Analytics for iOS

Analytics for iOS makes it dead simple to send your data to any analytics or marketing tool without having to learn, test or implement a new API every time.

All of our libraries are open-source, so you can view Analytics for iOS on Github, or check out our browser and server-side libraries too.

Want to stay updated on releases? Subscribe to the release feed.

Note: At the moment we can’t support tracking of watchkit extensions for the Apple watch. Email us if you’re interested in a watchkit SDK. For now we recommend tracking watch interactions via the iPhone app code.

Getting Started

Install the SDK

The recommended way to install Analytics for iOS is via Cocoapods, since it means you can create a build with specific destinations, and because it makes it dead simple to install and upgrade.

First, add the Analytics dependency to your Podfile, like so:

pod 'Analytics', '~> 3.0'

Then in your application delegate’s - application:didFinishLaunchingWithOptions: method, setup the SDK like so:

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];
configuration.trackApplicationLifecycleEvents = YES; // Enable this to record certain application events automatically!
configuration.recordScreenViews = YES; // Enable this to record screen views automatically!
[SEGAnalytics setupWithConfiguration:configuration];

Note: Automatically tracking lifecycle events (Application Opened, Application Installed, Application Updated) and screen views is optional via initialization config parameters, but highly recommended to hit the ground running with core events! See below for more info!

And of course, import the SDK in the files that you use it with:

#import <Analytics/SEGAnalytics.h>

Including SDKs for destinations using Device-mode

In the interest of keeping our SDK lightweight, the Analytics pod only installs the Segment destination. This means that all your data is sent using Segment’s servers to any tools you’ve enabled using the default Cloud-mode.

As described here, some integrations require or offer Device-mode connections. In those cases, you’ll need to take some additional steps as shown in the source documentation here.

Now that the SDK is installed and setup, you’re ready to…

Configure and Setup the SDK

The SEGAnalyticsConfiguration class provides a set of properties that control various policies of the SEGAnalytics instance. You initialize it with a writeKey like so:

[SEGAnalytics setupWithConfiguration:[SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"]];

Application Lifecycle Tracking

Our SDK can automatically instrument common application lifecycle events such as “Application Installed”, “Application Updated” and “Application Opened”. Simply enable this option when you initialize the SDK.

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];
configuration.trackApplicationLifecycleEvents = YES;
[SEGAnalytics setupWithConfiguration:configuration];

Automatic Screen Tracking

Our SDK can automatically instrument screen calls. It uses method swizzling to detect when ViewController’s are loaded and uses the label of the view controller (or the class name if a label is not available) as the screen name. It removes the string “ViewController” from the name (if present).

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];
configuration.recordScreenViews = YES;
[SEGAnalytics setupWithConfiguration:configuration];

Automatic Push Notification Tracking

Tracking Push notifications will automatically track Push Notification Received and Push Notification Tapped.

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];
configuration.trackPushNotifications = YES;
[SEGAnalytics setupWithConfiguration:configuration];

Automatic Deep Link Tracking

Tracking deep linking will automatically track Deep Link Clicked and Deep Link Opened.

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];
configuration.trackDeepLinks = YES;
[SEGAnalytics setupWithConfiguration:configuration];

Flushing

You can set the number of events should queue before flushing. Setting this to 1 will send events as they come in (i.e. not send batched events) and will use more battery. 20 by default.

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];
configuration.flushAt = 1;
[SEGAnalytics setupWithConfiguration:configuration];

You can also manually flush the queue:

[[SEGAnalytics sharedAnalytics] alias:@"glenncoco"];
[[SEGAnalytics sharedAnalytics] flush]

Data Collection

Now that the Segment SDK and any accompanying packaged SDKs are installed, you’re ready to collect some data!

Identify

identify lets you tie a user to their actions and record traits about them. It includes a unique User ID and any optional traits you know about them.

We recommend calling identify a single time when the user’s account is first created, and only identifying again later when their traits are changed.

Note: We automatically assign an anonymousId to users before you identify them. The userId is what connects anonymous activities across devices (e.g. iPhone and iPad).

