Overview
Building an Audience is essential to sending a relevant message to your users. Audiences allow you to segment your users based on their behaviors within your app (and your messages) and their profiles. Audiences can be built using Events, Sessions, or Profiles data.
Audiences are also highly customizable. Some profile and behavioral attributes are tracked by default with the Localytics SDK out of the box. While others you can create and track using your own additions to the SDK. Talk to your development team or your Mobile Engagement Consultant (MEC) for more information.
When building an Audience, you’ll want to keep two things in mind:
- Segmenting your users to give them the most possible value from the message
- Keeping the audience large enough that you’re actually getting a return on your investment in the message
Building an Audience
The first thing you’ll want to do is give your Audience a name. Make sure to give your Audience a descriptive name—as you’ll be able to access your Audiences from different parts of the product (messages, remarketing) by its name.
When you build an Audience, you can pick Behavioral Conditions or Profile Conditions. You can also build an Audience of both.
Behavioral Conditions
To add a Behavioral Condition, click the + Add Behavioral Condition button.
To begin, you’ll have the option of choosing to target behaviors on each individual device, or across a person’s devices. So if you’ve got a user who’s logged in on both their iPad and iPhone, you don’t have to worry about double messaging them.
You also have the option of choosing the timeframe of the behavior. The default Localytics options are:
- Last 180 days
- Last 90 days
- Last 60 days
- Last 30 days
- Last 7 days
If none of these options work for you, you can select your own last day option by choosing last… from the list, and filling in your own number of days.
You can also choose a static date range by selecting static range… from the drop-down.
Finally, you can get even more granular by choosing an hour window. Pick a window of time (ie, within the last 3 to 5 hours) to target users within.
Next up, you’ll want to pick the actual behavior users need to perform (or not perform) in order to qualify for the Audience.
You have four options for how your users need to interact in order to qualify for the Audience:
- Performed event
- Did not perform event
- Had a session
- Did not have a session
We’ll start will talking about performing (or not performing) a certain tagged event in your app. You’ll want to work with your development team to build out custom events for your app, but by default, the Localytics SDK will track:
- Localytics Inbox Message Viewed
- Localytics In-App Displayed
- Localytics Place Entered
- Localytics Place Visited
- Localytics Places Push Opened
- Localytics Places Push Received
- Localytics Push Opened
- Localytics Push Received
- Localytics Push Registered
- Localytics Push Sent
- Localytics Re-engagement
- Localytics Uninstall
If your SDK is installed properly, you should see these options in the drop-down menu labeled (choose event).
Once you choose an event, you’ll be given options to add more details to the behavior. Depending on which interaction you’ve selected from the dropdown (performed event, did not perform event, had a session, did not have a session), you’ll be given different options for further granularity.
Here’s a quick example:
Behavior: Across a person’s devices during the last 30 days
People who: Did not perform event Localytics Push Opened
In the past 30 days (but were active in the above range)
Where Push Enabled is one of Enabled
You can add up to 5 profile dimensions to your behavioral condition (ie, the drop-down menu labeled Where), or just skip them. We’ll talk more about profile attributes in the next section. Use the Add Filter button to add more filters, and use the minus button to delete them.
Finally, if you’re looking to further refine (or further open up) your Audience, you can add a second behavioral condition.
Again, choose your behavior from the dropdown list and refine it if necessary.
Between the two behaviors, you should see an AND button. This button allows you to specify how the two behaviors interact with each other. Your options are:
| Modifier | Meaning |
|---|---|
| AND | User must fulfill both of these behaviors |
| OR | User can fulfill either of these behaviors |
| AND THEN | User must fulfill both behaviors sequentially |
Once you’re satisfied with your Behavioral Conditions, feel free to leave the audience there, or to add Profile Conditions.
Profile Conditions
A Profile, at its base, is a user-level record. You can attach different properties (what we call Profile Attributes) like Gender or Last Website Visit. Profile Attributes can be dates, text fields, numeric values, or a non-ordered collection called a set.
Profiles allow you to select a group of users with a shared set of characteristics (what we call attributes). When building an Audience you can select up to 5 profile attributes.
It’s important to note that profiles are memory-less—so they can only contain the user’s current state, with no associated history.
To add a Profile Condition to your Audience, click the + Add Profile Condition button. This will open a menu for you to build your profile conditions.
Like Behavioral Conditions, Localytics will track a few Profile Attributes by default through the SDK. If your SDK is installed properly, you should see the following options available to you in the drop-down menu labeled (choose attribute):
| Scope | Profile Attribute | Description |
|---|---|---|
| App/Org | User type | Localytics detects whether a Profile is Known or Anonymous. |
| App/Org | Last session date | Localytics detects the last observed Session per Profile. |
| App | Last CustomDimensionName |
For each Custom Dimension tracked by your app, Localytics will preserve the last observed value per Profile. |
| App | Last app version | For the last logged Session per Profile, Localytics captures the app version. If a user updates the app subsequently without recording a Session, this Attribute may not reflect current app version. |
| App | Last OS Version | Localytics detects the last OS Version being used by a Profile. |
| App | Total sessions | Localytics computes the total number of observed Sessions per Profile. |
| App | First session date | Localytics detects the first observed Session per Profile |
| App | Push enabled | For the last observed Session per Profile, Localytics detects whether push messaging was enabled or disabled. If a user updates this subsequently without recording a Session, this Attribute may not reflect current status. |
| App | Last push opened date | For the last push opened, Localytics detects the date the Profile opened it. |
| App | Last in-app displayed date | For the last in-app displayed, Localytics detects the date the Profile opened it. |
| Org | Last Country | Localytics detects the last country a Profile was in. |
| Org | Last time zone | Localytics detects the last time zone a Profile was in. |
| Org | Last city | Localytics detects the last city a Profile was in. |
| Org | Language | Localytics detects the last language a Profile was using. |
You can also create as many custom attributes as you’d like. Read more about custom profile attributes here.
Select your profile dimension from the drop-down menu. Depending on the type of attribute (numeric value, text field, etc), you’ll be able to choose a condition to further define the attribute.
Here’s a quick example:
Profile
People where Last Push Opened is on or before 14 days ago
AND Language is one of English
And like with Behavioral Conditions, you can further define your profile attributes with the AND/OR selector. Clicking the button will flip between the two.
Add additional Profile attributes by clicking the Add Filter button. Again, you can add up to 5 attributes to your Audience.
Calculating the Size of your Audience
Finally, you can calculate the size of your Audience from within the Audience builder. Click the Calculate the number of users in this audience link.
Keep in mind, this is just an estimate, and might not fully align with the total number of users who end up receiving your message. For more information on why that is the case, check out this article.