting a time of day is not enough.
Device-Side Machine Learning
This branch of signals all employ machine learning (ML) to form an algorithm that will inform moment of delivery based on user device-side behavior. These include:
- Device-Decisions ML
- Interact ML
The Device-Decisions (DD-ML) signal has seen the most exciting results so far, with a 39% jump in CTR. DD-ML combines your users’ behaviors on-device into an algorithm that takes into account changes in their behavior pattern in real-time. For example, if your user wakes up 5 days a week and goes to a secondary location, where they stay for 9 hours and only sparingly use their device, DD-ML will block that out as a no-send window of time. It will send notifications once the user is back at their primary location (which is probably home, after work). However, if once a week the user goes to a third location (possibly a night class, or a bar for happy hour), DD-ML will take that into account, and hold off sending until a more convenient moment.
Usage and Device Segments
You can combine segment signals with other advanced signals, first to create a segment of users of certain characteristics and then to signal the moment for the notification to deliver. Usage Segments are the following:
- Last Open (times delivery according to the length of time since the app was last opened)
- App Open (signals an in-app message to deliver when the user opens the app)
- Custom Segments
- Custom Events (in which you can assign a signal to any event that occurs on the device to deliver a notification)
You can also segment users according to their device characteristics, including:
- Language (which language the device is set up in)
- Version (which operating system runs the device)
- Device (whether it’s an Android or iPhone)
- Mobile operator (which system the user has a contract with)
- Apps (which apps the user has installed on their device)
OpenBack Delivery Moment Signals
These signals leverage contextual data from the device, and can be combined to ascertain the perfect moment on the user’s end for them to receive a push notification.
- Unlock (sends notification when device is unlocked)
- Headphones (sends notification when the headphone jack is in use)
- Roaming (sends notification when device roaming data is engaged)
- Power (sends notification when the power button has been pressed)
- Airplane mode (sends notification when airplane mode is engaged)
- Volume (lets you assign moment of delivery to when the volume is at a certain level)
- Proximity (sends notification when the device is near a certain physical location)
- Signal (lets you assign moment of delivery to when data signal is strong)
- Brightness (lets you assign moment of delivery according to how strong the brightness on the screen is)
- Orientation (lets you assign moment of delivery according to which physical direction device is oriented)
- Noise (lets you assign moment of delivery when there is low or zero background noise)
- Connectivity (lets you assign moment of delivery according to nearby Wi-Fi strength)
- Memory (lets you assign moment of delivery according to memory levels on-device)
OpenBack Delivery Location Signals
These signals can be combined with others to deliver the notification according to the device’s physical location. These can be useful if, e.g. the user walks past a brick-and-mortar shop associated with your app. Or, in such AR reality games such as Pokemon GO, they can be used to send notifications about nearby elements that are part of the game.
- Location (ML): This can either pinpoint your user’s exact geolocation or their country
- Places (ML): This signal uses location patterns in the device’s location to determine whether your user is at work or home
To read more about location signals, how to use them, and how to convince users to allow your app to access their location, read our previous Newsroom post.
Advanced controls you can implement on the OpenBack dashboard include our Delay and Fallback features. On the app backend, you can also implement the Identity feature.
For more details on OpenBack Signals and how to use them, you can link to our Documentation page.
Or, to learn more, you can get in touch with one of our experts to schedule a demo.