Example identify call:

[[SEGAnalytics sharedAnalytics] identify:@"a user's id"
                                traits:@{ @"email": @"a user's email address" }];

This call identifies a user by his unique User ID (the one you know him by in your database) and labels him with name and email traits.

The identify call has the following fields:

Analytics for iOS works on its own background thread, so it will never block the main thread for the UI or the calling thread.

Calling - identify: with a userId will write that ID to disk to be used in subsequent calls. That ID can be removed either by uninstalling the app or by calling reset.

Find details on the identify method payload in our Spec.

Track

track lets you record the actions your users perform. Every action triggers what we call an “event”, which can also have associated properties.

To get started, our SDK can automatically track a few key common events with our Native Mobile Spec, such as the Application Installed, Application Updated and Application Opened. Simply enable this option during initialization.

You’ll also want to track events that are indicators of success for your mobile app, like Signed Up, Item Purchased or Article Bookmarked. We recommend tracking just a few important events. You can always add more later!

Example track call:

[[SEGAnalytics sharedAnalytics] track:@"Item Purchased"
                           properties:@{ @"item": @"Sword of Heracles", @"revenue": @2.95 }];

This example track call tells us that your user just triggered the Item Purchased event recording the item name of “Sword of Heracles” and revenue of 2.95.

track event properties can be anything you want to record. In this case, item and revenue.

The track call has the following fields:

Screen

The screen method lets you you record whenever a user sees a screen of your mobile app, along with optional extra information about the page being viewed.

You’ll want to record a screen event an event whenever the user opens a screen in your app. This could be a view, fragment, dialog or activity depending on your app.

Example screen call:

[[SEGAnalytics sharedAnalytics] screen:@"Photo Feed"
                            properties:@{ @"Feed Type": @"private" }];

The screen call has the following fields:

Find details on the screen payload in our Spec.

Group

group lets you associate an identified user user with a group. A group could be a company, organization, account, project or team! It also lets you record custom traits about the group, like industry or number of employees.

This is useful for tools like Intercom, Preact and Totango, as it ties the user to a group of other users.

Example group call:

[[SEGAnalytics sharedAnalytics] group:@"group123"
traits:@{ @"name": @"Initech", @"description": @"Accounting Software" }];

The group call has the following fields:

Find more details about group including the group payload in our Spec.

Alias

alias is how you associate one identity with another. This is an advanced method, but it is required to manage user identities successfully in some of our destinations.

In Mixpanel it’s used to associate an anonymous user with an identified user once they sign up. For KISSmetrics, if your user switches IDs, you can use ‘alias’ to rename the ‘userId’.

Example alias call:

[[SEGAnalytics sharedAnalytics] alias:@"some new id"];

The alias call has the following fields:

For more details about alias, including the alias call payload, check out our Spec.

AnonymousId

You can retrieve the anonymousId set by the library by using:

[[SEGAnalytics sharedAnalytics] getAnonymousId];

Reset

The - reset method clears the SDK’s internal stores for the current user and group. This is useful for apps where users can log in and out with different identities over time.

Clearing all information about the user is as simple as calling:

[[SEGAnalytics sharedAnalytics] reset];

Events queued on disk are not cleared and are uploaded the next time the app starts.

Note: Each time you call reset, a new AnonymousId is generated the next time the app is opened, which can impact the number of Monthly Tracked Users (MTUs) you process.

Disabling Data Collection for Users who opt out

Depending on the audience for your app (e.g. children) or the countries where you sell your app (e.g. the EU), you may need to offer the ability for users to opt-out of analytics data collection inside your app. You can turn off forwarding to ALL destinations including Segment itself:

[[SEGAnalytics sharedAnalytics] disable];

Or if they opt-back-in, you can re-enable data collection:

[[SEGAnalytics sharedAnalytics] enable];

Note: disabling the Segment SDK ensures that all data collection method invocations (eg. track, identify, etc) are ignored; however, it does not tear down inititialized SDKs. If your packaged SDKs are collecting data automatically or outside of Segment, disabling Segment does not address that. We recommend invoking corresponding disable methods in each of your packaged SDKs in response to user opt-out to ensure any automatic data collection is stopped.

Selecting Destinations

The alias, group, identify, page and track calls can all be passed an object of destinations that lets you turn certain destinations on or off. By default all destinations are enabled.

You can enable or disable destinations by specifying an NSDictionary * in the options parameter of our methods as follows:

options:@{
  @"integrations": @{
    @"ENABLED_INTEGRATION_NAME": @YES,
    @"DISABLED_INTEGRATION_NAME: @NO
  }
}

Here’s an example showing an - track: call that is sent to all enabled destinations except Mixpanel:

[[SEGAnalytics sharedAnalytics] track:@"Product Rated"
                           properties:nil
                              options:@{ @"integrations": @{ @"All": @YES, @"Mixpanel": @NO }}];

Destination flags are case sensitive and match the destination’s name in the docs (i.e. “AdLearn Open Platform”, “awe.sm”, “MailChimp”, etc.).

Note: Available at the business level, filtering track calls can be done right from the Segment UI on your source schema page. We recommend using the UI if possible since it’s a much simpler way of managing your filters and can be updated with no code changes on your side.

Destinations in Debugger

If you are seeing any of your destinations turned off in the raw version of requests in the Segment live debugger, but you haven’t added those to your requests, like this:

"integrations": {
  "Segment.io": false,
  "Google Analytics": false,
  "Localytics": false,
  "Mixpanel": false
}

These flags tell the Segment servers that a request was already made directly from the device through a packaged SDK. That way we don’t send a duplicate request via our servers to those services.

Logging

To see a trace of your data going through the SDK, you can enable debug logging with - debug::

[SEGAnalytics debug:YES];

Or disable it like this:

[SEGAnalytics debug:NO];

By default debug logging is disabled.

Middlewares

Middlewares are a powerful mechanism that can augment the events collected by the SDK. A middleware is a simple function that is invoked by the Segment SDK and can be used to monitor, modify or reject events. Middlewares are available on analytics-ios 3.6.0 and higher.

The middleware API is easily accessible in both Objective-C and Swift.

Usage

Middleware is any Objective-C class that conforms to the following protocol.

@protocol SEGMiddleware
@required
- (void)context:(SEGContext *_Nonnull)context next:(S
EGMiddlewareNext _Nonnull)next;
@end

We also provide a block-centric helper class to make it easier to create middlewares using anonymous functions on the fly. (See examples later on in the guide)

typedef void (^SEGMiddlewareBlock)(SEGContext *_Nonnull context, SEGMiddlewareNext _Nonnull next);

@interface SEGBlockMiddleware : NSObject <SEGMiddleware>
@property (nonnull, nonatomic, readonly) SEGMiddlewareBlock block;

- (instancetype _Nonnull)initWithBlock:(SEGMiddlewareBlock _Nonnull)block;

@end

context is an object that encapsulates everything about an event in the stream. next is a callback function that should be invoked when the current middleware is done processing the event and can pass the processed event down to the next middleware in the chain.

SEGContext object is not very information rich by itself. Typically you will need to use eventType and payload to get more information about an event.

@interface SEGContext : NSObject <NSCopying>

@property (nonatomic, readonly, nonnull) SEGAnalytics *_analytics;
@property (nonatomic, readonly) SEGEventType eventType;

@property (nonatomic, readonly, nullable) NSString *userId;
@property (nonatomic, readonly, nullable) NSString *anonymousId;
@property (nonatomic, readonly, nullable) NSError *error;
@property (nonatomic, readonly, nullable) SEGPayload *payload;
@property (nonatomic, readonly) BOOL debug;

- (instancetype _Nonnull)initWithAnalytics:(SEGAnalytics *_Nonnull)analytics;

- (SEGContext *_Nonnull)modify:(void (^_Nonnull)(id<SEGMutableContext> _Nonnull ctx))modify;

@end

If you look at SEGEventType more carefully, you’ll realize that middleware is not only capable of handling track , identify and other normal analytics APIs, even calls like reset , flush and openURL go through and can therefore be processed by the middleware pipeline.

typedef NS_ENUM(NSInteger, SEGEventType) {
    // Should not happen, but default state
    SEGEventTypeUndefined,
    // Core Tracking Methods
    SEGEventTypeIdentify,
    SEGEventTypeTrack,
    SEGEventTypeScreen,
    SEGEventTypeGroup,
    SEGEventTypeAlias,

    // General utility
    SEGEventTypeReset,
    SEGEventTypeFlush,

    // Remote Notification
    SEGEventTypeReceivedRemoteNotification,
    SEGEventTypeFailedToRegisterForRemoteNotifications,
    SEGEventTypeRegisteredForRemoteNotifications,
    SEGEventTypeHandleActionWithForRemoteNotification,

    // Application Lifecycle
    SEGEventTypeApplicationLifecycle,

    // Misc.
    SEGEventTypeContinueUserActivity,
    SEGEventTypeOpenURL,
};

There are almost as many SEGPayload subclasses as there are SEGEventType enums. Subclassed payloads may contain call specific information, For example, the SEGTrackPayload contains event as well as properties .

@interface SEGTrackPayload : SEGPayload

@property (nonatomic, readonly) NSString *event;

@property (nonatomic, readonly, nullable) NSDictionary *properties;

@end

Finally, to use a middleware, you will need to provide it to the SEGAnalyticsConfiguration object prior to the initialization of SEGAnalytics .

@interface SEGAnalyticsConfiguration : NSObject
/**
 * Set custom middlewares. Will be run before all integrations
 */
@property (nonatomic, strong, nullable) NSArray<id<SEGMiddleware>> *middlewares;

// ...
@end

Once initialized, the list of middlewares used in SEGAnalytics cannot be changed at this time.

Examples

The following examples will be written in Swift to demonstrate that the middleware API works just as well in Swift as in Objective-C. To start off, this is what initialization looks like

let config = SEGAnalyticsConfiguration(writeKey: "YOUR_WRITEKEY_HERE")
config.trackApplicationLifecycleEvents = true
config.trackDeepLinks = true
config.recordScreenViews = true
config.use(SEGMixpanelIntegrationFactory.instance())

config.middlewares = [
    turnScreenIntoTrack,
    enforceEventTaxonomy,
    customizeAllTrackCalls,
    sampleEventsToMixpanel,
    blockScreenCallsToAmplitude,
]
SEGAnalytics.setup(with: config)

Changing event names and adding custom attributes

let customizeAllTrackCalls = SEGBlockMiddleware { (context, next) in
    if context.eventType == .track {
        next(context.modify { ctx in
            guard let track = ctx.payload as? SEGTrackPayload else {
                return
            }
            let newEvent = "[New] \(track.event)"
            var newProps = track.properties ?? [:]
            newProps["customAttribute"] = "Hello"
            ctx.payload = SEGTrackPayload(
                event: newEvent,
                properties: newProps,
                context: track.context,
                integrations: track.integrations
            )
        })
    } else {
        next(context)
    }
}

Turn one kind call into another

let turnScreenIntoTrack = SEGBlockMiddleware { (context, next) in
    if context.eventType == .screen {
        next(context.modify { ctx in
            guard let screen = ctx.payload as? SEGScreenPayload else {
                return
            }
            let event = "\(screen.name) Screen Tracked"
            ctx.payload = SEGTrackPayload(
                event: event,
                properties: screen.properties,
                context: screen.context,
                integrations: screen.integrations
            )
            ctx.eventType = .track
        })
    } else {
        next(context)
    }
}

Completely block events

let   = SEGBlockMiddleware { (context, next) in
    let validEvents = [
        "Application Opened",
        "Order Completed",
        "Home Screen Tracked",
        "AnalyticsIOSTestApp. Screen Tracked",
    ]
    if let track = context.payload as? SEGTrackPayload {
        if !validEvents.contains(track.event) {
            showAlert(title: "Dropping Rogue Event",
                      message: track.event)
            return
        }
    }
    next(context)
}

Sample events to Mixpanel

let sampleEventsToMixpanel = SEGBlockMiddleware { (context, next) in
    if let track = context.payload as? SEGTrackPayload {
        let numberBetween0To4 = arc4random() % 5
        if numberBetween0To4 != 0 {
            next(context.modify { ctx in
                ctx.payload = SEGTrackPayload(
                    event: track.event,
                    properties: track.properties,
                    context: track.context,
                    integrations: ["Mixpanel": false]
                )
            })
            return
        }
    }
    next(context)
}

Block only screen calls to amplitude

let blockScreenCallsToAmplitude = SEGBlockMiddleware { (context, next) in
    if let screen = context.payload as? SEGScreenPayload {
        next(context.modify { ctx in
            ctx.payload = SEGScreenPayload(
                name: screen.name,
                properties: screen.properties,
                context: screen.context,
                integrations: ["Amplitude": false]
            )
        })
        return
    }
    next(context)
}

Proxy HTTP Calls

You can point the iOS SDK to your own hosted proxy of the Segment API. This will run the HTTP traffic for the Segment API through the proxy.

SEGAnalyticsConfiguration *configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];

// Set a custom request factory which allows you to modify the way the library creates an HTTP request.
// In this case, we're transforming the URL to point to our own custom non-Segment host.
configuration.requestFactory = ^(NSURL *url) {
    NSURLComponents *components = [NSURLComponents componentsWithURL:url resolvingAgainstBaseURL:NO];
    // Replace YOUR_PROXY_HOST with the address of your proxy, e.g. aba64da6.ngrok.io.
    components.host = @"YOUR_PROXY_HOST";
    NSURL *transformedURL = components.URL;
    return [NSMutableURLRequest requestWithURL:transformedURL];
};

// Set any other custom configuration options.
...

// Initialize the SDK with the configuration.
[SEGAnalytics setupWithConfiguration:configuration];

Submitting to the App Store

When you submit to the app store, be aware that Segment collects the IDFA for use in doing mobile install attribution with destinations like Mobile App Tracking. Even if you’re not currently doing mobile install attribution, if you get asked, “Does this app use the Advertising Identifier (IDFA)?” on this page, you’ll want to check the following three boxes:

1. “Attribute this app installation to a previously served advertisement”
2. “Attribute an action taken within this app to a previously served advertisement”
3. “I, YOUR_NAME, confirm that this app, and any third party…”

Note, you should not check the box labeled “Serve advertisements within the app” unless you are actually going to display ads.

Limited Ad Tracking

iOS users can opt into limited ad tracking (similar to ad-blocking for browsers). For those users that have opted in, adTrackingEnabled will come through as false; however there will still be an advertisingId present. Since the iOS 10 release, those who opt in for limited ad tracking will have adTrackingEnabled set to false AND there will either be no advertisingId or the advertisingId will be a series of zeroes.

Bleeding Edge Releases

We publish stable releases every second Wednesday, when we tag and release the master branch.

After releasing, we also merge the dev branch merged into master. In general, code will be available on master for two weeks before being tagged as a stable release. During this two week period, it is available for use via Cocoapods and Carthage — our equivalent of bleeding edge releases. We recommend using this version to try out upcoming features and fixes that have not been published yet.

To use the master branch for CocoaPods users, use this line in your Podfile:

pod 'Analytics', :git => 'https://github.com/segmentio/analytics-ios.git', :branch => 'master'

To use the master branch for Carthage users, use this line in your Cartfile:

github "segmentio/analytics-ios" "master"

Packaging Destinations using Device-mode

By default, our Analytics pod packages no external SDKs.

pod 'Analytics', '~> 3.1.0'

To add destinations using Device-mode, first add the dependencies you need. You can find these in our app when you open the destination sheet for any mobile destination with a Device-mode option.

pod 'Segment-Bugsnag'
pod 'Segment-Branch'
pod 'Segment-GoogleAnalytics'
...

After adding the dependency, you must register the destination with our SDK.

#import <Segment-GoogleAnalytics/SEGGoogleAnalyticsIntegrationFactory.h>
#import <Segment-Branch/BNCBranchIntegrationFactory.h>

SEGAnalyticsConfiguration *config = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];

// Add any of your Device-mode destinations.
[config use:[SEGGoogleAnalyticsIntegrationFactory instance]];
[config use:[BNCBranchIntegrationFactory instance]];
...

[SEGAnalytics setupWithConfiguration:config];

We recommend using Device-mode destinations sparingly to reduce the size of your application.

FAQ

How big is the Segment SDK?

The core Segment SDK is extremely lightweight. It weighs in at about 212kb.

Can I install the SDK Manually with a Dynamic Framework?

We highly recommend using Cocoapods.

However, if you cannot use Cocoapods or Carthage, you can manually install our dynamic framework allowing you to send data to Segment and on to enabled cloud-mode destinations. We do not support sending data to bundled, device-mode integrations outside of Cocoapods.

Here are the steps for installing manually:

1. Download the latest built SDK, and unzip the zip file.
2. Drag the unzipped Analytics.framework folder into your XCode project.
3. In the General Tab for your project, search for Embedded Binaries and add the Analytics.framework.

Once you’ve installed the framework, just import the header file and install as described above in Install the SDK.

Please note, if you are choosing to not use a dependency manager, you must keep files up-to-date with regularly scheduled, manual updates.

What if your SDK doesn’t support feature X?

If you’re using a Device-mode for a mobile destination, if you want to access a feature from a tool’s native SDK, you can include the header file and call the method just as normal.

For example, you might want access to Flurry’s location logging or Localytics’s attribution parameters. To use the destination’s SDK, just import the headers and then access the SDK as you would without Segment. We’ll still handle initialization, event, screen & user tracking, plus all the proxied services and data storage for you.

Here’s an example for Flurry location logging:

#import <Flurry-iOS-SDK/Flurry.h>

CLLocationManager *locationManager = [[CLLocationManager alloc] init];
[locationManager startUpdatingLocation];
CLLocation *location = locationManager.location;
[Flurry setLatitude:location.coordinate.latitude
          longitude:location.coordinate.longitude
 horizontalAccuracy:location.horizontalAccuracy
   verticalAccuracy:location.verticalAccuracy];

How Do I Use Push Notifications?

For services that send push notifications, you first want to create a Push SSL certificate following these steps. You then want to configure your application delegate to look like the code below, and replace your Segment source write key.

- (BOOL)application:(UIApplication *)application didFinishLaunchingWithOptions:(NSDictionary *)launchOptions {
  SEGAnalyticsConfiguration* configuration = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];

  // Use launchOptions to track tapped notifications
  configuration.launchOptions = launchOptions;
  [SEGAnalytics setupWithConfiguration:configuration];

  if ([[UIApplication sharedApplication] respondsToSelector:@selector(registerForRemoteNotifications)]) {
    UIUserNotificationType types = UIUserNotificationTypeAlert | UIUserNotificationTypeSound |
    UIUserNotificationTypeBadge;
    UIUserNotificationSettings *settings = [UIUserNotificationSettings settingsForTypes:types
    categories:nil];
    [[UIApplication sharedApplication] registerUserNotificationSettings:settings];
    [[UIApplication sharedApplication] registerForRemoteNotifications];
  } else {
    UIRemoteNotificationType types = UIRemoteNotificationTypeAlert | UIRemoteNotificationTypeSound |
    UIRemoteNotificationTypeBadge;
    [[UIApplication sharedApplication] registerForRemoteNotificationTypes:types];
  }
  return YES;
}

- (void)application:(UIApplication *)application didRegisterForRemoteNotificationsWithDeviceToken:(NSData *)deviceToken {
  [[SEGAnalytics sharedAnalytics] registeredForRemoteNotificationsWithDeviceToken:deviceToken];
}

// A notification has been received while the app is running in the foreground
- (void)application:(UIApplication *)application didReceiveRemoteNotification:(NSDictionary *)userInfo {
  [[SEGAnalytics sharedAnalytics] receivedRemoteNotification:userInfo];
}

// iOS 8+ only
- (void)application:(UIApplication *)application didRegisterUserNotificationSettings:(UIUserNotificationSettings *)notificationSettings {
  // register to receive notifications
  [application registerForRemoteNotifications];
}

How Do You Handle Unique Identifiers?

A key component of any analytics platform is consistently and accurately identifying users. Some kind of ID must be assigned and persisted on the device so that user actions can be effectively studied. This is especially important for funnel conversion analysis and retention analysis.

Naturally the Analytics SDK needs a unique ID for each user. To protect end-users’ privacy, Apple places restrictions on how these IDs can be generated and used. Here’s an explanation of these policies from Apple, and how we generate IDs in compliance.

Before iOS 5 developers had access to uniqueIdentifier which was a hardware-specific serial number that was consistent across different apps, vendors and installs. Starting with iOS 5, however, Apple deprecated access to this identifier. In iOS 6 Apple introduced the identifierForVendor which protects end-users from cross-app identification. In iOS 7 Apple restricted access to the device’s MAC address, which was being used by many developers as a workaround to get a device-specific serial number similar to like uniqueIdentifier.

Segment’s iOS library supports iOS 7+ by generating a UUID and storing it on disk. This is in line with the privacy policies required by Apple, maintains compatibility, and leaves open the option for multiple users on one device since the UUID can be regenerated.

Should I include each service’s SDK alongside Segment?

No, don’t include an SDK manually for a service we support. That will cause symbol conflicts/namespace collisions. Sometimes it can even fail silently :( So make sure you remove the old Google Analytics or Mixpanel SDK when you install Segment’s SDK.

How does the SDK queue API calls?

Our SDK queues API calls so that we don’t use up your user’s battery life by making a network request for each event tracked.

Here’s how queuing works for server-side destinations: When you make an API call (e.g. -track:) that call is added to the queue. The SDK sends the events to the server in batches (by default, the batch size is 100). The batches are then sent either when there are 20 or more events in the queue, on a scheduled timer every 30 seconds, or when the app goes to the background. To limit memory and disk usage, we only queue upto 1000 events.

When the app is terminated we persist the queue to disk and load that data at app launch so there is no data loss. The queue behavior may differ in packaged destinations.

Can I set user traits without a User ID?

Yes! Just pass a nil value for the userId into your identify call, like this:

[[SEGAnalytics sharedAnalytics] identify: nil
    traits:@{ @"email": @"example@example.com",
      @"Gender": @"F" }];

Do you support iOS 5?

Our SDK does not support iOS 5. If you need support for iOS 5 it’s possible by forking our iOS repo on GitHub and building the framework.

Is The Segment SDK Compatible with Swift?

Indeed! Swift’s compatibility with Objective-C lets you create a source that contains files written in either language, so to use our SDK from a Swift source just follow the instructions from Apple here.

Can I help develop a destination?

Yep! Our SDK is open-source. If you’d like to contribute, fix a bug, or add a destination - here’s documentation on how to do so. to add a destination, make sure you contact our partners team first.

How do I know when a destination is initialized?

The iOS library will post a notification to indicate when it initializes any destination so you can call it’s methods directly.

[[NSNotificationCenter defaultCenter] addObserver:self selector:@selector(integrationDidStart:) name:SEGAnalyticsIntegrationDidStart object:nil];

- (void)integrationDidStart:(nonnull NSNotification *)notification
{
    NSString *integration = notification.object;

    if ([integration.name isEqualToString:@"Mixpanel"]) {
        // Call Mixpanel library methods here.
    }
}

Can I anonymize IP addresses?

We collect IP address for client-side (iOS, Android, Analytics.js and Xamarin) events automatically.

If you don’t want us to record your tracked users’ IP in destinations and S3, you can set your event’s context.ip field to 0.0.0.0 . Our server won’t record the IP address of the client for libraries if the context.ip field is already set. An example would look like this:

[[SEGAnalytics sharedAnalytics] track: @"Clicked Button"
  properties:nil
  options:@{ @"context": @{@"ip": @"0.0.0.0"}}];

If you’d like to centralize this logic, you can write a middleware for it!

IDFA

Some destinations, particularly mobile attribution tools (e.g. Kochava), require the IDFA (identifier for advertisers). The IDFA shows up in Segment calls in the debugger under context.device.advertisingId. In order for this value to be captured by the Segment SDK, ensure that you include the AdSupport framework.

Once you enable this, you will see the context.device.advertisingId populate and the context.device.adTrackingEnabled flag set to true.

Note: While the network is deprecated, the relevant framework is not.

tvOS Support

As of Version 3.3.0 we now have support for tvOS through our Analytics-iOS sdk. You can follow the iOS quickstart documentation and you should be good to go! tvOS installation is only supported via Carthage and CocoaPods. The dynamic framework installation method is not supported for tvOS.

Troubleshooting

Target has transitive dependencies that include static binaries

This was due to an old CocoaPods limitation.

1. Verify you are not using any previously needed workarounds
2. Verify you are using at least CocoaPods 1.4

   $ pod --version
   1.5.3
   

3. Verify you are using the latest Segment pods ```bash $ pod outdated The following pod updates are available:
   • Segment-GoogleAnalytics 1.1.7 -> 1.1.7 (latest version 1.1.8) ```

No events in my debugger

1. Verify you have followed all Getting Started steps
2. Verify you have entered the correct writeKey for your source
   • If the writeKey you have entered is something other than a string or an empty string your app may crash
   • If the writeKey you have entered is a valid form but not the correct writeKey for your specific source, you will not see an error response. Data will be accepted by Segment but not able to be correctly routed to your source (debugger).
3. Enable logging to confirm if call is being sent to Segment

No events in my destinations

1. Verify that your destination is enabled
2. Verify your destination credentials entered in your Segment dashboard are correct
3. Make sure the destination can accept what you’re sending:
   • Does the integration have device-mode/cloud-mode support? Confirm you are sending via the correct connection mode.
   • Does the destination accept the type of call you are sending? Not all destinations accept all calls: page, track, etc.
4. If you are still not seeing data in your destination, continue debugging based on which type of connection mode you are using.

Debugging Device-mode Destinations

If you are using device-mode, you should see the value of that integration set to false in the integrations object. That means that the data is being sent from the device to the destination SDK, and not through Segment’s servers. This is expected if you chose to use a device-mode destination’s SDK with Segment’s during installation.

Enable verbose logging and trigger the call in question. You should see a call to Segment triggered as well as to the partner SDK. It will show you exactly which partner method was invoked and the arguments it was invoked with!

Debugging Cloud-mode Destinations

Look at the raw JSON in your debugger. Does the call look like what is expected?

Read through the docs for that destination to see expected event format, behavior and caveats for that destination.

Migrating to v3 from earlier releases

v3 was an API compatible release, but there are a few additional steps for packaging destinations using Device-mode and migrating from the older data format.

Firstly, we changed how the anonymousId was stored between v2 and v3. You’ll need to read the old anonymousId and set it so that it’s moved to the new location.

NSString *oldAnonymousId = loadOldAnonymousId();
if (oldAnonymousId) {
  [[SEGAnalytics sharedAnalytics] identify:userId
                           traits:traits
                              options:@{ @"anonymousId": oldAnonymousId }];
  deleteOldAnonymousId();
}

In version 3, we’ve organized the destinations to be make the core SDK even leaner and smaller. The Analytics/Segmentio pod is not available any longer. It has been renamed to Analytics (which previously packaged all possible destinations). Version 3 of Analytics only includes the core library which forwards data directly to . To add a destination using Device-mode, you must manually add that destination’s dependencies, like so:

pod 'Segment-Bugsnag'
pod 'Segment-Branch'
pod 'Segment-GoogleAnalytics'
...

…and then register them in your configuration when you initialize the SDK.

Note: If you are unsure about the name of the pod for a given SDK, you can always confirm on by searching for the destination in our app!

#import <Segment-GoogleAnalytics/SEGGoogleAnalyticsIntegrationFactory.h>
#import <Segment-Branch/BNCBranchIntegrationFactory.h>

SEGAnalyticsConfiguration *config = [SEGAnalyticsConfiguration configurationWithWriteKey:@"YOUR_WRITE_KEY"];

// Add any of your Device-mode destination components.
[config use:[SEGGoogleAnalyticsIntegrationFactory instance]];
[config use:[BNCBranchIntegrationFactory instance]];
...

[SEGAnalytics setupWithConfiguration:config];

Still having issues?

Feel free to reach out to us with the following information:

• The version of our SDK you are using
• Whether you are using device- or cloud-mode
• Logs of the call in question
• Screenshots of the event in the Segment debugger
• Screenshots of what you are seeing in your destination